Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A zero‐sample state evaluation model for valve‐side bushing of UHV converter transformer oriented to digital twin under attribute analysis

A zero‐sample state evaluation model for valve‐side bushing of UHV converter transformer oriented... INTRODUCTIONUnder the background of low carbon and energy structure adjustment, higher requirements are put forward for the optimal allocation of power resources [1]. The key to realizing the safe transmission of large capacity and long‐distance power is to improve the operation and maintenance management level of power equipment. The valve‐side bushing of the ultra‐high voltage (UHV) converter transformer is the “bridge” and weak link to realize power conversion [2], and its running state stability directly affects the reliability of DC transmission projects. According to statistics, in recent years, power system failures caused by converter valve‐side bushing of converter transformer account for 16.7% of all equipment failures [3], and the forms of failures also show diversity, such as SF6 gas leakage, aging of capacitor core, oil leakage, discharge, and so on [4–7]. At this stage, most of the research on the valve‐side bushing of UHV converter transformers remains in the fault occurrence mechanism, fault analysis, and physical structure optimization [8–10]. However, due to the variety of fault forms, difficulties in data collection, and poor correlation, there is less research on the evaluation of the running state.In the state research of valve‐side bushing of UHV converter transformers, the acquisition of its state data is the first prerequisite. With the development of digital twin technology in aerospace technology [11], medical equipment [12], rail transit [13], water conservancy engineering [14], and other fields, the integration technology between digital twin technology and electric power field is becoming closer. Tao et al. [15] used digital twin technology to establish a health status management system for wind turbines, and real‐time updates and fusion of twin data can be carried out through this system. Jia et al. [16, 17] established the CloudEPS platform by using digital twin technology to plan and process physical entities at the virtual level, thus providing a feasible scheme for the optimal control of the energy internet. To improve the operation and maintenance level of high‐voltage cables, He et al. [18] proposed the use of the digital twin framework to carry out full‐cycle state management of high‐voltage cables. As the heavy equipment in power systems, the development of digital twin technology also brings new opportunities and challenges to the valve‐side bushing of UHV converter transformers.To ensure the reliability of the state evaluation of the valve‐side bushing of the UHV converter, the analysis of its effective state feature information is a key step. Elsisi et al. [19] conducted fault diagnosis for transformers based on an effective deep‐learning platform. Cong et al. [20] tested the infrared spectrum of the transformer in running and established the electric heating fault diagnosis model of the transformer oil. Leong et al. [21] studied the transformer health state index and proposed the method of transformer state feature extraction using the ultraviolet‐visible spectrum for the first time. Du et al. [22] used currents of different frequencies as testing tools to analyse the influence of water in oil on bushing aging. Monga et al. [23] proposed the calculation method for bushing electric fields and studied the influence of key components on electric field distribution in combination with the bushing structure model. The research on bushing state characteristics also includes insulation features [24], dielectric features [25], partial discharge [26], and other objects, but the above feature objects have problems such as fewer data and lower sensitivity to training parameters in the process of bushing state evaluation.In the state evaluation model of the valve‐side bushing of the UHV converter transformer, Záliš et al. [27] proposed to establish an expert mapping system using historical fault data, but due to the differences between valve‐side bushing models and running environments, the stability of its weight parameters is poor. Liao et al. [28] used the dielectric model to analyse the water content of oil‐penetrating bushing and took the grey correlation degree as a parameter index to complete the state evaluation. Elsisi et al. [29–34] applied machine learning and optimization algorithms to automatic driving, hybrid battery control, robot arm cooperative control, etc., while there are relatively few state assessment optimization algorithms in ultra‐high voltage power equipment.The valve‐side bushing of the UHV converter transformer is different from other power insulation equipment. It has a large internal current and a complex external running environment, which will cause serious accidents when it fails [35]. Due to the rapid occurrence of valve‐side bushing accidents and the low accident rate, it is difficult to collect status data. The combination of digital twin technology and status feature analysis can solve the problem of insufficient status representation. When using the fuzzy clustering algorithm to establish the state evaluation model, it has the characteristics of zero‐samples [36], and the state evaluation results can be intuitively expressed through the similarity matching degree in the dynamic clustering graph.Based on the above analysis, this paper proposes to obtain the twin data of the valve‐side bushing of the UHV converter transformer under the attribute analysis based on the digital twin technology and realize the zero‐sample status evaluation combined with the fuzzy clustering algorithm. First, the physical entity is analyzed, including attribute parameters, electrical features, ambient temperature, and current‐carrying. To extract more details of the carried current of the bushing, empirical mode decomposition (EMD) is used to decompose the fundamental and harmonic components. Then, the COMSOL software is used to complete the mapping from physical entity to digital twin, which is verified and analyzed according to the axial heat distribution of the bushing, and use different carrier current components to obtain the temperature extremum inside and outside the bushing to establish the twin data. Finally, the initial state feature sets of valve‐side bushing under different defect degrees are established, the fuzzy clustering algorithm is used to process them, and the state evaluation of valve‐side bushing is completed with the similarity matching degree in the dynamic clustering diagram. Through the verification and analysis of examples, the state evaluation method can ensure the safe running of equipment and further improve the operation and maintenance level of the valve‐side bushing.The main contributions of this paper are as follows: (a) Analyse the basic physical attributes of the bushing, and then establish the physical entity in the digital twin technology, including not only the fundamental and harmonic components of the input current after EMD processing, but also the analysis of the bushing structural materials, electrical characteristics, and geometric dimensions. (b) The overheating mechanism of the bushing is simulated by COMSOL to obtain the overheating characteristics of internal defects, and then the bushing state feature set under different defects is established. (c) Aimed at the characteristic of a few bushing fault samples, a zero‐sample state evaluation model based on a fuzzy clustering algorithm is proposed. The bushing state evaluation is completed by similarity matching degree, which does not require advanced training.The structure of this paper is as follows: (1) Analysis of basic algorithms, mainly including digital twin framework, attribute analysis algorithm, and fuzzy clustering algorithm; (2) state evaluation of valve‐side bushing of UHV converter transformer, including physical entity analysis, virtual twin, state evaluation process; (3) example demonstration, mainly including the simulation of virtual space, the establishment of state feature set of data twins, and the evaluation of running state; 4) conclusion.ALGORITHM ANALYSISDigital twin framework for state assessment systemThe concept of digital twins was first proposed by Grieves [37, 38] and used to establish the mapping relationship between physical entities and abstract virtual entities. With the development of information technology, the internet of things, cloud computing, communications, and other technologies, higher requirements are put forward for digital twin technology in the operation and maintenance of specific equipment. The application of digital twin technology in the field of power equipment mainly consists of five parts: physical entity (PE), digital twin (DW), operations management (OM), digital link (DL), and Data twin (DT). The system framework is shown in Figure 1.1FIGURESystem frame of digital twin technologyThe specific meanings of digital twin technology are as follows:PE: When the research object of the physical entity is the valve‐side bushing, the characteristic description of the input object includes not only the geometric, material, and environment, but also the output current characteristics, defect distribution, sensing equipment, and other elements.DW: Digital twin is to complete the mapping from physical entity to virtual space, including product simulation, dynamic display, and operation parameters.OM: Operation and maintenance management is the premise to ensure the safe and stable running of equipment, mainly including fault diagnosis, state evaluation, and state prediction.DT: Data twinning runs through the whole physical entity, digital twinning, operation and maintenance management, and other links.DL: As a bridge, the digital link realizes the data interaction between physical entities, digital twins, operation, and maintenance management.The digital twin framework based on the state evaluation system fully maps the running characteristics of physical entities into the virtual space. While establishing the digital twin, it obtains the internal characteristics to realize the expansion of the running status features and solves the problems of online monitoring difficulty and a small sample size of fault defects. Thus, the whole life cycle of physical entities and digital twins can be synchronously evolved and evaluated.Attribute analysis algorithmThe attribute analysis algorithm is the mathematical basis for the dynamic analysis of the research object. The attribute analysis of the valve‐side bushing of the UHV converter transformer includes both the description of physical entities and the dynamic display principle of digital twins. The input current decomposition algorithm and heat distribution theory are introduced in detail below.1) Output current decomposition algorithm.EMD is the frequency domain analysis of the original signal according to the time scale. Assuming that the original signal sequence is x(t) and the natural mode function component is imfi(t), the EMD can be formulated as Equation (1).1xt=∑i=1nimfi(t)+r(t),$$\begin{equation}x\left( t \right) = \sum_{i = 1}^n {im{f_i}(t) + r(t)} ,\end{equation}$$where i is the variable of decomposition layer n, r(t) is the residual component, and different intrinsic mode function components imfi(t) meet the Dirichley conditions:The termination condition of the whole decomposition process is that r(t) is a monotone signal. The corresponding EMD is shown in Algorithm 1.ALGORITHM 1: EMDInput: Original signal x(t); Decomposition layers i; Intermediate variable k.Output: Intrinsic mode function component imfi(t), residual component r(t).1 // initialization2: i←1, k←03: set r(t) = x(t)4: r(t) local maximum point emax(t), r(t) local minimum point emin(t)5: // evaluate6: m(t)=(emax(t)+emin(t))/2$m(t) = ({e_{max}}(t) + {e_{min}}(t))/2$, update k = k+17: pk(t)=r(t)−m(t),r(t)=pk(t)${p_k}(t) = r(t) - m(t),{\kern 1pt} {\kern 1pt} {\kern 1pt} r(t) = {p_k}(t)$8: Ifpk(t)${p_k}(t)$satisfies two conditions then9 imfi(t)=pk(t)$im{f_i}(t) = {p_k}(t)$10 r(t)=r(t)−imfi(t)$r(t) = r(t) - im{f_i}(t)$11 and if r(t) is the monotone signal then12 x(t)=∑i=1nimfi(t)+r(t)$x( t ) = \sum_{i = 1}^n {im{f_i}(t) + r(t)} $13 else i = i+114 Return r(t) local maximum point emax(t), r(t) local minimum point emin(t)15 else16 Return r(t) local maximum point emax(t), r(t) local minimum point emin(t)2) Calculation principle of heat distribution.The heat source on the valve‐side bushing of the UHV converter transformer is generated by the Joule effect of the conductive tube; the heat conduction between different solid materials, the convection heat transfer between solid and non‐solid materials, and the thermal radiation of the solid itself need to be comprehensively considered in the heat distribution calculation.Heat conduction exists between two solids with a certain temperature difference. The solid materials of the valve‐side bushing include a catheter, capacitor core, flange, watchband contact, and other devices, and the heat conduction satisfies the mathematical definition shown in Equation (2) [39]:2λtr·∂∂rr∂T∂r+λt·∂2T∂z2+Qtc=0,$$\begin{equation}\frac{{{\lambda _t}}}{r} \cdot \frac{\partial }{{\partial r}}\left( {r\frac{{\partial T}}{{\partial r}}} \right) + {\lambda _t} \cdot \frac{{{\partial ^2}T}}{{\partial {z^2}}} + {Q_{tc}} = 0,\end{equation}$$where r is the geometric radius; T the temperature; λt the thermal conductivity; Qtc the amount of heat conduction between solids in contact.The convective heat transfer of the valve‐side bushing includes two contact modes: the external natural environment and the transformer insulating oil; the convective heat transfer satisfies the mathematical model of Equation (3) [40]:3Qi=2πkNu(Ti−To)ln(Do/Di),$$\begin{equation}{Q_i} = \frac{{2\pi k{N_u}({T_i} - {T_o})}}{{\ln ({D_o}/{D_i})}},\end{equation}$$where Qi is the convective heat conduction; k the convective thermal conductivity; Nu is the Nussel constant; Ti, To the temperature inside and outside the bushing surface; Di, Do the diameter inside and outside the bushing.The thermal radiation of the valve‐side bushing is also a part of the thermal distribution. Compared with heat conduction and convective heat transfer, thermal radiation does not need to contact the medium to transfer. The solid thermal radiation based on electromagnetic energy can be calculated by the Stefan–Boltzmann equation as shown in Equation (4) [41]:4Qr=εσA1F12(T14−T24)$$\begin{equation}{Q_r} = \varepsilon \sigma {A_1}{F_{12}}({T_1}^4 - {T_2}^4)\end{equation}$$where Qr is the thermal radiation; ε the emissivity (boldface); σ the Stefan–Boltzmann factor; A1 the surface area of the radiator; F12 the shape coefficient between surfaces of different radiating radiators; T1, T2 the absolute temperature of the surface of different radiators.State evaluation algorithmSuppose there are n objects to be inspected xt(t = 1, 2, …, n), and there are m characteristic indicators under each object to be inspected. According to the theory of fuzzy mathematics, a fuzzy clustering algorithm based on the initial feature matrix Xn * m is shown in Algorithm 2.ALGORITHM 2: Fuzzy Clustering AlgorithmInput: Original signal x(t); Decomposition layers i; Intermediate variable k.Output: Intrinsic mode function component imfi(t), residual component r(t).1: // initialization2: i←(1,2, …n), j←(1,2, …m)3: k←04: // evaluate6: Standardization (X→X′′${\bf{X^{\prime\prime}}}$)57:xik′=xij−x¯jsj(i=1,2,…,n;j=1,2,…,m)$$\begin{equation}7:{x^{\prime}_{ik}} = \frac{{{x_{ij}} - {{\bar x}_j}}}{{{s_j}}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} (i = 1,2, \ldots ,n;{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} j = 1,2, \ldots ,m)\end{equation}$$68:xik′′=x′ij−min1≤i≤n{x′ij}max1≤i≤n{x′ij}−min1≤i≤n{x′ij}(j=1,2,…,m)$$\begin{equation}8:\;{x^{\prime\prime}_{ik}} = \frac{{{{x^{\prime}}_{ij}} - \mathop {\min }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} }}{{\mathop {\max }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} - \mathop {\min }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} }}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} (j = 1,2, \ldots ,m)\end{equation}$$9: Similitude matrix (X′′${\bf{X^{\prime\prime}}}$→R)710:rij=1m∑k=1mexp−34⋅(xik−xjk)2sk2$$\begin{equation}10:{{r}_{ij}}=\frac{1}{m}\sum\limits_{k=1}^{m}{\exp \left[-\frac{3}{4}\centerdot \frac{{{({{x}_{ik}}-{{x}_{jk}})}^{2}}}{s_{k}^{2}}\right]}\end{equation}$$11: Equivalent Boolean Matrix (R→R∗${{\bf{R}}^*}$)812:R2=R∘R,R4=R2∘R2,…,$$\begin{equation} 12:{{\bf{R}}^2} = {\bf{R}} \circ {\bf{R}}{\kern 1pt} ,{{\bf{R}}^4} = {{\bf{R}}^2} \circ {{\bf{R}}^2}, \ldots ,\end{equation}$$13: If there is a natural K, which makes R2K=R2(K+1)${{\bf{R}}^{2K}} = {{\bf{R}}^{2(K + 1)}}$ then914:rij=1,rij≥λ0,rij<λ$$\begin{equation}14:\;{r_{ij}} = \left\{ \def\eqcellsep{&}\begin{array}{l} 1{\kern 1pt} {\kern 1pt} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {r_{ij}} \ge \lambda \\ 0{\kern 1pt} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {r_{ij}} &lt; \lambda \end{array} \right.\end{equation}$$Equation (5) of Algorithm 2 solves the dimensional problem so that the elements of the fuzzy characteristic matrix are in the interval [0, 1]; x¯j${\bar x_j}$ and sk, respectively, represents the mean value and variance, which are calculation as follows:10x¯j=1n∑i−1nxij(j=1,2,…,m),$$\begin{equation}{\bar x_j} = \frac{1}{n}\sum_{i - 1}^n {{x_{ij}}} (j = 1,2, \ldots ,m),\end{equation}$$11sk=1n∑i=1n(xij−x¯j)2.$$\begin{equation}{s_k} = \frac{1}{n}\sum_{i = 1}^n {{{({x_{ij}} - {{\bar x}_j})}^2}} .\end{equation}$$Finally, when the confidence factor λ changes from 1 to 0 (the computer automatically reduces by 0.001 steps), the columns with the same equivalent Boolean matrix are classified into one category. When λ=min1≤i,j≤n{rij}$\lambda = \mathop {\min }\limits_{1 \le i,j \le n} \{ {r_{ij}}\} $ completes the classification, the dynamic clustering graph is established accordingly.STATE EVALUATION OF VALVE‐SIDE BUSHING OF UHV CONVERTER TRANSFORMERThe valve‐side bushing of the UHV converter transformer belongs to the current‐carrying connection component. As a weak link in power transmission, its working state is not only affected by its high voltage, high temperature, mechanical vibration, material properties, and other factors but also affected by the external environment, insulation oil corrosion, and other factors. To realize the state evaluation of the valve‐side bushing, this paper first analyses the basic attributes of the physical entity. Then, use COMSOL to establish digital twins. Finally, the zero‐sample status evaluation process is analyzed.Physical entity of valve‐side bushing of UHV converter transformerThe main insulation of the valve‐side bushing is oil impregnated paper capacitor core, the air side is a silicone rubber composite insulator, and the transformer insulation oil side is a porcelain insulator or composite insulator. Its overall structure is shown in Figure 2.2FIGURESchematic structure of valve side bushing of converter transformerIn Figure 2, the connecting terminal is used to connect external input signals. The conservator is used to adjust the volume change of insulating oil. The upper and lower porcelain bushings are used to resist external discharge. The capacitor core is used to improve the electric field distribution of the bushing. The flange is used to install the bushing, and a measuring terminal is installed on one side to achieve effective grounding. The grading ring is used to eliminate the electric field imbalance.To realize the mapping between physical entities and twins, this paper takes the ±800kV valve‐side bushing of an electric power company as the experimental object to conduct physical analysis. The physical parameters are mainly composed of attribute parameters, electrical characteristics, and external inputs.1) Attribute parametersBasic attribute parameters of valve‐side bushing of UHV converter transformer are shown in Table 11TABLEAttribute parameters of the valve‐side bushing of the UHV converter transformerComponentAttributedataAluminium tubeLength/outside diameter/inside diameter/mm1 400/176/140Copper tubeLength/outside diameter/inside diameter/mm12 600/140/92Insulator skirtOutside diameter/mm708BushingLength/mm14850Rated voltageTransformer(AC+DC)/kV8442) Electrical characteristicsThe capacitor core is made of epoxy resin, and the electrical characteristics of other materials are shown in Table 2.2TABLEElectrical characteristics of the valve‐side bushing of UHV converter transformerMaterialRelative inductivityConductivity (S/m)Silicon rubber2.85 × 10−16Aluminium20003.5 × 107Copper20005.7 × 107Epoxy resin4.51 × 10−13Converter transformer insulating oil2.25 × 10−14SF61.0021 × 10−203) External inputsThe insulation structure of the bushing can be divided into two parts: the side of the converter transformer oil and the side of the valve hall. Therefore, the external environment includes both the ambient temperature of the valve hall and the insulating oil characteristics of the converter transformer. According to the provisions of IEC 60137 and GB/T 4109, in the experiment of thermal characteristics analysis, the temperature of the valve hall is 50°C±2°C, and the temperature of the oil is 90°C±2°C, and the theoretical difference between the two is not more than 60K±2K. The bushing excitation is also an important part of external input. Figure 3 shows the current waveforms at both ends of the bushing when a UHV power plant is running at ±800kV full load.3FIGURECurrent waveform at both ends of the valve‐side bushing of the UHV converter transformerEstablishment of a digital twin for the valve‐side bushing UHV converter transformerTo realize the establishment of a physical entity into a digital twin, COMSOL software is used as the experimental simulation platform here. The model constraint problem is described as follows:First, the physical entity of valve‐side bushing is analyzed, including physical attribute parameters: geometric dimension, material attribute, electrical characteristics, etc., as well as environmental parameters: insulation oil temperature, valve chamber temperature, etc.Then, there are many reasons for bushing failure, such as contact finger corrosion of the strap, SF6 gas leakage, and so on. Through the simulation tests, the temperature change of the failed bushing is the most obvious electrical connection device. Therefore, this paper defines the contact resistance of the electrical connector as the bushing degradation feature. The boundary conditions are set according to the basic principles of heat conduction, convective heat transfer, and thermal radiation in Section 2.2.Finally, the rated current of the valve‐side bushing is AC+DC, when the spectrum analysis of the measured current waveform at both ends of the bushing in Figure 3 is carried out; the current includes other harmonic components in addition to the 50 Hz basic component, so the skin‐effect of harmonic current must be considered in the heat loss calculation [40]. Assuming that the g‐th harmonic current component is Ig, its heat loss is calculated as shown in Equation (12):12P(t)=∑g=1GIg2Rg(t′)$$\begin{equation}P(t) = \sum_{g = 1}^G {I_g^2} {R_g}(t^{\prime})\end{equation}$$where Rg(t′)${R_g}(t^{\prime})$is the harmonic resistance, whose calculation method is shown in Equation (13):13Rg(t′)=ρt′Lπ▵dg(D−▵dg)▵dg=ρt′πfgμ$$\begin{equation} \left\{ \begin{aligned} {{R}_{g}}({t}')&=\frac{{{\rho }_{{{t}'}}}L}{\pi \vartriangle {d}_{g}( D-\vartriangle {d}_{g} )} \\ \vartriangle {{d}_{g}}&=\sqrt{\frac{{{\rho }_{{{t}'}}}}{\pi {{f}_{g}}\mu }} \end{aligned}\right.\end{equation}$$where D is the bushing diameter; ρt′${\rho _{t^{\prime}}}$the resistivity; L the length of the conduit; μ the magnetic resistance; fg the current frequency.Based on the analysis of the above steps and combined with the physical experiment platform, the digital twin built by COMSOL is shown in Figure 4.4FIGUREConstruction of digital twin for valve‐side bushing of UHV converter transformerZero‐sample state evaluation process based on attribute analysisThe factors that affect the running state of the valve‐side bushing are diverse, which inevitably leads to problems such as diverse failure forms and weak correlation. Typical failures include SF6 gas leakage, partial discharge, aging of the connection parts, and contact corrosion of the watchband. In the face of different fault types, their fault mechanisms are also very different, so it is difficult to extract fault feature data.To solve the above problems, this paper proposes a zero‐sample state evaluation method for valve‐side bushing based on digital twin technology and attributes analysis. The specific state evaluation flow diagram is shown in Figure 5.5FIGUREEvaluation process of valve‐side bushing state of UHV converter transformerIt can be seen from Figure 5 that the zero‐sample state evaluation of valve‐side bushing based on attribute analysis is mainly composed of four parts: physical entity analysis, simulation analysis of virtual space, the establishment of state feature set based on data twins, and running state evaluation.1) Physical entity analysisBased on the analysis of physical characteristics (attribute parameters and electrical characteristics of ±800kV valve‐side bushing in Section 3.1), the physical entity analysis of valve‐side bushing is completed in combination with running environment temperature (insulation oil temperature and hall temperature).2) Simulation of virtual spaceFirst, COMSOL is used to set geometric, material properties and physical fields; then, the measured carrier current is analyzed in the frequency domain, and it is decomposed into decomposition components with different frequency centres by EMD; Finally, combined with the heat loss calculation model in Section 3.2, the digital twins of valve side bushing are established.3) Establishment of state feature set based on data twinThrough the analysis of the digital twin, the bushing heat distribution is taken as the research object, and the characteristics data of the bushing external maximum temperature, external minimum temperature, internal maximum temperature, and internal minimum temperature under different carrier current decomposition components are obtained.4) Running status evaluationThe initial feature matrix is established by using the bushings with different defect degrees, and the running state evaluation of the valve‐side bushing is completed by combining the fuzzy clustering algorithm flow in Section 2.3.EXAMPLE VERIFICATION AND ANALYSISConstruction of digital twins for the valve‐side bushingCurrent decomposition of the valve‐side bushingThe carrier voltage and current of the valve‐side bushing are related to the working mode of the converter transformer. To establish the corresponding digital twins of the bushing and extract more detailed features, it is necessary to decompose other detailed components except for the 50 Hz fundamental component.Taking the dry‐type SF6 gas insulated bushing of phase A at Y‐Y side of ±800kV UHV as an example, its input signal waveform is shown in Figure 3. The initial current waveform is decomposed at different centre frequencies using EMD algorithm. The decomposed intrinsic mode function component imf(t) and residual component r(t) are shown in Figure 6.6FIGUREEMD decomposition of bushing current at valve‐side of UHV converter transformerAs shown in Figure 6, after the bushing current is processed by the EMD algorithm, there are four intrinsic mode componentsimf1(t)$im{f_{\rm{1}}}(t)$∼imf4(t)$im{f_{\rm{4}}}(t)$and one residual component r(t), and r(t) meets the termination condition of EMD.Simulation analysis of virtual space based on sample attributesWhen the valve‐side bushing has fault defects, the most direct features are the changes in DC impedance and carrier current. Although the external thermal feature distribution of the bushing can be obtained through infrared testing, it is difficult to obtain the internal features. Combined with the basic attributes, DC impedance, and other features of the physical entity of the valve‐side bushing in Section 3.1, the EMD results of the carried current are input into the virtual model, and the digital twin technology is used to obtain the internal characteristics under the simulation analysis.The simulation results of UHV ±800kV dry‐type SF6 gas‐insulated bushing in COMSOL are shown in Figure 7.7FIGURESimulation results of axial heat distribution of valve‐side bushingAs shown in Figure 7, the temperature of the simulated bushing at the insulating oil side of the transformer is about 73°C, and the highest temperature in the axial temperature distribution of the bushing occurs at the connection of the aluminium–copper conduit. When the bushing fails, the most obvious change is the temperature at the connection. Therefore, to obtain the state features of the bushing, the section at the connection is simulated and intercepted as shown in Figure 8.8FIGURESimulation results of radial heat distribution of valve‐side bushingEffectiveness analysis of thermal distribution of twinsTo verify the effectiveness of the digital twin simulation analysis in the above section, the axial thermal distribution of physical entities and virtual entities of UHV ±800kV valve‐side dry‐type SF6 gas‐insulated bushing is compared and analyzed. The maximum temperature results are shown in Figure 9.9FIGUREComparison of axial temperature distribution between the physical entity and digital twin of the valve‐side bushingAs shown in Figure 9, the maximum temperature range of the bushing is 69.37–100.15°C, while the normal reference value is 120°C, so the bushing is under normal operation. By comparing the axial heat distribution of physical entities and digital twins, the average error is not more than 5%, which meets the one‐to‐one mapping relationship. Then, the internal characteristics of the bushing can be obtained through the digital twin, which lays the foundation for state data twinning.Defect analysis of bushing and establishment of state feature setTaking a ±500 kV power plant as an example, the picture of the capacitor core when the valve‐side bushing is faulty is shown in Figure 10.10FIGUREThe capacitor core of the valve‐side bushing is faultyAccording to field detection, the blackening substance of the capacitor core is the substance produced after the local temperature rises due to the rise of the impedance at the copper–aluminium connection, and then local flashover occurs. Analyse other faults such as SF6 gas leakage, and the change of impedance of the bushing conductive tube is the first. Therefore, in the face of the diversity of fault defect forms of the bushing and the features of fewer fault data in operation and maintenance management, the use of digital twins to obtain twin data under different impedance changes is conducive to achieving zero‐sample state evaluation under online monitoring.In the process of establishing the state feature set, first, the digital twin is established according to the decomposition result of the carrier current and the impedance of the catheter in the physical entity, and then the maximum temperature and minimum temperature of the catheter are obtained. Then, the initial state feature set is established. Finally, to reduce the dimension of the feature set, principal component analysis is used to map the initial state features into one‐dimensional space. Based on the analysis of ±800kV valve‐side dry‐type SF6 gas‐insulated bushing under normal conditions in Section 4.1, the state features under different decomposition components are shown in Table 3.3TABLEState features of valve‐side bushing under normal conditionsDecomposition of componentimf1(t)imf2(t)imf3(t)imf4(t)r(t)Maximum temperature of conduit (°C)75.875.875.875.875.8Minimum temperature of conduit (°C))48.948.948.948.948.9Maximum external temperature (°C)35.135.135.135.135.1Minimum external temperature (°C)20.820.820.820.820.8To complete the dimension reduction analysis of the feature set, the principal component analysis of the feature set in Table 3 under normal conditions is carried out, and the results are as follows:xt=184.376.190.510.51$$\begin{equation*}{{\bf{x}}_t}{\rm{ = }}\left[ {{\rm{184}}{\rm{.37}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{6}}{\rm{.19}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{0}}{\rm{.51}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{0}}{\rm{.51}}} \right]\end{equation*}$$Case study on evaluation of bushing state at converter valve‐sideTo realize the state evaluation of the valve‐side bushing of the UHV converter transformer, the state feature vectors of the bushing under different defects are established according to the method of establishing the feature vectors under normal conditions in the above section. In the example demonstration, x1∼x5 are used to represent the state feature vectors of the different bushing. x1 is the bushing with the impedance of 52.2 μΩ under normal condition, and x2∼x5 are the bushing with defect impedance of 120.4, 180.6, 200.3, and 260.8 μΩ.Grey relation analysisTo further verify the effectiveness of the fuzzy clustering algorithm in the state evaluation, this paper analyses the initial matrix X by citing the grey correlation coefficient, in which the calculation method of the grey correlation coefficient is as follows:ξi(k)=minimink|x1(k)−xi(k)|+ρmaximaxk|x1(k)−xi(k)||x1(k)−xi(k)|+ρmaximaxk|x1(k)−xi(k)|,$$\begin{equation*}{\xi _i}(k) = \frac{{\mathop {\min }\limits_i \mathop {\min }\limits_k |{x_1}(k) - {x_i}(k)| + \rho \mathop {\max }\limits_i \mathop {\max }\limits_k |{x_1}(k) - {x_i}(k)|}}{{|{x_1}(k) - {x_i}(k)| + \rho \mathop {\max }\limits_i \mathop {\max }\limits_k |{x_1}(k) - {x_i}(k)|}},\end{equation*}$$where ξi(k)${\xi _i}(k)$is the correlation coefficient of x0 to xi at point k, minimink|x0(k)−xi(k)|$\mathop {\min }\limits_i \mathop {\min }\limits_k |{x_0}(k) - {x_i}(k)|$is the absolute value of the second level minimum difference of x1 to xi at point k, ρ Is the grey resolution coefficient, generally 0.5, and the final grey correlation is ri:ri=1n∑k=1nξi(k).$$\begin{equation*}{r_i} = \frac{1}{n}\sum_{k = 1}^n {{\xi _i}(k)} .\end{equation*}$$Finally, the grey correlation coefficient of x1 in the initial characteristic matrix X is shown in Figure 11.11FIGUREDistribution of grey correlation coefficientAs shown in the Figure 11, the grey correlation coefficients of x1 are very similar; the grey correlation degrees are 0.776, 0.795, 0.785, and 0.795, respectively. Moreover, the correlation degrees of x3 and x4 are the same, so effective bushing state evaluation cannot be conducted.Fuzzy clustering analysisAccording to the fuzzy clustering algorithm shown in Algorithm 2 in Section 2.3, the zero‐sample state evaluation process of the valve‐side bushing is as follows.1) Initial feature matrix X.X=[x1;x2;x3;x4;x5]$$\begin{equation*}{\bf{X}} = [{{\bf{x}}_1};{{\bf{x}}_2};{{\bf{x}}_3};{{\bf{x}}_4};{{\bf{x}}_5}]\end{equation*}$$2) Standardize with Equations (5) and (6).X′′=000.2140.5050.3360.533000.5560.7210.6220.7740.7350.8740.6540.7771111$$\begin{equation*} {{\bf{X}}^{{\bf{^{\prime\prime}}}}} = \left[ \def\eqcellsep{&}\begin{array}{cccc} 0 & 0 & 0.214 & 0.505\\[2pt] 0.336 & 0.533 & 0& 0\\[2pt] 0.556 & 0.721 & 0.622 & 0.774\\[2pt] 0.735 & 0.874 & 0.654 & 0.777\\[2pt] 1 & 1 & 1& 1\end{array} \right]\end{equation*}$$3) Use Equation (7) to establish matrix RR=10.6630.4440.2670.0040.66310.7800.6030.3400.4440.78010.8220.5590.2670.6030.82210.7360.0040.3400.5590.7361$$\begin{equation*}{\bf{R}} = \left[ \def\eqcellsep{&}\begin{array}{ccccc} 1 & 0.663 & 0.444 & 0.267 & 0.004\\[2pt] 0.663 & 1 & 0.780 & 0.603 & 0.340\\[2pt] 0.444 & 0.780 & 1 & 0.822 & 0.559\\[2pt] 0.267 & 0.603 & 0.822 & 1 & 0.736\\[2pt] 0.004 & 0.340 & 0.559 & 0.736 & 1\end{array} \right]\end{equation*}$$4) As shown in Equations (8) and (9), the matrix is established after making matrix R transitive, and the cluster diagram formed when λ=min1≤i,j≤n{rij}$\lambda = \mathop {\min }\limits_{1 \le i,j \le n} \{ {r_{ij}}\} $is shown in Figure 12.12FIGURECluster diagram of valve‐side bushing statue evaluationAs shown in Figure 12, when λ is 0.664, the bushing x1 and other bushings are classified, so the similarity λ = 0.664 can be used as a criterion for bushing evaluation. The higher the similarity with normal conditions, the better the bushing status. The lower the similarity with normal conditions, the lower the bushing status.In conclusion, using the attribute parameters of physical entities to establish a digital twin, the internal features of the data twin can be obtained, and then the zero‐sample state evaluation of the valve‐side bushing can be realized.CONCLUSIONThe valve‐side bushing of the UHV converter transformer is the key equipment to realize the safe transmission of electric power. This paper conducts attribute analysis and research on the physical entity and uses the digital twin technology to obtain the internal state feature set under the defect state. Through case analysis and research, when the similarity of the fuzzy clustering algorithm is taken as the index, its zero‐sample state evaluation can be achieved, and the following conclusions are drawn:In the physical entity analysis of the valve‐side bushing, the research includes both the basic physical characteristics and the electrical characteristics. To decompose the different centre frequency components of the carrier current, the EMD algorithm is used to decompose the 50 Hz fundamental component and other harmonic components, which further improves the accuracy of establishing the digital twin.In the process of establishing the digital twin of the valve‐side bushing, the mechanism of the bushing with different defects is analyzed, its state characteristics are analyzed according to the thermal distribution phenomenon, and the validity of the data twin is verified by using the axial temperature distribution of the digital twin.In the zero‐sample state evaluation of valve‐side bushing, the initial state feature set is established by using the temperature distribution at the copper aluminium joint and the external temperature distribution under different decomposition components of bushing current carrying, and the fuzzy clustering algorithm is used to calculate the valve‐side bushing under different defect degrees. Finally, the zero‐sample state evaluation is realized by similarity matching, which provides a guarantee for equipment maintenance and safe operation.Herein, the preliminary assessment and analysis of the valve‐side bushing state have been realized. The following problems need to be solved in future research:More problems such as stability and applicability of valve‐side bushing state assessment under different application environments need to be solved urgently.How to realize the residual life prediction under complex running conditions is conducive to the establishment of a modern health management system.AUTHOR CONTRIBUTIONSZheng Li: writing ‐ original draft. Kai Liu: data curation; formal analysis; writing ‐ review and editing. Mu Lin: investigation; validation. Dongli Xin: data curation. Hao Tang: resources. Guangning Wu: methodology.ACKNOWLEDGEMENTThis research was supported in part by the Science and Technology Project of Headquarter of SGCC: SGTYHT/19‐JS‐215.CONFLICT OF INTERESTThere is no conflict of interest.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.REFERENCESWang, L., Tao, Z., Zhu, L., et al.: Optimal dispatch of integrated energy system considering integrated demand response resource trading. IET Gener. Transm. Distrib. 16, 1727–1742 (2022)Chen, M., Liu, X., Shao, Y., et al.: Cracking risk analysis and control for high‐voltage dry‐type valve‐side bushings. IET Gener. Transm. Distrib. 14, 6555–6561 (2020)Mikhak‐Beyranvand, M., Faiz, J. Rezaeealam, B.: Thermal analysis and derating of a power transformer with harmonic loads. IET Gener. Transm. Distrib. 14, 1233–1241 (2020)Maximov, S., Olivares‐Galvan, J. C., Magdaleno‐Adame, S., et al.: New analytical formulas for electromagnetic field and eddy current losses in bushing regions of transformers. IEEE Trans. on Magn. 51(4), 1–10 (2015)Su, C., Branch, D.: On‐site Testing requirements of ultra‐high voltage converter transformer insulating Oil and SF_6 Gas. Elect. Engin. (07): 146–149 +160 (2016)Zhang, S., Peng, Z., Peng, L., et al.: Design and dielectric characteristics of the ±1100 kV UHVDC wall bushing in china. IEEE Trans. on Diel. & Elect. Insul. 22(1), 409–419 (2015)Chen, M., Liu, X., Liang, C., et al.: Study on surface charge Accumulation characteristics of resin impregnated paper wall bushing core under positive DC voltage. Energies. 12, 4420 (2019)Zhang, L., Shuai, X., Han, C., et al.: Valve side lead exit insulation design used in ±800 UHV converter transformer. In Proceedings of the International Conference on Condition Monitoring and Diagnosis (CMD), pp. 452–455. IEEE, Piscataway, NJ 2016.Liu, K., Yang, Z., Wei, W., et al.: Novel detection approach for thermal defects: study on its feasibility and application to vehicle cables. High Voltage. 1–10 (2022)Shi, Y. U., Nie, D., Weimin, M. A., et al.: Overvoltage and insulation coordination of Jinping‐Sunan ±800 kV UHVDC project converter stations. High Volt. Engin. 38(12), 3219–3223 (2012)Liu, Z., Chen, C., Zhang, C., et al.: Data super‐network fault prediction model and maintenance strategy for mechanical product based on digital twin. IEEE Access 7, 177284–177296 (2019)Alarcón, Serrano, ngel, Madrid, et al.: The role of digital twins in personalized sleep medicine. In German‐Italian Workshop, Social Innovation in Long‐Term Care through Digitalization. Springer, Cham (2022)Jim, G.: Software specialist, rFpro develops digital modelling tool of Applus+ IDIADA proving grounds for real world trials of autonomous vehicles. Truck & Bus Builder: The International Newsletter of Commercial Vehicle Manufacturing, Developments. 40(7), 10–20 (2018)Shi, Y., Xu, J., Du, W.: Discussion on the new operation management mode of hydraulic engineering based on the digital twin technique. J. Phys.: Confer. Seri., 1168, 022044 (2019)Tao, F., Zhang, M., Liu, Y., et al.: Digital twin driven prognostics and health management for complex equipment. CIRP Anna. 67(1), 169–172 (2018)Jia, M., Shen, C., Chen, Y., et al.: Digital twin of the energy internet and its application. Global Ener. Inter. 3(1), 1–13 (2020)Zhou, M., Yan, J., Feng, D.: Digital twin framework and its application to power grid online analysis. CSEE J. Power Ener. Syst. 5(3), 391–398 (2019)He, W., Wu, Z., Ren, X., et al.: Research on the application of digital twin technique in high voltage cable. In 2020 4th International Conference on Power and Energy Engineering (ICPEE), pp. 90–93. IEEE, Piscataway, NJ (2022)Elsisi M., Tran M., Mahmoud K., et al.: Effective IoT‐based deep learning platform for online fault diagnosis of power transformers against cyberattacks and data uncertainties. Measurement 190, 110686 (2022)Li, S., Cao, B., Cui, Y., et al.: Terahertz‐based insulation delamination defect inspection of vehicle cable terminals. IEEE Transactions on Transportation Electrification, 1–9 (2022)Leong, Y., Ker, P., Jamaludin, M. M.: UV‐vis spectroscopy: A new approach for assessing the color index of transformer insulating oil. Sensors. 18(7), 2175 (2018)Du, Z., Nie, D., Zhang, L., et al.: Insulation state evaluation of high voltage bushing based on frequency domain dielectric spectrum analysis. High Voltage Appliances. 49(11), 6–11 (2013)Monga, S., Gorur, R. S., Hansen, P., et al.: Design optimization of high voltage bushing using electric field computations. IEEE Trans. on Diel. & Elect.l Insul. 13(6), 1217–1224 (2006)Sarkar, S., Sharma, T., Baral, A., et al.: An expert system approach for transformer insulation diagnosis combining conventional diagnostic tests and PDC, RVM data. IEEE Transactions on Dielectrics and Electrical Insulation. 21(2), 882–891 (2014)Nie, D., Zhang, H., Chen, Z., et al.: Optimization design of grading ring and electrical field analysis of 800 kV UHVDC wall wall bushing. IEEE Trans. on Dielec. & Elec. Insulation. 20(4), 1361–1368 (2013)Radakovic, Z., Cardillo, E., Schaefer, M., et al.: Design of the winding‐bushing interconnections in large power transformers. Elect. Engin. 88(3), 183–190 (2006)Záliš, K.: Using expert systems in evaluation of the state of high voltage machine insulation systems. Acta Polytechnica. 40(5), 185–190 (2000)Liao, R., Du, Y., Yang, L., et al.: Quantitative diagnosis of moisture content in oil‐paper condenser bushing insulation based on frequency domain spectroscopy and polarisation and depolarisation current. IET Gener. Trans. Distr. 11(6), 1420–1426 (2017)Elsisi M.: Improved grey wolf optimizer based on opposition and quasi learning approaches for optimization: Case study autonomous vehicle including vision system. Arti. Intel. Revi. 55, 5597–5620 (2022)Ismail M. M., Bendary A. F., Elsisi M.: Optimal design of battery charge management controller for hybrid system PV/wind cell with storage battery. Int. J. Power Ener. Conv. 11(4), 412 (2020)Elsisi M., Tran M. Q.: Development of an iot architecture based on a deep neural network against cyber attacks for automated guided vehicles. Sensors 21(24), 8467 (2021)Tran M. Q., Elsisi M., Liu M. K., et al.: Reliable deep learning and iot‐based monitoring system for secure computer numerical control machines against cyber‐attacks with experimental verification. IEEE Access 10, 23186–23197 (2022)Elsisi M., Soliman M., Aboelela M., et al.: ABC based design of PID controller for two area load frequency control with nonlinearities. Telk. Indo. J. Elec. Eng. 16(1), 58–64 (2015)Elsisi M., Zaini H. G., Mahmoud K., et al.: Improvement of trajectory tracking by robot manipulator based on a new co‐operative optimization algorithm. Mathematics 9(24), 3231 (2021)Li, S., Cao, B., Cui, Y., et al.: Nonintrusive inspection of moisture damp in composited insulation structure based on terahertz technology. IEEE Transactions on Instrumentation and Measurement. 70, 1–10 (2021)Yu, Z., Chen, H., You, J., et al.: Adaptive fuzzy consensus clustering framework for clustering analysis of cancer data. IEEE/ACM Trans. Comp. Biol. Bioin. 12(4), 887–901 (2015)Grieves, M. W., Virtually Perfect: Driving Innovative and Lean Products through Product Lifecycle Management. Space Coast Press, Cocoa Beach (2011)Grieves, M. W., Vickers, M.: Digital twin :mitigatingunpredictable, undesirable emergent behavior in complexsystems. Transdisciplinary Perspectives on Complex Systems, pp. 85–113. Cham, Springer (2016)Maximov, S., Escarela‐Perez, R., Magdaleno‐Adame, S., et al.: Calculation of nonlinear electromagnetic fields in the steel wall vicinity of transformer bushings. IEEE Trans. on Magn. 51(6), 1–11 (2015)Wen, M., Lin, X., Zhong, S.: Numerical simulation and experimental study of heat transfer characteristics in transformer bushing, High Volt. Eng. 42(9), 2956–2962 (2016)Du, B., Tian, M., Su, J., et al.: Electrical tree growth characteristics in epoxy resin with harmonic superimposed DC voltage. IEEE Access 7, 47273–47281 (2019)Li, C., Cerrada, M., Cabrera, D., et al.: A comparison of fuzzy clustering algorithms for bearing fault diagnosis. J. Intell. Fuzzy Syst. 34(6), 3565–3580 (2018) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png IET Generation Transmission & Distribution Wiley

A zero‐sample state evaluation model for valve‐side bushing of UHV converter transformer oriented to digital twin under attribute analysis

Download PDF
12 pages

Loading next page...
 
/lp/wiley/a-zero-sample-state-evaluation-model-for-valve-side-bushing-of-uhv-zsBBYwdvMo
Publisher
Wiley
Copyright
© 2023 The Institution of Engineering and Technology.
eISSN
1751-8695
DOI
10.1049/gtd2.12721
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONUnder the background of low carbon and energy structure adjustment, higher requirements are put forward for the optimal allocation of power resources [1]. The key to realizing the safe transmission of large capacity and long‐distance power is to improve the operation and maintenance management level of power equipment. The valve‐side bushing of the ultra‐high voltage (UHV) converter transformer is the “bridge” and weak link to realize power conversion [2], and its running state stability directly affects the reliability of DC transmission projects. According to statistics, in recent years, power system failures caused by converter valve‐side bushing of converter transformer account for 16.7% of all equipment failures [3], and the forms of failures also show diversity, such as SF6 gas leakage, aging of capacitor core, oil leakage, discharge, and so on [4–7]. At this stage, most of the research on the valve‐side bushing of UHV converter transformers remains in the fault occurrence mechanism, fault analysis, and physical structure optimization [8–10]. However, due to the variety of fault forms, difficulties in data collection, and poor correlation, there is less research on the evaluation of the running state.In the state research of valve‐side bushing of UHV converter transformers, the acquisition of its state data is the first prerequisite. With the development of digital twin technology in aerospace technology [11], medical equipment [12], rail transit [13], water conservancy engineering [14], and other fields, the integration technology between digital twin technology and electric power field is becoming closer. Tao et al. [15] used digital twin technology to establish a health status management system for wind turbines, and real‐time updates and fusion of twin data can be carried out through this system. Jia et al. [16, 17] established the CloudEPS platform by using digital twin technology to plan and process physical entities at the virtual level, thus providing a feasible scheme for the optimal control of the energy internet. To improve the operation and maintenance level of high‐voltage cables, He et al. [18] proposed the use of the digital twin framework to carry out full‐cycle state management of high‐voltage cables. As the heavy equipment in power systems, the development of digital twin technology also brings new opportunities and challenges to the valve‐side bushing of UHV converter transformers.To ensure the reliability of the state evaluation of the valve‐side bushing of the UHV converter, the analysis of its effective state feature information is a key step. Elsisi et al. [19] conducted fault diagnosis for transformers based on an effective deep‐learning platform. Cong et al. [20] tested the infrared spectrum of the transformer in running and established the electric heating fault diagnosis model of the transformer oil. Leong et al. [21] studied the transformer health state index and proposed the method of transformer state feature extraction using the ultraviolet‐visible spectrum for the first time. Du et al. [22] used currents of different frequencies as testing tools to analyse the influence of water in oil on bushing aging. Monga et al. [23] proposed the calculation method for bushing electric fields and studied the influence of key components on electric field distribution in combination with the bushing structure model. The research on bushing state characteristics also includes insulation features [24], dielectric features [25], partial discharge [26], and other objects, but the above feature objects have problems such as fewer data and lower sensitivity to training parameters in the process of bushing state evaluation.In the state evaluation model of the valve‐side bushing of the UHV converter transformer, Záliš et al. [27] proposed to establish an expert mapping system using historical fault data, but due to the differences between valve‐side bushing models and running environments, the stability of its weight parameters is poor. Liao et al. [28] used the dielectric model to analyse the water content of oil‐penetrating bushing and took the grey correlation degree as a parameter index to complete the state evaluation. Elsisi et al. [29–34] applied machine learning and optimization algorithms to automatic driving, hybrid battery control, robot arm cooperative control, etc., while there are relatively few state assessment optimization algorithms in ultra‐high voltage power equipment.The valve‐side bushing of the UHV converter transformer is different from other power insulation equipment. It has a large internal current and a complex external running environment, which will cause serious accidents when it fails [35]. Due to the rapid occurrence of valve‐side bushing accidents and the low accident rate, it is difficult to collect status data. The combination of digital twin technology and status feature analysis can solve the problem of insufficient status representation. When using the fuzzy clustering algorithm to establish the state evaluation model, it has the characteristics of zero‐samples [36], and the state evaluation results can be intuitively expressed through the similarity matching degree in the dynamic clustering graph.Based on the above analysis, this paper proposes to obtain the twin data of the valve‐side bushing of the UHV converter transformer under the attribute analysis based on the digital twin technology and realize the zero‐sample status evaluation combined with the fuzzy clustering algorithm. First, the physical entity is analyzed, including attribute parameters, electrical features, ambient temperature, and current‐carrying. To extract more details of the carried current of the bushing, empirical mode decomposition (EMD) is used to decompose the fundamental and harmonic components. Then, the COMSOL software is used to complete the mapping from physical entity to digital twin, which is verified and analyzed according to the axial heat distribution of the bushing, and use different carrier current components to obtain the temperature extremum inside and outside the bushing to establish the twin data. Finally, the initial state feature sets of valve‐side bushing under different defect degrees are established, the fuzzy clustering algorithm is used to process them, and the state evaluation of valve‐side bushing is completed with the similarity matching degree in the dynamic clustering diagram. Through the verification and analysis of examples, the state evaluation method can ensure the safe running of equipment and further improve the operation and maintenance level of the valve‐side bushing.The main contributions of this paper are as follows: (a) Analyse the basic physical attributes of the bushing, and then establish the physical entity in the digital twin technology, including not only the fundamental and harmonic components of the input current after EMD processing, but also the analysis of the bushing structural materials, electrical characteristics, and geometric dimensions. (b) The overheating mechanism of the bushing is simulated by COMSOL to obtain the overheating characteristics of internal defects, and then the bushing state feature set under different defects is established. (c) Aimed at the characteristic of a few bushing fault samples, a zero‐sample state evaluation model based on a fuzzy clustering algorithm is proposed. The bushing state evaluation is completed by similarity matching degree, which does not require advanced training.The structure of this paper is as follows: (1) Analysis of basic algorithms, mainly including digital twin framework, attribute analysis algorithm, and fuzzy clustering algorithm; (2) state evaluation of valve‐side bushing of UHV converter transformer, including physical entity analysis, virtual twin, state evaluation process; (3) example demonstration, mainly including the simulation of virtual space, the establishment of state feature set of data twins, and the evaluation of running state; 4) conclusion.ALGORITHM ANALYSISDigital twin framework for state assessment systemThe concept of digital twins was first proposed by Grieves [37, 38] and used to establish the mapping relationship between physical entities and abstract virtual entities. With the development of information technology, the internet of things, cloud computing, communications, and other technologies, higher requirements are put forward for digital twin technology in the operation and maintenance of specific equipment. The application of digital twin technology in the field of power equipment mainly consists of five parts: physical entity (PE), digital twin (DW), operations management (OM), digital link (DL), and Data twin (DT). The system framework is shown in Figure 1.1FIGURESystem frame of digital twin technologyThe specific meanings of digital twin technology are as follows:PE: When the research object of the physical entity is the valve‐side bushing, the characteristic description of the input object includes not only the geometric, material, and environment, but also the output current characteristics, defect distribution, sensing equipment, and other elements.DW: Digital twin is to complete the mapping from physical entity to virtual space, including product simulation, dynamic display, and operation parameters.OM: Operation and maintenance management is the premise to ensure the safe and stable running of equipment, mainly including fault diagnosis, state evaluation, and state prediction.DT: Data twinning runs through the whole physical entity, digital twinning, operation and maintenance management, and other links.DL: As a bridge, the digital link realizes the data interaction between physical entities, digital twins, operation, and maintenance management.The digital twin framework based on the state evaluation system fully maps the running characteristics of physical entities into the virtual space. While establishing the digital twin, it obtains the internal characteristics to realize the expansion of the running status features and solves the problems of online monitoring difficulty and a small sample size of fault defects. Thus, the whole life cycle of physical entities and digital twins can be synchronously evolved and evaluated.Attribute analysis algorithmThe attribute analysis algorithm is the mathematical basis for the dynamic analysis of the research object. The attribute analysis of the valve‐side bushing of the UHV converter transformer includes both the description of physical entities and the dynamic display principle of digital twins. The input current decomposition algorithm and heat distribution theory are introduced in detail below.1) Output current decomposition algorithm.EMD is the frequency domain analysis of the original signal according to the time scale. Assuming that the original signal sequence is x(t) and the natural mode function component is imfi(t), the EMD can be formulated as Equation (1).1xt=∑i=1nimfi(t)+r(t),$$\begin{equation}x\left( t \right) = \sum_{i = 1}^n {im{f_i}(t) + r(t)} ,\end{equation}$$where i is the variable of decomposition layer n, r(t) is the residual component, and different intrinsic mode function components imfi(t) meet the Dirichley conditions:The termination condition of the whole decomposition process is that r(t) is a monotone signal. The corresponding EMD is shown in Algorithm 1.ALGORITHM 1: EMDInput: Original signal x(t); Decomposition layers i; Intermediate variable k.Output: Intrinsic mode function component imfi(t), residual component r(t).1 // initialization2: i←1, k←03: set r(t) = x(t)4: r(t) local maximum point emax(t), r(t) local minimum point emin(t)5: // evaluate6: m(t)=(emax(t)+emin(t))/2$m(t) = ({e_{max}}(t) + {e_{min}}(t))/2$, update k = k+17: pk(t)=r(t)−m(t),r(t)=pk(t)${p_k}(t) = r(t) - m(t),{\kern 1pt} {\kern 1pt} {\kern 1pt} r(t) = {p_k}(t)$8: Ifpk(t)${p_k}(t)$satisfies two conditions then9 imfi(t)=pk(t)$im{f_i}(t) = {p_k}(t)$10 r(t)=r(t)−imfi(t)$r(t) = r(t) - im{f_i}(t)$11 and if r(t) is the monotone signal then12 x(t)=∑i=1nimfi(t)+r(t)$x( t ) = \sum_{i = 1}^n {im{f_i}(t) + r(t)} $13 else i = i+114 Return r(t) local maximum point emax(t), r(t) local minimum point emin(t)15 else16 Return r(t) local maximum point emax(t), r(t) local minimum point emin(t)2) Calculation principle of heat distribution.The heat source on the valve‐side bushing of the UHV converter transformer is generated by the Joule effect of the conductive tube; the heat conduction between different solid materials, the convection heat transfer between solid and non‐solid materials, and the thermal radiation of the solid itself need to be comprehensively considered in the heat distribution calculation.Heat conduction exists between two solids with a certain temperature difference. The solid materials of the valve‐side bushing include a catheter, capacitor core, flange, watchband contact, and other devices, and the heat conduction satisfies the mathematical definition shown in Equation (2) [39]:2λtr·∂∂rr∂T∂r+λt·∂2T∂z2+Qtc=0,$$\begin{equation}\frac{{{\lambda _t}}}{r} \cdot \frac{\partial }{{\partial r}}\left( {r\frac{{\partial T}}{{\partial r}}} \right) + {\lambda _t} \cdot \frac{{{\partial ^2}T}}{{\partial {z^2}}} + {Q_{tc}} = 0,\end{equation}$$where r is the geometric radius; T the temperature; λt the thermal conductivity; Qtc the amount of heat conduction between solids in contact.The convective heat transfer of the valve‐side bushing includes two contact modes: the external natural environment and the transformer insulating oil; the convective heat transfer satisfies the mathematical model of Equation (3) [40]:3Qi=2πkNu(Ti−To)ln(Do/Di),$$\begin{equation}{Q_i} = \frac{{2\pi k{N_u}({T_i} - {T_o})}}{{\ln ({D_o}/{D_i})}},\end{equation}$$where Qi is the convective heat conduction; k the convective thermal conductivity; Nu is the Nussel constant; Ti, To the temperature inside and outside the bushing surface; Di, Do the diameter inside and outside the bushing.The thermal radiation of the valve‐side bushing is also a part of the thermal distribution. Compared with heat conduction and convective heat transfer, thermal radiation does not need to contact the medium to transfer. The solid thermal radiation based on electromagnetic energy can be calculated by the Stefan–Boltzmann equation as shown in Equation (4) [41]:4Qr=εσA1F12(T14−T24)$$\begin{equation}{Q_r} = \varepsilon \sigma {A_1}{F_{12}}({T_1}^4 - {T_2}^4)\end{equation}$$where Qr is the thermal radiation; ε the emissivity (boldface); σ the Stefan–Boltzmann factor; A1 the surface area of the radiator; F12 the shape coefficient between surfaces of different radiating radiators; T1, T2 the absolute temperature of the surface of different radiators.State evaluation algorithmSuppose there are n objects to be inspected xt(t = 1, 2, …, n), and there are m characteristic indicators under each object to be inspected. According to the theory of fuzzy mathematics, a fuzzy clustering algorithm based on the initial feature matrix Xn * m is shown in Algorithm 2.ALGORITHM 2: Fuzzy Clustering AlgorithmInput: Original signal x(t); Decomposition layers i; Intermediate variable k.Output: Intrinsic mode function component imfi(t), residual component r(t).1: // initialization2: i←(1,2, …n), j←(1,2, …m)3: k←04: // evaluate6: Standardization (X→X′′${\bf{X^{\prime\prime}}}$)57:xik′=xij−x¯jsj(i=1,2,…,n;j=1,2,…,m)$$\begin{equation}7:{x^{\prime}_{ik}} = \frac{{{x_{ij}} - {{\bar x}_j}}}{{{s_j}}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} (i = 1,2, \ldots ,n;{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} j = 1,2, \ldots ,m)\end{equation}$$68:xik′′=x′ij−min1≤i≤n{x′ij}max1≤i≤n{x′ij}−min1≤i≤n{x′ij}(j=1,2,…,m)$$\begin{equation}8:\;{x^{\prime\prime}_{ik}} = \frac{{{{x^{\prime}}_{ij}} - \mathop {\min }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} }}{{\mathop {\max }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} - \mathop {\min }\limits_{1 \le i \le n} \{ {{x^{\prime}}_{ij}}\} }}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} (j = 1,2, \ldots ,m)\end{equation}$$9: Similitude matrix (X′′${\bf{X^{\prime\prime}}}$→R)710:rij=1m∑k=1mexp−34⋅(xik−xjk)2sk2$$\begin{equation}10:{{r}_{ij}}=\frac{1}{m}\sum\limits_{k=1}^{m}{\exp \left[-\frac{3}{4}\centerdot \frac{{{({{x}_{ik}}-{{x}_{jk}})}^{2}}}{s_{k}^{2}}\right]}\end{equation}$$11: Equivalent Boolean Matrix (R→R∗${{\bf{R}}^*}$)812:R2=R∘R,R4=R2∘R2,…,$$\begin{equation} 12:{{\bf{R}}^2} = {\bf{R}} \circ {\bf{R}}{\kern 1pt} ,{{\bf{R}}^4} = {{\bf{R}}^2} \circ {{\bf{R}}^2}, \ldots ,\end{equation}$$13: If there is a natural K, which makes R2K=R2(K+1)${{\bf{R}}^{2K}} = {{\bf{R}}^{2(K + 1)}}$ then914:rij=1,rij≥λ0,rij<λ$$\begin{equation}14:\;{r_{ij}} = \left\{ \def\eqcellsep{&}\begin{array}{l} 1{\kern 1pt} {\kern 1pt} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {r_{ij}} \ge \lambda \\ 0{\kern 1pt} ,{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {r_{ij}} &lt; \lambda \end{array} \right.\end{equation}$$Equation (5) of Algorithm 2 solves the dimensional problem so that the elements of the fuzzy characteristic matrix are in the interval [0, 1]; x¯j${\bar x_j}$ and sk, respectively, represents the mean value and variance, which are calculation as follows:10x¯j=1n∑i−1nxij(j=1,2,…,m),$$\begin{equation}{\bar x_j} = \frac{1}{n}\sum_{i - 1}^n {{x_{ij}}} (j = 1,2, \ldots ,m),\end{equation}$$11sk=1n∑i=1n(xij−x¯j)2.$$\begin{equation}{s_k} = \frac{1}{n}\sum_{i = 1}^n {{{({x_{ij}} - {{\bar x}_j})}^2}} .\end{equation}$$Finally, when the confidence factor λ changes from 1 to 0 (the computer automatically reduces by 0.001 steps), the columns with the same equivalent Boolean matrix are classified into one category. When λ=min1≤i,j≤n{rij}$\lambda = \mathop {\min }\limits_{1 \le i,j \le n} \{ {r_{ij}}\} $ completes the classification, the dynamic clustering graph is established accordingly.STATE EVALUATION OF VALVE‐SIDE BUSHING OF UHV CONVERTER TRANSFORMERThe valve‐side bushing of the UHV converter transformer belongs to the current‐carrying connection component. As a weak link in power transmission, its working state is not only affected by its high voltage, high temperature, mechanical vibration, material properties, and other factors but also affected by the external environment, insulation oil corrosion, and other factors. To realize the state evaluation of the valve‐side bushing, this paper first analyses the basic attributes of the physical entity. Then, use COMSOL to establish digital twins. Finally, the zero‐sample status evaluation process is analyzed.Physical entity of valve‐side bushing of UHV converter transformerThe main insulation of the valve‐side bushing is oil impregnated paper capacitor core, the air side is a silicone rubber composite insulator, and the transformer insulation oil side is a porcelain insulator or composite insulator. Its overall structure is shown in Figure 2.2FIGURESchematic structure of valve side bushing of converter transformerIn Figure 2, the connecting terminal is used to connect external input signals. The conservator is used to adjust the volume change of insulating oil. The upper and lower porcelain bushings are used to resist external discharge. The capacitor core is used to improve the electric field distribution of the bushing. The flange is used to install the bushing, and a measuring terminal is installed on one side to achieve effective grounding. The grading ring is used to eliminate the electric field imbalance.To realize the mapping between physical entities and twins, this paper takes the ±800kV valve‐side bushing of an electric power company as the experimental object to conduct physical analysis. The physical parameters are mainly composed of attribute parameters, electrical characteristics, and external inputs.1) Attribute parametersBasic attribute parameters of valve‐side bushing of UHV converter transformer are shown in Table 11TABLEAttribute parameters of the valve‐side bushing of the UHV converter transformerComponentAttributedataAluminium tubeLength/outside diameter/inside diameter/mm1 400/176/140Copper tubeLength/outside diameter/inside diameter/mm12 600/140/92Insulator skirtOutside diameter/mm708BushingLength/mm14850Rated voltageTransformer(AC+DC)/kV8442) Electrical characteristicsThe capacitor core is made of epoxy resin, and the electrical characteristics of other materials are shown in Table 2.2TABLEElectrical characteristics of the valve‐side bushing of UHV converter transformerMaterialRelative inductivityConductivity (S/m)Silicon rubber2.85 × 10−16Aluminium20003.5 × 107Copper20005.7 × 107Epoxy resin4.51 × 10−13Converter transformer insulating oil2.25 × 10−14SF61.0021 × 10−203) External inputsThe insulation structure of the bushing can be divided into two parts: the side of the converter transformer oil and the side of the valve hall. Therefore, the external environment includes both the ambient temperature of the valve hall and the insulating oil characteristics of the converter transformer. According to the provisions of IEC 60137 and GB/T 4109, in the experiment of thermal characteristics analysis, the temperature of the valve hall is 50°C±2°C, and the temperature of the oil is 90°C±2°C, and the theoretical difference between the two is not more than 60K±2K. The bushing excitation is also an important part of external input. Figure 3 shows the current waveforms at both ends of the bushing when a UHV power plant is running at ±800kV full load.3FIGURECurrent waveform at both ends of the valve‐side bushing of the UHV converter transformerEstablishment of a digital twin for the valve‐side bushing UHV converter transformerTo realize the establishment of a physical entity into a digital twin, COMSOL software is used as the experimental simulation platform here. The model constraint problem is described as follows:First, the physical entity of valve‐side bushing is analyzed, including physical attribute parameters: geometric dimension, material attribute, electrical characteristics, etc., as well as environmental parameters: insulation oil temperature, valve chamber temperature, etc.Then, there are many reasons for bushing failure, such as contact finger corrosion of the strap, SF6 gas leakage, and so on. Through the simulation tests, the temperature change of the failed bushing is the most obvious electrical connection device. Therefore, this paper defines the contact resistance of the electrical connector as the bushing degradation feature. The boundary conditions are set according to the basic principles of heat conduction, convective heat transfer, and thermal radiation in Section 2.2.Finally, the rated current of the valve‐side bushing is AC+DC, when the spectrum analysis of the measured current waveform at both ends of the bushing in Figure 3 is carried out; the current includes other harmonic components in addition to the 50 Hz basic component, so the skin‐effect of harmonic current must be considered in the heat loss calculation [40]. Assuming that the g‐th harmonic current component is Ig, its heat loss is calculated as shown in Equation (12):12P(t)=∑g=1GIg2Rg(t′)$$\begin{equation}P(t) = \sum_{g = 1}^G {I_g^2} {R_g}(t^{\prime})\end{equation}$$where Rg(t′)${R_g}(t^{\prime})$is the harmonic resistance, whose calculation method is shown in Equation (13):13Rg(t′)=ρt′Lπ▵dg(D−▵dg)▵dg=ρt′πfgμ$$\begin{equation} \left\{ \begin{aligned} {{R}_{g}}({t}')&=\frac{{{\rho }_{{{t}'}}}L}{\pi \vartriangle {d}_{g}( D-\vartriangle {d}_{g} )} \\ \vartriangle {{d}_{g}}&=\sqrt{\frac{{{\rho }_{{{t}'}}}}{\pi {{f}_{g}}\mu }} \end{aligned}\right.\end{equation}$$where D is the bushing diameter; ρt′${\rho _{t^{\prime}}}$the resistivity; L the length of the conduit; μ the magnetic resistance; fg the current frequency.Based on the analysis of the above steps and combined with the physical experiment platform, the digital twin built by COMSOL is shown in Figure 4.4FIGUREConstruction of digital twin for valve‐side bushing of UHV converter transformerZero‐sample state evaluation process based on attribute analysisThe factors that affect the running state of the valve‐side bushing are diverse, which inevitably leads to problems such as diverse failure forms and weak correlation. Typical failures include SF6 gas leakage, partial discharge, aging of the connection parts, and contact corrosion of the watchband. In the face of different fault types, their fault mechanisms are also very different, so it is difficult to extract fault feature data.To solve the above problems, this paper proposes a zero‐sample state evaluation method for valve‐side bushing based on digital twin technology and attributes analysis. The specific state evaluation flow diagram is shown in Figure 5.5FIGUREEvaluation process of valve‐side bushing state of UHV converter transformerIt can be seen from Figure 5 that the zero‐sample state evaluation of valve‐side bushing based on attribute analysis is mainly composed of four parts: physical entity analysis, simulation analysis of virtual space, the establishment of state feature set based on data twins, and running state evaluation.1) Physical entity analysisBased on the analysis of physical characteristics (attribute parameters and electrical characteristics of ±800kV valve‐side bushing in Section 3.1), the physical entity analysis of valve‐side bushing is completed in combination with running environment temperature (insulation oil temperature and hall temperature).2) Simulation of virtual spaceFirst, COMSOL is used to set geometric, material properties and physical fields; then, the measured carrier current is analyzed in the frequency domain, and it is decomposed into decomposition components with different frequency centres by EMD; Finally, combined with the heat loss calculation model in Section 3.2, the digital twins of valve side bushing are established.3) Establishment of state feature set based on data twinThrough the analysis of the digital twin, the bushing heat distribution is taken as the research object, and the characteristics data of the bushing external maximum temperature, external minimum temperature, internal maximum temperature, and internal minimum temperature under different carrier current decomposition components are obtained.4) Running status evaluationThe initial feature matrix is established by using the bushings with different defect degrees, and the running state evaluation of the valve‐side bushing is completed by combining the fuzzy clustering algorithm flow in Section 2.3.EXAMPLE VERIFICATION AND ANALYSISConstruction of digital twins for the valve‐side bushingCurrent decomposition of the valve‐side bushingThe carrier voltage and current of the valve‐side bushing are related to the working mode of the converter transformer. To establish the corresponding digital twins of the bushing and extract more detailed features, it is necessary to decompose other detailed components except for the 50 Hz fundamental component.Taking the dry‐type SF6 gas insulated bushing of phase A at Y‐Y side of ±800kV UHV as an example, its input signal waveform is shown in Figure 3. The initial current waveform is decomposed at different centre frequencies using EMD algorithm. The decomposed intrinsic mode function component imf(t) and residual component r(t) are shown in Figure 6.6FIGUREEMD decomposition of bushing current at valve‐side of UHV converter transformerAs shown in Figure 6, after the bushing current is processed by the EMD algorithm, there are four intrinsic mode componentsimf1(t)$im{f_{\rm{1}}}(t)$∼imf4(t)$im{f_{\rm{4}}}(t)$and one residual component r(t), and r(t) meets the termination condition of EMD.Simulation analysis of virtual space based on sample attributesWhen the valve‐side bushing has fault defects, the most direct features are the changes in DC impedance and carrier current. Although the external thermal feature distribution of the bushing can be obtained through infrared testing, it is difficult to obtain the internal features. Combined with the basic attributes, DC impedance, and other features of the physical entity of the valve‐side bushing in Section 3.1, the EMD results of the carried current are input into the virtual model, and the digital twin technology is used to obtain the internal characteristics under the simulation analysis.The simulation results of UHV ±800kV dry‐type SF6 gas‐insulated bushing in COMSOL are shown in Figure 7.7FIGURESimulation results of axial heat distribution of valve‐side bushingAs shown in Figure 7, the temperature of the simulated bushing at the insulating oil side of the transformer is about 73°C, and the highest temperature in the axial temperature distribution of the bushing occurs at the connection of the aluminium–copper conduit. When the bushing fails, the most obvious change is the temperature at the connection. Therefore, to obtain the state features of the bushing, the section at the connection is simulated and intercepted as shown in Figure 8.8FIGURESimulation results of radial heat distribution of valve‐side bushingEffectiveness analysis of thermal distribution of twinsTo verify the effectiveness of the digital twin simulation analysis in the above section, the axial thermal distribution of physical entities and virtual entities of UHV ±800kV valve‐side dry‐type SF6 gas‐insulated bushing is compared and analyzed. The maximum temperature results are shown in Figure 9.9FIGUREComparison of axial temperature distribution between the physical entity and digital twin of the valve‐side bushingAs shown in Figure 9, the maximum temperature range of the bushing is 69.37–100.15°C, while the normal reference value is 120°C, so the bushing is under normal operation. By comparing the axial heat distribution of physical entities and digital twins, the average error is not more than 5%, which meets the one‐to‐one mapping relationship. Then, the internal characteristics of the bushing can be obtained through the digital twin, which lays the foundation for state data twinning.Defect analysis of bushing and establishment of state feature setTaking a ±500 kV power plant as an example, the picture of the capacitor core when the valve‐side bushing is faulty is shown in Figure 10.10FIGUREThe capacitor core of the valve‐side bushing is faultyAccording to field detection, the blackening substance of the capacitor core is the substance produced after the local temperature rises due to the rise of the impedance at the copper–aluminium connection, and then local flashover occurs. Analyse other faults such as SF6 gas leakage, and the change of impedance of the bushing conductive tube is the first. Therefore, in the face of the diversity of fault defect forms of the bushing and the features of fewer fault data in operation and maintenance management, the use of digital twins to obtain twin data under different impedance changes is conducive to achieving zero‐sample state evaluation under online monitoring.In the process of establishing the state feature set, first, the digital twin is established according to the decomposition result of the carrier current and the impedance of the catheter in the physical entity, and then the maximum temperature and minimum temperature of the catheter are obtained. Then, the initial state feature set is established. Finally, to reduce the dimension of the feature set, principal component analysis is used to map the initial state features into one‐dimensional space. Based on the analysis of ±800kV valve‐side dry‐type SF6 gas‐insulated bushing under normal conditions in Section 4.1, the state features under different decomposition components are shown in Table 3.3TABLEState features of valve‐side bushing under normal conditionsDecomposition of componentimf1(t)imf2(t)imf3(t)imf4(t)r(t)Maximum temperature of conduit (°C)75.875.875.875.875.8Minimum temperature of conduit (°C))48.948.948.948.948.9Maximum external temperature (°C)35.135.135.135.135.1Minimum external temperature (°C)20.820.820.820.820.8To complete the dimension reduction analysis of the feature set, the principal component analysis of the feature set in Table 3 under normal conditions is carried out, and the results are as follows:xt=184.376.190.510.51$$\begin{equation*}{{\bf{x}}_t}{\rm{ = }}\left[ {{\rm{184}}{\rm{.37}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{6}}{\rm{.19}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{0}}{\rm{.51}}{\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\kern 1pt} {\rm{0}}{\rm{.51}}} \right]\end{equation*}$$Case study on evaluation of bushing state at converter valve‐sideTo realize the state evaluation of the valve‐side bushing of the UHV converter transformer, the state feature vectors of the bushing under different defects are established according to the method of establishing the feature vectors under normal conditions in the above section. In the example demonstration, x1∼x5 are used to represent the state feature vectors of the different bushing. x1 is the bushing with the impedance of 52.2 μΩ under normal condition, and x2∼x5 are the bushing with defect impedance of 120.4, 180.6, 200.3, and 260.8 μΩ.Grey relation analysisTo further verify the effectiveness of the fuzzy clustering algorithm in the state evaluation, this paper analyses the initial matrix X by citing the grey correlation coefficient, in which the calculation method of the grey correlation coefficient is as follows:ξi(k)=minimink|x1(k)−xi(k)|+ρmaximaxk|x1(k)−xi(k)||x1(k)−xi(k)|+ρmaximaxk|x1(k)−xi(k)|,$$\begin{equation*}{\xi _i}(k) = \frac{{\mathop {\min }\limits_i \mathop {\min }\limits_k |{x_1}(k) - {x_i}(k)| + \rho \mathop {\max }\limits_i \mathop {\max }\limits_k |{x_1}(k) - {x_i}(k)|}}{{|{x_1}(k) - {x_i}(k)| + \rho \mathop {\max }\limits_i \mathop {\max }\limits_k |{x_1}(k) - {x_i}(k)|}},\end{equation*}$$where ξi(k)${\xi _i}(k)$is the correlation coefficient of x0 to xi at point k, minimink|x0(k)−xi(k)|$\mathop {\min }\limits_i \mathop {\min }\limits_k |{x_0}(k) - {x_i}(k)|$is the absolute value of the second level minimum difference of x1 to xi at point k, ρ Is the grey resolution coefficient, generally 0.5, and the final grey correlation is ri:ri=1n∑k=1nξi(k).$$\begin{equation*}{r_i} = \frac{1}{n}\sum_{k = 1}^n {{\xi _i}(k)} .\end{equation*}$$Finally, the grey correlation coefficient of x1 in the initial characteristic matrix X is shown in Figure 11.11FIGUREDistribution of grey correlation coefficientAs shown in the Figure 11, the grey correlation coefficients of x1 are very similar; the grey correlation degrees are 0.776, 0.795, 0.785, and 0.795, respectively. Moreover, the correlation degrees of x3 and x4 are the same, so effective bushing state evaluation cannot be conducted.Fuzzy clustering analysisAccording to the fuzzy clustering algorithm shown in Algorithm 2 in Section 2.3, the zero‐sample state evaluation process of the valve‐side bushing is as follows.1) Initial feature matrix X.X=[x1;x2;x3;x4;x5]$$\begin{equation*}{\bf{X}} = [{{\bf{x}}_1};{{\bf{x}}_2};{{\bf{x}}_3};{{\bf{x}}_4};{{\bf{x}}_5}]\end{equation*}$$2) Standardize with Equations (5) and (6).X′′=000.2140.5050.3360.533000.5560.7210.6220.7740.7350.8740.6540.7771111$$\begin{equation*} {{\bf{X}}^{{\bf{^{\prime\prime}}}}} = \left[ \def\eqcellsep{&}\begin{array}{cccc} 0 & 0 & 0.214 & 0.505\\[2pt] 0.336 & 0.533 & 0& 0\\[2pt] 0.556 & 0.721 & 0.622 & 0.774\\[2pt] 0.735 & 0.874 & 0.654 & 0.777\\[2pt] 1 & 1 & 1& 1\end{array} \right]\end{equation*}$$3) Use Equation (7) to establish matrix RR=10.6630.4440.2670.0040.66310.7800.6030.3400.4440.78010.8220.5590.2670.6030.82210.7360.0040.3400.5590.7361$$\begin{equation*}{\bf{R}} = \left[ \def\eqcellsep{&}\begin{array}{ccccc} 1 & 0.663 & 0.444 & 0.267 & 0.004\\[2pt] 0.663 & 1 & 0.780 & 0.603 & 0.340\\[2pt] 0.444 & 0.780 & 1 & 0.822 & 0.559\\[2pt] 0.267 & 0.603 & 0.822 & 1 & 0.736\\[2pt] 0.004 & 0.340 & 0.559 & 0.736 & 1\end{array} \right]\end{equation*}$$4) As shown in Equations (8) and (9), the matrix is established after making matrix R transitive, and the cluster diagram formed when λ=min1≤i,j≤n{rij}$\lambda = \mathop {\min }\limits_{1 \le i,j \le n} \{ {r_{ij}}\} $is shown in Figure 12.12FIGURECluster diagram of valve‐side bushing statue evaluationAs shown in Figure 12, when λ is 0.664, the bushing x1 and other bushings are classified, so the similarity λ = 0.664 can be used as a criterion for bushing evaluation. The higher the similarity with normal conditions, the better the bushing status. The lower the similarity with normal conditions, the lower the bushing status.In conclusion, using the attribute parameters of physical entities to establish a digital twin, the internal features of the data twin can be obtained, and then the zero‐sample state evaluation of the valve‐side bushing can be realized.CONCLUSIONThe valve‐side bushing of the UHV converter transformer is the key equipment to realize the safe transmission of electric power. This paper conducts attribute analysis and research on the physical entity and uses the digital twin technology to obtain the internal state feature set under the defect state. Through case analysis and research, when the similarity of the fuzzy clustering algorithm is taken as the index, its zero‐sample state evaluation can be achieved, and the following conclusions are drawn:In the physical entity analysis of the valve‐side bushing, the research includes both the basic physical characteristics and the electrical characteristics. To decompose the different centre frequency components of the carrier current, the EMD algorithm is used to decompose the 50 Hz fundamental component and other harmonic components, which further improves the accuracy of establishing the digital twin.In the process of establishing the digital twin of the valve‐side bushing, the mechanism of the bushing with different defects is analyzed, its state characteristics are analyzed according to the thermal distribution phenomenon, and the validity of the data twin is verified by using the axial temperature distribution of the digital twin.In the zero‐sample state evaluation of valve‐side bushing, the initial state feature set is established by using the temperature distribution at the copper aluminium joint and the external temperature distribution under different decomposition components of bushing current carrying, and the fuzzy clustering algorithm is used to calculate the valve‐side bushing under different defect degrees. Finally, the zero‐sample state evaluation is realized by similarity matching, which provides a guarantee for equipment maintenance and safe operation.Herein, the preliminary assessment and analysis of the valve‐side bushing state have been realized. The following problems need to be solved in future research:More problems such as stability and applicability of valve‐side bushing state assessment under different application environments need to be solved urgently.How to realize the residual life prediction under complex running conditions is conducive to the establishment of a modern health management system.AUTHOR CONTRIBUTIONSZheng Li: writing ‐ original draft. Kai Liu: data curation; formal analysis; writing ‐ review and editing. Mu Lin: investigation; validation. Dongli Xin: data curation. Hao Tang: resources. Guangning Wu: methodology.ACKNOWLEDGEMENTThis research was supported in part by the Science and Technology Project of Headquarter of SGCC: SGTYHT/19‐JS‐215.CONFLICT OF INTERESTThere is no conflict of interest.DATA AVAILABILITY STATEMENTThe data that support the findings of this study are available from the corresponding author upon reasonable request.REFERENCESWang, L., Tao, Z., Zhu, L., et al.: Optimal dispatch of integrated energy system considering integrated demand response resource trading. IET Gener. Transm. Distrib. 16, 1727–1742 (2022)Chen, M., Liu, X., Shao, Y., et al.: Cracking risk analysis and control for high‐voltage dry‐type valve‐side bushings. IET Gener. Transm. Distrib. 14, 6555–6561 (2020)Mikhak‐Beyranvand, M., Faiz, J. Rezaeealam, B.: Thermal analysis and derating of a power transformer with harmonic loads. IET Gener. Transm. Distrib. 14, 1233–1241 (2020)Maximov, S., Olivares‐Galvan, J. C., Magdaleno‐Adame, S., et al.: New analytical formulas for electromagnetic field and eddy current losses in bushing regions of transformers. IEEE Trans. on Magn. 51(4), 1–10 (2015)Su, C., Branch, D.: On‐site Testing requirements of ultra‐high voltage converter transformer insulating Oil and SF_6 Gas. Elect. Engin. (07): 146–149 +160 (2016)Zhang, S., Peng, Z., Peng, L., et al.: Design and dielectric characteristics of the ±1100 kV UHVDC wall bushing in china. IEEE Trans. on Diel. & Elect. Insul. 22(1), 409–419 (2015)Chen, M., Liu, X., Liang, C., et al.: Study on surface charge Accumulation characteristics of resin impregnated paper wall bushing core under positive DC voltage. Energies. 12, 4420 (2019)Zhang, L., Shuai, X., Han, C., et al.: Valve side lead exit insulation design used in ±800 UHV converter transformer. In Proceedings of the International Conference on Condition Monitoring and Diagnosis (CMD), pp. 452–455. IEEE, Piscataway, NJ 2016.Liu, K., Yang, Z., Wei, W., et al.: Novel detection approach for thermal defects: study on its feasibility and application to vehicle cables. High Voltage. 1–10 (2022)Shi, Y. U., Nie, D., Weimin, M. A., et al.: Overvoltage and insulation coordination of Jinping‐Sunan ±800 kV UHVDC project converter stations. High Volt. Engin. 38(12), 3219–3223 (2012)Liu, Z., Chen, C., Zhang, C., et al.: Data super‐network fault prediction model and maintenance strategy for mechanical product based on digital twin. IEEE Access 7, 177284–177296 (2019)Alarcón, Serrano, ngel, Madrid, et al.: The role of digital twins in personalized sleep medicine. In German‐Italian Workshop, Social Innovation in Long‐Term Care through Digitalization. Springer, Cham (2022)Jim, G.: Software specialist, rFpro develops digital modelling tool of Applus+ IDIADA proving grounds for real world trials of autonomous vehicles. Truck & Bus Builder: The International Newsletter of Commercial Vehicle Manufacturing, Developments. 40(7), 10–20 (2018)Shi, Y., Xu, J., Du, W.: Discussion on the new operation management mode of hydraulic engineering based on the digital twin technique. J. Phys.: Confer. Seri., 1168, 022044 (2019)Tao, F., Zhang, M., Liu, Y., et al.: Digital twin driven prognostics and health management for complex equipment. CIRP Anna. 67(1), 169–172 (2018)Jia, M., Shen, C., Chen, Y., et al.: Digital twin of the energy internet and its application. Global Ener. Inter. 3(1), 1–13 (2020)Zhou, M., Yan, J., Feng, D.: Digital twin framework and its application to power grid online analysis. CSEE J. Power Ener. Syst. 5(3), 391–398 (2019)He, W., Wu, Z., Ren, X., et al.: Research on the application of digital twin technique in high voltage cable. In 2020 4th International Conference on Power and Energy Engineering (ICPEE), pp. 90–93. IEEE, Piscataway, NJ (2022)Elsisi M., Tran M., Mahmoud K., et al.: Effective IoT‐based deep learning platform for online fault diagnosis of power transformers against cyberattacks and data uncertainties. Measurement 190, 110686 (2022)Li, S., Cao, B., Cui, Y., et al.: Terahertz‐based insulation delamination defect inspection of vehicle cable terminals. IEEE Transactions on Transportation Electrification, 1–9 (2022)Leong, Y., Ker, P., Jamaludin, M. M.: UV‐vis spectroscopy: A new approach for assessing the color index of transformer insulating oil. Sensors. 18(7), 2175 (2018)Du, Z., Nie, D., Zhang, L., et al.: Insulation state evaluation of high voltage bushing based on frequency domain dielectric spectrum analysis. High Voltage Appliances. 49(11), 6–11 (2013)Monga, S., Gorur, R. S., Hansen, P., et al.: Design optimization of high voltage bushing using electric field computations. IEEE Trans. on Diel. & Elect.l Insul. 13(6), 1217–1224 (2006)Sarkar, S., Sharma, T., Baral, A., et al.: An expert system approach for transformer insulation diagnosis combining conventional diagnostic tests and PDC, RVM data. IEEE Transactions on Dielectrics and Electrical Insulation. 21(2), 882–891 (2014)Nie, D., Zhang, H., Chen, Z., et al.: Optimization design of grading ring and electrical field analysis of 800 kV UHVDC wall wall bushing. IEEE Trans. on Dielec. & Elec. Insulation. 20(4), 1361–1368 (2013)Radakovic, Z., Cardillo, E., Schaefer, M., et al.: Design of the winding‐bushing interconnections in large power transformers. Elect. Engin. 88(3), 183–190 (2006)Záliš, K.: Using expert systems in evaluation of the state of high voltage machine insulation systems. Acta Polytechnica. 40(5), 185–190 (2000)Liao, R., Du, Y., Yang, L., et al.: Quantitative diagnosis of moisture content in oil‐paper condenser bushing insulation based on frequency domain spectroscopy and polarisation and depolarisation current. IET Gener. Trans. Distr. 11(6), 1420–1426 (2017)Elsisi M.: Improved grey wolf optimizer based on opposition and quasi learning approaches for optimization: Case study autonomous vehicle including vision system. Arti. Intel. Revi. 55, 5597–5620 (2022)Ismail M. M., Bendary A. F., Elsisi M.: Optimal design of battery charge management controller for hybrid system PV/wind cell with storage battery. Int. J. Power Ener. Conv. 11(4), 412 (2020)Elsisi M., Tran M. Q.: Development of an iot architecture based on a deep neural network against cyber attacks for automated guided vehicles. Sensors 21(24), 8467 (2021)Tran M. Q., Elsisi M., Liu M. K., et al.: Reliable deep learning and iot‐based monitoring system for secure computer numerical control machines against cyber‐attacks with experimental verification. IEEE Access 10, 23186–23197 (2022)Elsisi M., Soliman M., Aboelela M., et al.: ABC based design of PID controller for two area load frequency control with nonlinearities. Telk. Indo. J. Elec. Eng. 16(1), 58–64 (2015)Elsisi M., Zaini H. G., Mahmoud K., et al.: Improvement of trajectory tracking by robot manipulator based on a new co‐operative optimization algorithm. Mathematics 9(24), 3231 (2021)Li, S., Cao, B., Cui, Y., et al.: Nonintrusive inspection of moisture damp in composited insulation structure based on terahertz technology. IEEE Transactions on Instrumentation and Measurement. 70, 1–10 (2021)Yu, Z., Chen, H., You, J., et al.: Adaptive fuzzy consensus clustering framework for clustering analysis of cancer data. IEEE/ACM Trans. Comp. Biol. Bioin. 12(4), 887–901 (2015)Grieves, M. W., Virtually Perfect: Driving Innovative and Lean Products through Product Lifecycle Management. Space Coast Press, Cocoa Beach (2011)Grieves, M. W., Vickers, M.: Digital twin :mitigatingunpredictable, undesirable emergent behavior in complexsystems. Transdisciplinary Perspectives on Complex Systems, pp. 85–113. Cham, Springer (2016)Maximov, S., Escarela‐Perez, R., Magdaleno‐Adame, S., et al.: Calculation of nonlinear electromagnetic fields in the steel wall vicinity of transformer bushings. IEEE Trans. on Magn. 51(6), 1–11 (2015)Wen, M., Lin, X., Zhong, S.: Numerical simulation and experimental study of heat transfer characteristics in transformer bushing, High Volt. Eng. 42(9), 2956–2962 (2016)Du, B., Tian, M., Su, J., et al.: Electrical tree growth characteristics in epoxy resin with harmonic superimposed DC voltage. IEEE Access 7, 47273–47281 (2019)Li, C., Cerrada, M., Cabrera, D., et al.: A comparison of fuzzy clustering algorithms for bearing fault diagnosis. J. Intell. Fuzzy Syst. 34(6), 3565–3580 (2018)

Journal

IET Generation Transmission & Distribution – Wiley

Published: Mar 1, 2023

Keywords: contact resistance; finite element analysis; HVDC power transmission; polymer insulators; SF6 insulation; state estimation

References

  • Optimal dispatch of integrated energy system considering integrated demand response resource trading
    Wang, L.; Tao, Z.; Zhu, L.
  • Cracking risk analysis and control for high‐voltage dry‐type valve‐side bushings
    Chen, M.; Liu, X.; Shao, Y.
  • Thermal analysis and derating of a power transformer with harmonic loads
    Mikhak‐Beyranvand, M.; Faiz, J.; Rezaeealam, B.
  • New analytical formulas for electromagnetic field and eddy current losses in bushing regions of transformers
    Maximov, S.; Olivares‐Galvan, J. C.; Magdaleno‐Adame, S.
  • On‐site Testing requirements of ultra‐high voltage converter transformer insulating Oil and SF_6 Gas
    Su, C.; Branch, D.
  • Design and dielectric characteristics of the ±1100 kV UHVDC wall bushing in china
    Zhang, S.; Peng, Z.; Peng, L.
  • Study on surface charge Accumulation characteristics of resin impregnated paper wall bushing core under positive DC voltage
    Chen, M.; Liu, X.; Liang, C.
  • Valve side lead exit insulation design used in ±800 UHV converter transformer
    Zhang, L.; Shuai, X.; Han, C.
  • Novel detection approach for thermal defects: study on its feasibility and application to vehicle cables
    Liu, K.; Yang, Z.; Wei, W.
  • Overvoltage and insulation coordination of Jinping‐Sunan ±800 kV UHVDC project converter stations
    Shi, Y. U.; Nie, D.; Weimin, M. A.
  • Data super‐network fault prediction model and maintenance strategy for mechanical product based on digital twin
    Liu, Z.; Chen, C.; Zhang, C.
  • The role of digital twins in personalized sleep medicine
    Alarcón, Serrano; ngel, Madrid
  • Software specialist, rFpro develops digital modelling tool of Applus+ IDIADA proving grounds for real world trials of autonomous vehicles
    Jim, G.
  • Discussion on the new operation management mode of hydraulic engineering based on the digital twin technique
    Shi, Y.; Xu, J.; Du, W.
  • Digital twin driven prognostics and health management for complex equipment
    Tao, F.; Zhang, M.; Liu, Y.
  • Digital twin of the energy internet and its application
    Jia, M.; Shen, C.; Chen, Y.
  • Digital twin framework and its application to power grid online analysis
    Zhou, M.; Yan, J.; Feng, D.
  • Research on the application of digital twin technique in high voltage cable
    He, W.; Wu, Z.; Ren, X.
  • Effective IoT‐based deep learning platform for online fault diagnosis of power transformers against cyberattacks and data uncertainties
    Elsisi, M.; Tran, M.; Mahmoud, K.
  • Terahertz‐based insulation delamination defect inspection of vehicle cable terminals
    Li, S.; Cao, B.; Cui, Y.
  • UV‐vis spectroscopy: A new approach for assessing the color index of transformer insulating oil
    Leong, Y.; Ker, P.; Jamaludin, M. M.
  • Insulation state evaluation of high voltage bushing based on frequency domain dielectric spectrum analysis
    Du, Z.; Nie, D.; Zhang, L.
  • Design optimization of high voltage bushing using electric field computations
    Monga, S.; Gorur, R. S.; Hansen, P.
  • An expert system approach for transformer insulation diagnosis combining conventional diagnostic tests and PDC, RVM data
    Sarkar, S.; Sharma, T.; Baral, A.
  • Optimization design of grading ring and electrical field analysis of 800 kV UHVDC wall wall bushing
    Nie, D.; Zhang, H.; Chen, Z.
  • Design of the winding‐bushing interconnections in large power transformers
    Radakovic, Z.; Cardillo, E.; Schaefer, M.
  • Using expert systems in evaluation of the state of high voltage machine insulation systems
    Záliš, K.
  • Quantitative diagnosis of moisture content in oil‐paper condenser bushing insulation based on frequency domain spectroscopy and polarisation and depolarisation current
    Liao, R.; Du, Y.; Yang, L.
  • Improved grey wolf optimizer based on opposition and quasi learning approaches for optimization: Case study autonomous vehicle including vision system
    Elsisi, M.
  • Optimal design of battery charge management controller for hybrid system PV/wind cell with storage battery
    Ismail, M. M.; Bendary, A. F.; Elsisi, M.
  • Development of an iot architecture based on a deep neural network against cyber attacks for automated guided vehicles
    Elsisi, M.; Tran, M. Q.
  • Reliable deep learning and iot‐based monitoring system for secure computer numerical control machines against cyber‐attacks with experimental verification
    Tran, M. Q.; Elsisi, M.; Liu, M. K.
  • ABC based design of PID controller for two area load frequency control with nonlinearities
    Elsisi, M.; Soliman, M.; Aboelela, M.
  • Improvement of trajectory tracking by robot manipulator based on a new co‐operative optimization algorithm
    Elsisi, M.; Zaini, H. G.; Mahmoud, K.
  • Nonintrusive inspection of moisture damp in composited insulation structure based on terahertz technology
    Li, S.; Cao, B.; Cui, Y.
  • Adaptive fuzzy consensus clustering framework for clustering analysis of cancer data
    Yu, Z.; Chen, H.; You, J.
  • Virtually Perfect: Driving Innovative and Lean Products through Product Lifecycle Management
    Grieves, M. W.
  • Transdisciplinary Perspectives on Complex Systems
    Grieves, M. W.; Vickers, M.
  • Calculation of nonlinear electromagnetic fields in the steel wall vicinity of transformer bushings
    Maximov, S.; Escarela‐Perez, R.; Magdaleno‐Adame, S.
  • Numerical simulation and experimental study of heat transfer characteristics in transformer bushing
    Wen, M.; Lin, X.; Zhong, S.
  • Electrical tree growth characteristics in epoxy resin with harmonic superimposed DC voltage
    Du, B.; Tian, M.; Su, J.
  • A comparison of fuzzy clustering algorithms for bearing fault diagnosis
    Li, C.; Cerrada, M.; Cabrera, D.