An ultra-high voltage converter valve state evaluation method, medium and system

Abstract

The application provides a method, medium and system for evaluating the state of an extra-high voltage converter valve, and belongs to the technical field of state evaluation of an extra-high voltage converter valve.The method, medium and system comprise: obtaining and preprocessing operation parameters of the converter valve; establishing a layered and graded state evaluation index system of the converter valve according to the structural features of the converter valve; calculating the weight of each level evaluation index in the evaluation index system; calculating each level index data in the evaluation index system according to the operation parameters; establishing a membership function of each evaluation index based on fuzzy theory; calculating the membership of each level evaluation index according to the preprocessing index data by using the membership function; and performing graded fuzzy comprehensive evaluation on the obtained membership, and visually outputting the membership probability of the normal state, the attention state, the abnormal state and the serious state to which the converter valve and its components belong.The application can solve the technical problem of inaccurate evaluation of the state of an extra-high voltage converter valve in the prior art.

Description

Technical Field

[0001] This invention belongs to the technical field of ultra-high voltage converter valve condition assessment, specifically, it relates to an ultra-high voltage converter valve condition assessment method, medium, and system. Background Technology

[0002] With the increasing demands for high reliability and energy efficiency in power systems, higher requirements are being placed on the condition monitoring and fault diagnosis technologies for key power equipment. As a core component of UHVDC transmission projects, the operating status of ultra-high voltage (UHV) converter valves is crucial to the safety and stability of the DC system. In the context of UHV and a high proportion of renewable energy, the application of artificial intelligence (AI) technology to power systems can fully utilize grid resources and save on operation and maintenance costs, showing broad application prospects. The smart grid is extending and developing from system intelligence to equipment intelligence, which will unify and standardize the information system, monitoring system, evaluation system, and management system of power equipment. UHV converter valves are one of the key components in UHVDC transmission systems, and their operating status directly affects the safe and stable operation of the entire DC transmission system. Traditional converter valve condition monitoring and fault diagnosis methods have many shortcomings, failing to provide accurate early warning and location diagnosis of converter valve faults. Therefore, there is an urgent need to develop an advanced converter valve condition evaluation method.

[0003] Currently, the main methods for evaluating the condition of converter valves include fault tree analysis, expert system methods, and fuzzy evaluation methods. Fault tree analysis constructs a fault tree model of the converter valve, analyzes the logical relationships between various fault events, determines the root cause and probability of occurrence of converter valve faults, and provides early warning of faults. However, this method relies on expert experience and has a certain degree of subjectivity. Expert system methods use knowledge bases and reasoning mechanisms to simulate experts in fault analysis, but the construction of its rule base also depends on expert experience. Fuzzy evaluation methods use membership functions and fuzzy reasoning to qualitatively describe the condition of the converter valve and evaluate the health status of the equipment, but the quantitative analysis of the evaluation results is insufficient. Overall, existing evaluation methods suffer from problems such as inaccurate assessments. Summary of the Invention

[0004] In view of this, the present invention provides a method, medium and system for evaluating the status of ultra-high voltage converter valves, which can solve the technical problem of inaccurate evaluation of the status of ultra-high voltage converter valves in the prior art.

[0005] This invention is implemented as follows:

[0006] The first aspect of the present invention provides a method for evaluating the condition of an ultra-high voltage converter valve, comprising the following steps:

[0007] S10. Obtain the operating parameters of the converter valve and perform preprocessing. The operating parameters include voltage, current, sound, and image.

[0008] S20. Based on the structural characteristics of the converter valve, establish a hierarchical and graded evaluation index system for the converter valve status, wherein the evaluation index system includes multiple evaluation indicators.

[0009] S30. Calculate the weight of each level of evaluation index in the evaluation index system;

[0010] S40. Calculate the indicator data of each level in the evaluation indicator system based on the operating parameters;

[0011] S50. Based on fuzzy theory, establish the membership function of each evaluation index;

[0012] S60. Calculate the membership degree of each level of evaluation index based on the preprocessed index data using the membership function;

[0013] S70. Perform hierarchical fuzzy comprehensive evaluation on the obtained membership degrees, and visualize the membership probabilities of the converter valve and its components to the normal state, attention state, abnormal state and severe state.

[0014] Preferably, the method for preprocessing the voltage and current in the operating parameters is a multiphysics model simulation verification.

[0015] Based on the above technical solution, the ultra-high voltage converter valve condition evaluation method of the present invention can be further improved as follows:

[0016] The step of acquiring and preprocessing the operating parameters of the converter valve specifically includes:

[0017] Acquire raw signals from the converter valve collected by sensors or video terminals, including voltage, current, sound, and image signals;

[0018] Preprocessing is performed on the original signal, including filtering and noise reduction.

[0019] Effective features are extracted from the preprocessed signal and used as operating parameters for state assessment.

[0020] The beneficial effects of adopting the above-mentioned improvement scheme are: the accuracy and reliability of state assessment can be improved by using preprocessing and feature extraction.

[0021] The evaluation index system is divided into three levels according to the structural characteristics of the converter valve: the overall converter valve level, the internal component level, and the specific device level.

[0022] The method for calculating the weights of evaluation indicators at each level in the evaluation index system is the hierarchical single-ranking method.

[0023] The step of calculating the indicator data of each level in the evaluation indicator system based on the operating parameters also includes the step of performing fuzzy calculation on the indicator data of each level to obtain the quantitative value of the indicator.

[0024] The step of establishing the membership functions of each evaluation index based on fuzzy theory specifically includes:

[0025] Determine the linguistic variables and select linguistic variables to describe the state based on the status of the converter valve.

[0026] Establish membership functions and determine the membership functions corresponding to different language values ​​for each evaluation index to describe the membership relationship between index values ​​and language values.

[0027] Standardization transformation: Standardize each evaluation indicator and map it to the [0,1] interval;

[0028] Construct a membership matrix by calculating the membership degree of each language value corresponding to the index data based on the membership function of each evaluation index, and thus forming the membership matrix.

[0029] The step of performing hierarchical fuzzy comprehensive evaluation on the obtained membership degrees specifically includes:

[0030] The status of the converter valve is divided into four levels: normal, warning, abnormal, and critical.

[0031] Calculate the membership degree based on the hierarchical relationship of the indicator system;

[0032] Based on the obtained membership degree, the corresponding maximum state level is found and used as the result of hierarchical fuzzy comprehensive evaluation.

[0033] Furthermore, the membership probability is the membership degree corresponding to the hierarchical fuzzy comprehensive evaluation result.

[0034] A third aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, which, when executed, are used to perform the above-described ultra-high voltage converter valve status evaluation method.

[0035] The present invention provides an ultra-high voltage converter valve condition evaluation system, wherein the system includes the aforementioned computer-readable storage medium.

[0036] Compared with existing technologies, the beneficial effects of the ultra-high voltage converter valve condition evaluation method, medium, and system provided by this invention are:

[0037] 1. The evaluation indicator system is comprehensive and systematic.

[0038] This invention constructs a hierarchical evaluation index system for converter valve status. This system comprehensively considers various factors affecting the operating status of the converter valve from three levels: the overall converter valve, internal components, and specific devices. Multiple evaluation indicators are set under each level. This comprehensive evaluation index system can provide a three-dimensional evaluation of the converter valve status.

[0039] 2. The evaluation results are quantitatively accurate.

[0040] This invention achieves a quantitative description of the converter valve's state by collecting various operating parameters, calculating the membership degrees of indicators, and then performing fuzzy comprehensive evaluation. Compared with qualitative evaluation, the quantitative evaluation results are more accurate and reliable, and different fault states can be clearly quantitatively distinguished, providing a foundation for subsequent fault early warning and fault location.

[0041] 3. Accurate fault early warning and location

[0042] Based on the quantitative evaluation results of this invention, the overall health status probability of the converter valve and its different components can be intuitively output, such as the membership degree of normal, warning, abnormal, and severe fault states. This provides support for accurate fault early warning. Furthermore, by comparing the evaluation results of different levels of indicators, the root cause component leading to the fault can be identified, thereby improving the accuracy of fault location.

[0043] 4. The method is highly practical.

[0044] This invention utilizes fuzzy theory to achieve state evaluation, which closely aligns with the actual operation of converter valves. The weights of the indicators can be adjusted according to actual needs, making the method highly practical. Furthermore, the evaluation results can be displayed intuitively and visually, facilitating quick assessment of equipment status by on-site operators.

[0045] In summary, this invention solves the technical problem of inaccurate evaluation of the status of ultra-high voltage converter valves in the prior art. Attached Figure Description

[0046] Figure 1 A flowchart of a method for evaluating the condition of an ultra-high voltage converter valve provided by the present invention;

[0047] Figure 2 This is a schematic diagram illustrating the simulation verification of the multiphysics model in the operating parameters in an embodiment of the present invention;

[0048] Figure 3 This is a hierarchical diagram of the hierarchical evaluation index system in an embodiment of the present invention. Detailed Implementation

[0049] like Figure 1 The diagram shown is a flowchart of a first embodiment of a method for evaluating the condition of an ultra-high voltage converter valve provided by the first aspect of the present invention. The method in this embodiment includes the following steps:

[0050] S10. Obtain and preprocess the operating parameters of the converter valve. The operating parameters include voltage, current, sound, and image.

[0051] S20. Based on the structural characteristics of the converter valve, establish a hierarchical and graded evaluation index system for the converter valve status. The evaluation index system includes multiple evaluation indicators.

[0052] S30. Calculate the weight of each level of evaluation index in the evaluation index system;

[0053] S40. Calculate the data of each level of the evaluation index system based on the operating parameters;

[0054] S50. Based on fuzzy theory, establish the membership function of each evaluation index;

[0055] S60. Calculate the membership degree of each level of evaluation index based on the preprocessed index data using the membership function;

[0056] S70. Perform hierarchical fuzzy comprehensive evaluation on the obtained membership degrees, and visualize the membership probabilities of the converter valve and its components to the normal state, attention state, abnormal state and severe state.

[0057] The steps of obtaining and preprocessing the operating parameters of the converter valve in the above technical solution specifically include:

[0058] Acquire raw signals from the converter valve collected by sensors or video terminals, including voltage, current, sound, and image signals;

[0059] Preprocessing is performed on the original signal, including filtering and noise reduction.

[0060] Effective features are extracted from the preprocessed signal and used as operating parameters for state assessment.

[0061] The method for preprocessing voltage and current in the operating parameters is a multiphysics model simulation verification.

[0062] The voltage and current preprocessing method for the operating parameters is a multiphysics model simulation verification, which can be implemented as follows: Based on the existing control and protection system, valve control system, water cooling system, and external temperature and pressure sensors of the converter valve, a panoramic data set of electrical and multiphysics quantities of the converter valve equipment is collected. A dual-path multiphysics simulation model is built: one is a circuit simulation model based on Simulink, and the other is a thermal and pressure field simulation model based on Ansys. The circuit model is mainly based on the monitored voltage and current electrical quantities, while the thermal and pressure field models are mainly based on the temperature and pressure data collected by the multiphysics sensors. The two simulation models verify each other. The thermal and pressure fields of the converter valve under different operating conditions are converted into updated circuit parameters through an interactive program, while the electrical quantities in the circuit simulation reflect aging losses. The thermal and pressure fields are updated by updating the losses, thus verifying each other and ultimately obtaining an accurate multiphysics model of the converter valve that reflects the actual operating conditions, providing an accurate data source for the subsequent establishment of the indicator system.

[0063] In the above technical solution, the evaluation index system is divided into three levels according to the structural characteristics of the converter valve: the overall converter valve level, the internal component level, and the specific device level.

[0064] In the above technical solution, the method for calculating the weight of each level of evaluation index in the evaluation index system is the hierarchical single ranking method.

[0065] In the above technical solution, the step of calculating the indicator data of each level in the evaluation indicator system based on the operating parameters also includes the step of performing fuzzy calculation on the indicator data of each level to obtain the quantitative value of the indicator.

[0066] In the above technical solution, the step of establishing the membership function of each evaluation index based on fuzzy theory specifically includes:

[0067] Determine the linguistic variables and select linguistic variables to describe the state based on the status of the converter valve.

[0068] Establish membership functions and determine the membership functions corresponding to different language values ​​for each evaluation index to describe the membership relationship between index values ​​and language values.

[0069] Standardization transformation: Standardize each evaluation indicator and map it to the [0,1] interval;

[0070] Construct a membership matrix by calculating the membership degree of each language value corresponding to the index data based on the membership function of each evaluation index, and thus forming the membership matrix.

[0071] The step of performing hierarchical fuzzy comprehensive evaluation on the obtained membership degrees in the above technical solution specifically includes:

[0072] The status of the converter valve is divided into four levels: normal, warning, abnormal, and critical.

[0073] Calculate the membership degree based on the hierarchical relationship of the indicator system;

[0074] Based on the obtained membership degree, the corresponding maximum state level is found and used as the result of hierarchical fuzzy comprehensive evaluation.

[0075] Furthermore, in the above technical solution, the membership probability is the membership degree corresponding to the hierarchical fuzzy comprehensive evaluation result.

[0076] The specific implementation method of step S10 is described as follows:

[0077] Data Acquisition

[0078] The first step is to obtain the operating parameters of the converter valve, including voltage, current, sound, and image.

[0079] 1.1 Voltage Signal Acquisition

[0080] The voltage signal U(t) at the converter valve terminal is measured in real time using a voltage sensor, with a sampling frequency of 1000Hz.

[0081] 1.2 Current Signal Acquisition

[0082] The current signal I(t) at the converter valve terminal is measured in real time using a current sensor, with a sampling frequency of 1000Hz.

[0083] 1.3 Acquisition of Sound Signals

[0084] The sound signal S(t) generated during the operation of the converter valve is measured in real time using a specialized sound sensor. Considering the frequency characteristics of the sound signal, the sampling frequency is selected as 44100Hz.

[0085] 1.4 Image Signal Acquisition

[0086] The operating status of the converter valve is captured using a video terminal, resulting in an image sequence P(x,y,t), where P(x,y,t) represents the pixel value at coordinate point (x,y) in the original image. Considering the spatial resolution of the image, the video terminal's frame rate is set to 50fps, and the resolution is 1280×720.

[0087] Signal preprocessing

[0088] The acquired raw signals may contain noise interference, and directly using them for condition assessment can easily introduce errors. Therefore, necessary preprocessing is required.

[0089] 2.1 Voltage Signal Preprocessing

[0090] The acquired voltage signal U(t) is preprocessed as follows:

[0091] (1) Median filtering to remove scattered noise;

[0092] U′(t)=med{U(tT),...,U(t),...,U(t+T)};t represents the sampling time and T represents the sampling duration;

[0093] (2) Wavelet transform denoising

[0094] Design a low-pass filter based on wavelet transform to remove high-frequency noise. Let the wavelet transform be W(U′)(a,b), then the filtering process is as follows:

[0095] U″(t)=IW s (U′)(a,b)

[0096] Where I represents the inverse wavelet transform, W sLet a and b represent wavelet transform, and a and b represent scale and translation parameters.

[0097] 2.2 Current Signal Preprocessing

[0098] The same preprocessing method is used for the current signal I(t) as for the voltage signal.

[0099] Median filtering:

[0100] I′(t)=med{I(tT),...,I(t),...,I(t+T)};

[0101] Wavelet transform filtering:

[0102] I″(t)=IW s (I′)(a,b);

[0103] 2.3 Audio signal preprocessing

[0104] For the sound signal S(t):

[0105] (1) Bandpass filtering, retaining the main audio frequency band [100, 1000] Hz;

[0106] (2) Adaptive noise reduction: using the NLMS algorithm to adaptively filter out noise;

[0107] e(n) = d(n) - y(n);

[0108]

[0109] y(n)=W T (n)X(n);

[0110] Where d(n) is the input signal, y(n) is the filtered output signal, W(n) is the filter weight, X(n) is the block of the input signal, e(n) is the error, μ and δ are algorithm adjustment parameters, usually μ = 1 and δ = 0. μ and δ can be adjusted according to the noise reduction effect in actual use, and n is the number of audio signal sampling points, which is usually 1.

[0111] 2.4 Image Signal Preprocessing

[0112] Preprocess the image sequence P(x,y,t):

[0113] (1) Median filtering for noise reduction;

[0114] P′(x,y,t)=med{P(xM,yM,t),...,P(x,y,t),...,P(x+M,y+M,t)};M represents half the size of the filter template, and x and y are the horizontal and vertical coordinates of the image.

[0115] (2) Motion compensation: The motion vector field MV(x,y,t) is estimated using an optical flow algorithm, and motion compensation is performed.

[0116] P″(x,y,t)=P′(x+MV x ,y+MV y ,t);

[0117] In the formula, MV(x,y,t) represents the motion vector at position (x,y) at time t, which is estimated by the optical flow algorithm; MV x Represents the x-component of the motion vector; MV y Represents the y-component of the motion vector;

[0118] Parameter extraction

[0119] Extract the feature parameters needed for state assessment from the preprocessed signal.

[0120] 3.1 Electrical Parameter Extraction

[0121] Extract effective values ​​and frequency analysis features from U″(t) and I″(t).

[0122] RMS voltage value:

[0123]

[0124] RMS current value:

[0125]

[0126] Voltage spectrum characteristics:

[0127] P U (f) = |FFT(U″(t))|;

[0128] Current spectrum characteristics:

[0129] P I (f) = |FFT(I″(t))|;

[0130] 3.2 Acoustic Parameter Extraction

[0131] Features such as sound power and Mel-frequency cepstral coefficients (MFCC) are extracted from the sound signal S″(t). Sound power:

[0132]

[0133] MFCC characteristics:

[0134] MFCC=DCT(log(Mel(S″(t))));

[0135] Where Mel is the Mel filter bank and DCT is the Discrete Cosine Transform.

[0136] 3.3 Image Parameter Extraction

[0137] Features such as motion and texture can be extracted from the image sequence P″(x,y,t).

[0138] Image motion features:

[0139]

[0140] In the formula, This indicates that the summation result is normalized to obtain the average motion amplitude;

[0141] Image texture features:

[0142] F(t) = GLCM(P″(x,y,t));

[0143] Among them, GLCM is the gray-level co-occurrence matrix algorithm.

[0144] Preprocessing and feature extraction can improve the accuracy and reliability of state assessment.

[0145] The specific implementation of step S20 is as follows:

[0146] Indicator System Architecture

[0147] Based on the structural characteristics of the converter valve, evaluation indicators are selected and a hierarchical evaluation indicator system is established, such as... Figure 3 As shown:

[0148] Overall level of converter valve

[0149] Insulation parameters include air gap insulation and coil insulation.

[0150] Temperature rise indicators: including the heat generation level of the converter valve;

[0151] Air tightness index: including the air tightness performance of specific components;

[0152] The internal component hierarchy includes:

[0153] Main contact index: Monitors the contact status of the main contacts;

[0154] Auxiliary contact index: Monitors the contact status of auxiliary contacts;

[0155] Control switch indicators: Monitor the status of the control switch;

[0156] Filter metrics: the filtering effect of the filter;

[0157] The specific device level refers to the specific devices that make up each component.

[0158] The indicator weights include:

[0159] Assign weights to the indicators based on the severity of their impact on equipment condition:

[0160] The overall hierarchical weight of the converter valve is 0.4;

[0161] The internal component hierarchy weight is 0.3;

[0162] The specific device-level weight is 0.3;

[0163] The mathematical model establishment includes: using the analytic hierarchy process (AHP) to determine the weights of each indicator.

[0164] (1) Construct the judgment matrix

[0165] For each pair of indicators, an importance comparison is performed to establish a judgment matrix A = [a ij ];

[0166] Where a ij This indicates the comparison results of the importance of indicator i and indicator j.

[0167] (2) Hierarchical single sorting

[0168] Calculate the eigenvectors W of matrix A:

[0169] A W =λ max W;

[0170] W is the weight vector of each indicator.

[0171] (3) Consistency check

[0172] Calculate the consistency index:

[0173] m represents the number of indicators.

[0174] RI is the average random consistency index, when When the matrix passes the consistency test, CR represents the proportion of consistency indicators.

[0175] Fuzzy comprehensive evaluation includes: using the fuzzy comprehensive evaluation method of fuzzy mathematics to perform quantitative conversion of qualitative indicators.

[0176] 1. Define the linguistic evaluation set of the indicators: V = {Excellent, Good, Slightly Good, Average, Poor, Very Poor, Extremely Poor};

[0177] 2. Construct the membership function for language evaluation.

[0178] Determine the membership function

[0179] For each evaluation metric, determine the membership function for different language values:

[0180] Taking the temperature rise index of the converter valve as an example:

[0181]

[0182]

[0183] Similarly, other membership functions are established.

[0184] 3. Perform fuzzy calculations:

[0185]

[0186] Among them: A ij R is the language evaluation of the i-th indicator and the j-th indicator. ij This corresponds to the membership degree, B. i This is a comprehensive evaluation result; in this step, R... ij It is an unknown quantity.

[0187] 4 pairs of B i By maximizing the membership degree, we obtain a quantitative indicator, and simultaneously, we obtain R. ij The specific value.

[0188] This step allows us to obtain a scientific and reasonable evaluation index system for the condition of the converter valve.

[0189] In step S30, the weights of each level of the evaluation index system are calculated using a genetic algorithm to establish and iteratively optimize the weights. The objective function for optimization is set as the CR value of the hierarchical sorting. The optimal weight is obtained when the CR is minimized.

[0190] Step 1. Initialize the weight group

[0191] Randomly generate N weight vectors as the initial population: W1, W2, ..., W N

[0192] Each weight vector contains the weights of all M evaluation metrics: W i =(w i 1,w i 2,...,w i M), i = 1, 2, ..., N;

[0193] The weights are initialized to a range of [0,1] and then normalized.

[0194] Step 2. Evaluate FITNESS = CR;

[0195] For each weight vector W i Construct a judgment matrix and calculate the CR value;

[0196] The CR value serves as the fitness value for this weight vector, denoted as: FITNESS(W i ) = CR(W i )

[0197] Step 3. Selection, crossover, and mutation;

[0198] The optimal solution is selected based on the fitness probability and retained in the next generation;

[0199] Perform a crossover operation on the selected weight vectors to generate new weight vectors;

[0200] Further mutation operations are performed to enhance population diversity;

[0201] Selection, crossover, and mutation yield new weighted groups;

[0202] Step 4. Obtain the new weighted group

[0203] The new weight vectors obtained through selection, crossover, and mutation are used as the next generation population;

[0204] The new group contains a better weight vector;

[0205] Step 5. Repeat steps 2-4 until the termination condition is met;

[0206] Evaluate the fitness of the new group and find the optimal weights;

[0207] Repeat the optimization iterations until the termination condition is met:

[0208] The maximum number of iterations has been reached.

[0209] The population fits converge, and the populations are similar across multiple generations.

[0210] Genetic algorithms can effectively search for better weight vectors, that is, more scientific and reasonable evaluation index weights.

[0211] Step S30 allows for the objective and reasonable determination of the weights of each indicator in the state assessment model.

[0212] The specific implementation method of step S40 is described as follows:

[0213] Electrical parameter calculation

[0214] In step S10, the voltage signal U(t) and current signal I(t) of the converter valve were obtained and preprocessed. Now, the electrical parameters are calculated:

[0215] RMS voltage value:

[0216]

[0217] Where T is the sampling time length.

[0218] RMS current value:

[0219]

[0220] Voltage harmonic content:

[0221] Perform a fast Fourier transform to obtain the voltage spectrum X(f):

[0222] X(f)=FFT(U(t))

[0223] Calculate harmonic content:

[0224]

[0225] f1 and f2 represent the harmonic frequency range.

[0226] Current harmonic content:

[0227] Similarly, perform FFT to calculate the current harmonic content (IHD).

[0228] Sound parameter calculation

[0229] Obtain the sound signal S(t) in S10 and calculate the sound parameters:

[0230] Sound power:

[0231]

[0232] Sound entropy:

[0233]

[0234] Where p i It is the probability distribution of each sampling point of the sound signal.

[0235] Acoustic characteristics:

[0236] Perform FFT to calculate the acoustic characteristic spectrum:

[0237] F(f) = FFT(S(t))

[0238] The image parameters use the image motion feature M(t) and image texture feature F(t) from S10;

[0239] Fuzzy computing

[0240] Using MATLAB's fuzzy tools, fuzzy calculations are performed based on the mapping relationship between the indicator system and parameters to obtain the fuzzy quantification values ​​of each evaluation indicator.

[0241] Step S40 allows the extraction of characteristic parameter information required for evaluating the state from the signal.

[0242] The specific implementation method of step S50 is described as follows:

[0243] Determine language variables

[0244] Determine the set of linguistic variables based on the severity of the converter valve's condition:

[0245] V = {Excellent, Good, Average, Poor, Very Poor}

[0246] Determine the membership function

[0247] For each evaluation metric, determine the membership function for different language values:

[0248] Taking the temperature rise index of the converter valve as an example:

[0249]

[0250]

[0251] Similarly, other membership functions are established.

[0252] 3. Standardization Conversion

[0253] Because the evaluation indicators have different dimensions, they need to be standardized.

[0254]

[0255] Map the index value to the [0,1] interval.

[0256] 4. Fuzzy Matrix

[0257] Construct the membership matrix R:

[0258]

[0259] Step S50 achieves the fuzzy mapping from indicator data to language words, preparing for subsequent fuzzy comprehensive evaluation.

[0260] The specific implementation of step S60 is described as follows: the membership degree of each level of evaluation index is calculated by substituting the collected index data into the membership function of each level of evaluation index.

[0261] The specific implementation method of step S70 is described as follows:

[0262] Clustering classification

[0263] The membership degrees of each level of evaluation index obtained in S60 are clustered using the mean clustering algorithm to obtain the membership degrees of each state of the converter valve, which are represented by {excellent, good, average, poor, very poor}, and used to determine the state of the converter valve.

[0264] Visual presentation: The membership degree of each state is displayed intuitively using probability graphs and other methods.

[0265] Step S70 enables accurate assessment and intuitive display of the operating status of the converter valve.

[0266] The following is a second embodiment of the method of the present invention. In the second embodiment, the specific constants and variables are redefined, which are different from the specific constant and variable definitions used in the first embodiment:

[0267] Step 1: First, establish an evaluation index system for hierarchical and graded converter valves.

[0268] A bottom-up three-tiered evaluation index system for converter valves is constructed, comprising device evaluation, component evaluation, and converter valve evaluation. Specifically, evaluation index parameters are established for converter valves and different components, and are divided into two different levels according to the importance of the index parameters: general level and important level. The index parameters are divided into general state index parameters and important index parameters. Different weights are assigned to different levels. The calculation formula for the evaluation index parameters is shown in Equation (1).

[0269] Z=α*X+β*Y (1)

[0270] Where Z represents the evaluation result value, X is the sum of evaluation values ​​of general state index parameters, Y is the sum of evaluation values ​​of important index parameters, α represents the weight of general state index parameters, and β represents the weight of important index parameters. In this invention, the weights of α and β are selected as 0.2 and 0.8, respectively.

[0271] The indicators are categorized into four levels based on how close they are to the normal range: within the normal range, close to the normal boundary, outside the normal boundary, and severely outside the normal boundary. Each level is assigned a different score: 1 for within the normal range, 5 for close to the normal boundary, 8 for outside the normal boundary, and 10 for severely outside the normal boundary.

[0272] In order to unify the calculation of evaluation index results for converter valves and different components, this invention considers taking the average of the sum of general state index parameters and important state index parameters as the comprehensive evaluation value, and converting the comprehensive evaluation value into a percentage system. The calculation formulas are as follows (2) and (3).

[0273]

[0274]

[0275] Where, x i Let y be the evaluation value of the i-th general state index parameter. i Let be the evaluation value of the i-th important state indicator parameter, and n be the number of state indicator parameters.

[0276] Step 2: Evaluation and classification method based on fuzzy theory

[0277] The evaluation results of the converter valve and its components are divided into four levels: normal state, warning state, abnormal state, and critical state. The definitions of the evaluation result states are as follows.

[0278] (1) Normal state is defined as the status index parameters of the converter valve and each group of components are all within the normal range, and the equipment is operating normally.

[0279] (2) Note that the status definition mainly includes two situations. In one situation, the value of a single status indicator parameter changes, the value of an important status indicator parameter is close to the boundary of the normal range, or the value of a single general status indicator parameter exceeds the boundary of the normal range, and the equipment is operating normally without any abnormalities. In the other situation, the values ​​of multiple status indicator parameters change, the trend of multiple general status indicator parameter values ​​changes is close to the boundary of the normal range value, but does not exceed the boundary value, and the equipment is operating normally without any abnormalities.

[0280] (3) An abnormal state is defined as a change in the value of a single important state indicator parameter that exceeds the normal range boundary, or a change in the values ​​of multiple state parameters that exceeds the normal range boundary.

[0281] (4) A severe state is defined as a single important state indicator parameter that seriously exceeds the normal range boundary, or multiple state parameter values ​​that exceed the normal range boundary.

[0282] Based on the evaluation results, four membership functions for equipment states are constructed. The membership function for the normal state is shown in formula (4):

[0283]

[0284] Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the normal state.

[0285] Note that the membership function of a state is shown in formula (5):

[0286]

[0287] Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the attention state.

[0288] The membership function for abnormal states is shown in formula (6):

[0289]

[0290] Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the abnormal state.

[0291] The membership function for the severe state is shown in formula (7):

[0292]

[0293] Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the severe state.

[0294] Based on the evaluation index system and evaluation method, the evaluation result values ​​are obtained. Inputting these values ​​into formulas (4), (5), (6), and (7) yields the probability that the equipment's state belongs to different states. For example, inputting evaluation result values ​​of 43, 47, and 52 yields the probabilities of belonging to normal, alert, abnormal, and severe states as (0, 0.7, 0.3, 0), (0, 0.3, 0.7, 0), and (0, 0, 1, 0), respectively. Based on the magnitude of the probability values, state with an evaluation result value of 43 is highly likely to be in the alert state, but the probability of an abnormal state is 0.3. State with an evaluation result value of 47 is highly likely to be in the abnormal state, with the probability of the alert state gradually decreasing. State with an evaluation result value of 52 is in the abnormal state. This fuzzy theory-based evaluation and classification method for converter valves is more suitable for engineering applications.

[0295] In the above embodiments, the method for preprocessing voltage and current in the operating parameters is verified by multiphysics model simulation, which can be referred to as... Figure 2 The following method was adopted: Based on the existing control and protection system, valve control system, water cooling system, and external temperature and pressure sensors of the converter valve, panoramic data of electrical and multi-physical quantities of the converter valve equipment were collected. A dual-channel multi-physics simulation model was built: one channel was a circuit simulation model based on Simulink, and the other channel was a thermal and pressure field simulation model based on Ansys. The circuit model was mainly based on monitored electrical quantities such as voltage and current, while the thermal and pressure field models were mainly based on temperature and pressure data collected by the multi-physics sensors. The two simulation models were used for mutual verification. The thermal and pressure fields of the converter valve under different operating conditions were converted into updated circuit parameters through an interactive program. The electrical quantities in the circuit simulation reflected aging losses, and the thermal and pressure fields were updated by updating the losses. Through mutual verification, an accurate multi-physics model of the converter valve reflecting the actual operating conditions was obtained, providing an accurate data source for the subsequent establishment of the indicator system.

[0296] A third aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, which, when executed, are used to perform the above-described ultra-high voltage converter valve status evaluation method.

[0297] The present invention provides an ultra-high voltage converter valve condition evaluation system, wherein the system includes the aforementioned computer-readable storage medium.

[0298] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for evaluating the condition of an ultra-high voltage converter valve, characterized in that... This includes the following steps: S10. Obtain and preprocess the operating parameters of the converter valve, including voltage, current, sound, and image; S20. Based on the structural characteristics of the converter valve, establish a hierarchical and graded evaluation index system for the converter valve status, wherein the evaluation index system includes multiple evaluation indicators. S30. Calculate the weights of evaluation indicators at each level in the evaluation indicator system; S40. Calculate the indicator data for each level in the evaluation indicator system based on the operating parameters; S50. Based on fuzzy theory, establish the membership function of each evaluation index; S60. Calculate the membership degree of each level of evaluation index based on the preprocessed index data using the membership function; S70. Perform hierarchical fuzzy comprehensive evaluation on the obtained membership degrees, and visualize the membership probabilities of the converter valve and its components to the normal state, attention state, abnormal state and severe state. The step of establishing the membership functions of each evaluation index based on fuzzy theory specifically includes: Determine the linguistic variables and select linguistic variables to describe the state based on the status of the converter valve. Establish membership functions and determine the membership functions corresponding to different language values ​​for each evaluation index to describe the membership relationship between index values ​​and language values. Standardization transformation: Standardize each evaluation indicator and map it to the [0,1] interval; Construct a membership matrix and calculate the membership degree of each language value corresponding to the index data based on the membership function of each evaluation index to form a membership matrix; The step of performing hierarchical fuzzy comprehensive evaluation on the obtained membership degrees specifically includes: The status of the converter valve is divided into four levels: normal, warning, abnormal, and critical. Calculate the membership degree based on the hierarchical relationship of the indicator system; Based on the obtained membership degree, the corresponding maximum state level is found and used as the hierarchical fuzzy comprehensive evaluation result, where the membership probability is the membership degree corresponding to the hierarchical fuzzy comprehensive evaluation result. The membership function for the normal state is shown in formula (4): （4） Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the normal state; Note that the membership function of a state is shown in formula (5): （5） Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the attention state; The membership function for abnormal states is shown in formula (6): （6） Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the abnormal state; The membership function for the severe state is shown in formula (7): （7） Where x represents the evaluation result value of the membership function input, and y is the probability value of belonging to the severe state; Based on the evaluation index system and evaluation method, the evaluation result value is obtained. The evaluation result value is input into formulas (4), (5), (6), and (7) to obtain the probability that the equipment state belongs to different states.

2. The method for evaluating the condition of an ultra-high voltage converter valve according to claim 1, characterized in that, The method for preprocessing voltage and current in the operating parameters is a multiphysics model simulation verification.

3. The method for evaluating the condition of an ultra-high voltage converter valve according to claim 1, characterized in that, The step of acquiring and preprocessing the operating parameters of the converter valve specifically includes: Acquire raw signals from the converter valve collected by sensors or video terminals, including voltage, current, sound, and image signals; Preprocessing is performed on the original signal, including filtering and noise reduction. Effective features are extracted from the preprocessed signal and used as operating parameters for state assessment.

4. The method for evaluating the condition of an ultra-high voltage converter valve according to claim 1, characterized in that, The evaluation index system is divided into three levels according to the structural characteristics of the converter valve: the overall converter valve level, the internal component level, and the specific device level.

5. The method for evaluating the condition of an ultra-high voltage converter valve according to claim 1, characterized in that, The method for calculating the weights of evaluation indicators at each level in the evaluation index system is the hierarchical single-ranking method.

6. The method for evaluating the condition of an ultra-high voltage converter valve according to claim 1, characterized in that, The step of calculating the indicator data of each level in the evaluation indicator system based on the operating parameters also includes the step of performing fuzzy calculation on the indicator data of each level to obtain the quantitative value of the indicator.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, which, when executed, are used to perform the UHV converter valve status evaluation method according to any one of claims 1-6.

8. A condition evaluation system for ultra-high voltage converter valves, characterized in that, It includes the computer-readable storage medium of claim 7.

Images