Method and system for analyzing operation reliability of wind power equipment under time-varying conditions

Abstract

The application provides a wind power equipment operation reliability analysis method and system based on time-varying conditions, relates to the technical field of complex equipment safety management and maintenance decision analysis, obtains maximum likelihood estimators and distribution functions of various distribution types in a wind power equipment system based on a maximum likelihood estimation method and wind power equipment system fault data, and performs fitting processing on the distribution functions of the various distribution types according to an improved D-S evidence theory to obtain a fault data probability density function; the scoring process of various fault modes in the FMEA method is optimized according to a rough set theory, and the fault mode hazard degree is calculated; a dynamic reliability network model is constructed according to reliability data of the wind power equipment system, and the reliability index function and the reliability sensitivity of various risk nodes in the dynamic reliability network model are obtained. The application helps to improve the accuracy of fault prediction and the objectivity of risk assessment.

Description

Technical Field

[0001] This invention relates to the field of safety management and maintenance decision analysis technology for complex equipment, and in particular to a method and system for analyzing the operational reliability of wind power equipment under time-varying conditions. Background Technology

[0002] Industry is a crucial component of my country's equipment manufacturing industry, significantly impacting the national economy and national defense security. As a strategic industry, it drives the development of heavy industry, light industry, and electronics and information technology, while also promoting international trade. Power systems are key supporting equipment; 98% of ships use wind power equipment, and its reliability directly affects normal operation and mission capabilities. With technological advancements, power systems have evolved from steam engines to wind power, gas turbines, nuclear power, and all-electric propulsion. Wind power equipment, due to its economic efficiency and ease of control, is widely used in civilian applications such as merchant ships and fishing vessels.

[0003] Chinese patent CN111310789B discloses a wind power interval prediction method and its implementation system based on kernel density estimation. First, it uses a support vector machine modeling method based on continuous time-time clustering to perform deterministic point prediction of wind power, obtaining a comparison chart of predicted and actual wind power values. Then, it uses kernel density estimation and optimal window width selection to establish a probability density function for the prediction error within each power partition, and compares and analyzes the probability density curves and error frequency histograms obtained by recursive and sliding window methods. Finally, it calculates wind power prediction intervals that meet a certain confidence probability based on the deterministic point prediction results, and compares and analyzes the interval coverage and average interval width as evaluation indicators with recursive and sliding window methods. However, the method provided in the above application only studies the system as a whole, without comparing and selecting the best from multiple distribution types, which easily leads to relatively low fitting accuracy. Therefore, it is necessary to provide a method and system for wind power equipment operation reliability analysis under time-varying conditions to improve the accuracy of fault prediction and the objectivity of risk assessment. Summary of the Invention

[0004] In view of this, this invention proposes a method and system for analyzing the operational reliability of wind power equipment under time-varying conditions. By combining maximum likelihood estimation and improved DS evidence theory, the accuracy of identifying fault data distribution characteristics is improved. At the same time, by combining rough set theory to optimize the FMEA scoring process, objective fault mode hazard assessment is achieved, thereby improving the accuracy of fault prediction and the objectivity of risk assessment.

[0005] This invention provides a method for analyzing the operational reliability of wind power equipment under time-varying conditions, the method comprising:

[0006] Based on the maximum likelihood estimation method and wind power equipment system fault data, the maximum likelihood estimators and distribution functions of each distribution type in the wind power equipment system are obtained. The distribution functions of each distribution type are fitted according to the improved DS evidence theory to obtain the probability density function of the fault data.

[0007] The scoring process for each failure mode in the FMEA method is optimized based on rough set theory, and the hazard of failure modes is calculated. The hazard of failure modes includes the roughness number of the scoring corresponding to the failure mode and the ranking of the risk level of the failure mode.

[0008] Based on the reliability data of the wind power equipment system, a dynamic reliability network model is constructed, and the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model are obtained. The dynamic reliability network model includes multiple risk nodes.

[0009] Based on the above technical solutions, preferably, the step of obtaining the maximum likelihood estimator and distribution function of each distribution type in the wind power equipment system based on the maximum likelihood estimation method and wind power equipment system fault data specifically includes:

[0010] The number of groups is determined based on parameter estimation and empirical formulas. The unknown parameters in the probability density function are calculated based on the maximum likelihood estimation method and the fault data of the wind power equipment system.

[0011] Based on the maximum likelihood estimation method and the distribution types in the fault data of the wind power equipment system, the maximum likelihood estimate of each distribution type is calculated.

[0012] Based on the above technical solutions, preferably, the step of fitting the distribution function of each distribution type according to the improved DS evidence theory to obtain the probability density function of fault data specifically includes:

[0013] Calculate the slope between two adjacent points of the probability density function corresponding to the fault data, the median of two adjacent points of the probability density function corresponding to the fault data, and the difference of the median of the probability density function corresponding to the fault data, respectively.

[0014] The slope and the median difference are processed by maximum value image processing respectively, and the gray correlation degree between the probability density functions corresponding to the fault data is calculated to obtain the probability density function of the fault data.

[0015] More preferably, the optimization of the failure mode scoring process in the FMEA method based on rough set theory, and the calculation of the failure mode severity, specifically includes:

[0016] Obtain the scoring matrix for each failure mode and the weight vector corresponding to each risk factor. Based on the scoring transformation function, convert the risk factor scores corresponding to all scoring matrices into rough scores.

[0017] Under different risk factors, the distance from the rough number of each failure mode score to the ideal solution is calculated to obtain the group utility maximization ranking, individual regret minimization ranking, and compromise ranking of each failure mode.

[0018] The failure modes are sorted according to a compromise ranking to obtain a failure mode risk level sequence.

[0019] More preferably, the step of constructing a dynamic reliability network model based on the reliability data of the wind power equipment system specifically includes:

[0020] Reliability data of wind power equipment systems are collected, and various types of data in the reliability data are used as risk nodes. The reliability data includes data on system components, failure modes, failure causes, and reliability drivers.

[0021] Based on the rough score number corresponding to each risk node, calculate the ideal solution and negative ideal solution of each risk node under different risk factors, so as to obtain the distance and compromise ranking from the rough score number corresponding to each risk node to the ideal solution.

[0022] Using the compromise sorting as the node hazard degree, the reliability hazard degree corresponding to the wind power equipment system components, failure modes and failure causes is obtained respectively, and each reliability hazard degree is used as the node weight of the corresponding risk node;

[0023] Based on the failure data in the failure modes, the time function of the failure rate under different propagation relationships is calculated, and the time function is used as the edge weight of the corresponding risk node to construct a dynamic reliability network model.

[0024] More preferably, the method further includes evaluating the reliability level of the wind power equipment system under the current state based on the first system reliability level influence function and the second system reliability level influence function, wherein the expressions for the first system reliability level influence function and the second system reliability level influence function are:

[0025] C max =I max (B f E)W f→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0026] C t =I t (B f E)Wf→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0027] Among them, C max I represents the function that influences the reliability level of the first system. max B represents the matrix of maximum values ​​affecting reliability indicators. f The weight matrix represents the nodes causing the failure, E represents the expert rating matrix, and W represents the weight matrix of the nodes causing the failure. f→f W represents the influence weight matrix among failure causes. f→F B represents the weighting matrix of the influence between failure causes and typical system failure modes. F W represents the node weight matrix for typical failure modes of the system. F→F W represents the influence weight matrix among typical failure modes of the system. F→P B represents the influence weight matrix between typical failure modes of the system and system components. P W represents the system component node weight matrix. P→P W represents the influence weight matrix between system components. P→C C represents the influence weight matrix between system components and the central node. t I represents the function that influences the reliability level of the second system. t This represents the impact matrix of the monitoring indicators.

[0028] More preferably, the expression for the reliability index function is:

[0029]

[0030] Where C represents the influence of the central node, k represents the number of groups, P represents the influence of system component nodes, i represents the i-th risk node, j represents the j-th risk node, F represents the influence of typical system failure mode nodes, f represents the influence of failure cause nodes, and B f B represents the weight matrix of the node causing the failure. F B represents the node weight matrix for typical failure modes of the system. P This represents the system component node weight matrix. Represents system component node P j To system component node P i Influence weight, Represents system component node P i The influence weight of the central node Indicates fault mode F i To Fault Mode Fj Influence weight, Indicates fault mode F i To system component P k Influence weight, Node f represents the cause of failure. j With the failure cause node f i Influence weight matrix Node f represents the cause of failure. j With the system's typical failure mode node F k The influence weight matrix, where dist() represents the distance function. Represents the impact matrix of monitoring indicators, Interval Ii Represents the interval number.

[0031] A second aspect of this application provides a wind power equipment operation reliability analysis system based on time-varying conditions. The wind power equipment operation reliability analysis system includes a function fitting module, a risk ranking module, and a reliability analysis module, wherein...

[0032] The function fitting module is used to obtain the maximum likelihood estimate and distribution function of each distribution type in the wind power equipment system based on the maximum likelihood estimation method and the fault data of the wind power equipment system, and to perform fitting processing on the distribution function of each distribution type according to the improved DS evidence theory to obtain the probability density function of the fault data.

[0033] The risk ranking module is used to optimize the scoring process of each failure mode in the FMEA method according to rough set theory and calculate the hazard of failure modes. The hazard of failure modes includes the roughness number of the scoring corresponding to the failure mode and the risk level ranking of the failure mode.

[0034] The reliability analysis module is used to construct a dynamic reliability network model based on the reliability data of the wind power equipment system, and to obtain the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model, wherein the dynamic reliability network model includes multiple risk nodes.

[0035] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory.

[0036] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of a method for analyzing the operational reliability of wind power equipment under time-varying conditions.

[0037] The wind power equipment operation reliability analysis method and system based on time-varying conditions provided by this invention have the following advantages over existing technologies:

[0038] (1) By combining maximum likelihood estimation and improved DS evidence theory, the identification accuracy of fault data distribution characteristics is improved. The obtained probability density function can more accurately reflect the fault characteristics of wind power equipment system. At the same time, the FMEA scoring process is optimized by combining rough set theory to achieve objective fault mode hazard assessment. The reliability index and sensitivity of each risk node are analyzed by using dynamic reliability network model. Thus, a complete technical chain from data processing, risk assessment to reliability analysis is constructed, which not only improves the accuracy of fault prediction and the objectivity of risk assessment, but also enhances the reliability analysis capability.

[0039] (2) By constructing a scoring matrix and introducing a weight vector, the risk factor scores are transformed into rough scores. At the same time, the ideal solution distance calculation is combined to achieve the dual optimization of maximizing group utility and minimizing individual regret. Finally, an objective failure mode risk level sequence is obtained, which effectively improves the scientificity and reliability of the FMEA scoring process, overcomes the defects of strong subjectivity in traditional scoring methods, and provides a more accurate and reliable quantitative basis for the risk assessment of wind power equipment failure modes.

[0040] (3) By collecting comprehensive reliability data including system components, failure modes, failure causes and reliability drivers, and combining the score rough number to calculate the ideal solution and negative ideal solution to obtain an objective compromise ranking, the reliability hazard degree is converted into node weights. At the same time, the failure rate time function is considered as the edge weight, and finally a dynamic reliability network model is constructed, realizing the improvement from static assessment to dynamic analysis. This not only improves the accuracy and comprehensiveness of risk assessment, but also reflects the characteristics of system reliability changes over time, providing a more timely and targeted scientific basis for equipment maintenance decisions. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating the wind power equipment operation reliability analysis method based on time-varying conditions provided by this invention;

[0043] Figure 2 A schematic diagram of the reliability modeling process provided by the present invention;

[0044] Figure 3 Example diagram of the structure of the reliability network model provided by this invention;

[0045] Figure 4 This is a schematic diagram of the wind power equipment operation reliability analysis system provided by the present invention;

[0046] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention.

[0047] Explanation of reference numerals in the attached figures: 1. Wind power equipment operation reliability analysis system; 11. Function fitting module; 12. Risk ranking module; 13. Reliability analysis module; 2. Electronic equipment; 21. Processor; 22. Communication bus; 23. User interface; 24. Network interface; 25. Memory. Detailed Implementation

[0048] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0049] This invention discloses a method for analyzing the operational reliability of wind power equipment under time-varying conditions, with reference to... Figure 1 The method includes steps S1 to S3.

[0050] Step S1: Based on the maximum likelihood estimation method and the fault data of the wind power equipment system, obtain the maximum likelihood estimator and distribution function of each distribution type in the wind power equipment system, and perform fitting processing on the distribution function of each distribution type according to the improved DS evidence theory to obtain the probability density function of the fault data.

[0051] This step also includes steps S11 to S14.

[0052] Step S11: Determine the number of groups based on parameter estimation and empirical formulas, and calculate the unknown parameters in the probability density function based on the maximum likelihood estimation method and wind power equipment system fault data.

[0053] In this step, based on parameter estimation, the number of groups is determined using an empirical formula to optimize the probability density function fitting of the fault data and reduce fitting error. The calculation formula is as follows:

[0054]

[0055] Where, n F Indicates the number of samples.

[0056] Furthermore, based on the maximum likelihood estimation method, the unknown parameters in the probability density function are calculated using fault data from wind power equipment systems.

[0057] If (x1, x2, ..., x n Let be a sample from the population, and its sampling probability be... Construct the likelihood function:

[0058]

[0059] Where (x1,x2,…,x) n ) represents a set of samples, Let θ represent a set of sampling probabilities, θ = [θ1, θ2, ..., θ2]. n [] represents the overall parameter to be estimated.

[0060] Step S12: Based on the maximum likelihood estimation method and the distribution types in the wind power equipment system fault data, calculate the maximum likelihood estimate for each distribution type.

[0061] The maximum likelihood estimators for each distribution type are calculated using the formulas shown in Table 1.

[0062] Table 1 Maximum Likelihood Estimators for Each Distribution Type

[0063]

[0064] Furthermore, when using a rectangular kernel function for estimation under small sample conditions, a smooth Gaussian kernel function is selected for estimation to obtain a smooth probability density function curve. The calculation formula is as follows:

[0065]

[0066] Where N represents the number of samples, h represents the bandwidth, and x n This represents the nth sample.

[0067] Based on kernel density estimation, an appropriate bandwidth is selected to ensure the smoothness and accuracy of the probability density function of fault data, thereby optimizing the estimation results.

[0068] Since bandwidth is a free parameter, its suitability directly affects the accuracy of the probability density function estimation results for fault data. Therefore, the choice of bandwidth h should minimize the error of the probability density function estimation results. Generally, the mean integral squared error (MISE) is used to measure the error of the estimation results, and the calculation formula is as follows:

[0069]

[0070] To minimize MISE(h), we need to find the minimum point of MISE(h). The bandwidth h that minimizes MISE(h) can be obtained by applying the above equation. AMISE The calculation formula is:

[0071]

[0072] R(g)=∫g(x) 2 dx; m2(K)=∫x 2 K(x)dx

[0073] Once the kernel function is determined, R, m, and f″ in the above formula can all be determined, and the bandwidth calculation formula is:

[0074]

[0075] in, This represents the sample standard deviation, and N represents the sample size.

[0076] Based on the principle of kernel density estimation, the deviation of the kernel density estimation principle is calculated using the following formula:

[0077]

[0078]

[0079] in, Assuming the kernel function has [-1, 1] as its boundary, when |u| > 1, k(u) = 0. Therefore, when x > h, a0(x) = 1, a1(x) = 0. However, when 0 < x < h, a0(x) ≠ 1, a1(x) ≠ 0, and the bias always exists. Using xk(x) as the kernel function to estimate f(x), we construct a new kernel function such that when 0 < x < h, the bias of the kernel density estimation is 0.

[0080] Step S13: Calculate the slope between two adjacent points of the probability density function corresponding to the fault data, the median of two adjacent points of the probability density function corresponding to the fault data, and the difference in median of the probability density function corresponding to the fault data.

[0081] In this step, the linear correlation between the system fault data and its probability density function is determined based on the linear correlation coefficient. The calculation formula is as follows:

[0082]

[0083] Where, x i This represents the value of the i-th fault data. x represents i The mean of y i Indicates that it is located at x i The function value at that point, Indicates y i The mean.

[0084] Calculate the slope between two adjacent points of the fault data and its probability density function:

[0085]

[0086] Where n represents the number of samples, f(x) j ) represents sample x j The statistical function value at f i (x j ) represents the probability density function in x j The function value at point K i K represents the slope of the fault data between the j-th sample point and the (j+1)-th sample point. ij This represents the slope of the probability density function between the j-th sample point and the (j+1)-th sample point.

[0087] Calculate the median of the fault data and its probability density function between two adjacent points:

[0088]

[0089] Calculate the difference between the fault data and the slope and median of its probability density function:

[0090]

[0091] Maximize the difference between slope and median:

[0092]

[0093] Step S14: Perform maximum value image processing on the slope and median difference respectively, and calculate the gray correlation degree between the probability density functions corresponding to the fault data to obtain the probability density function of the fault data.

[0094] The grey relational degree is calculated based on the slope of the fault data and its probability density function between two adjacent points, thereby determining the goodness of fit of the curve. The calculation formula is as follows:

[0095]

[0096] After standardizing the grey relational degree, we get:

[0097]

[0098] Based on the improved DS evidence theory method, the goodness-of-fit indices of the same distribution function are fused to optimize the five target distribution functions. The calculation formula is as follows:

[0099]

[0100] Where, m CCM (θ i ), m GRM (θ i ), m MDM (θ i ), m CSM (θ i ) represent the evidence for the i-th distribution function in the correlation coefficient method, grey relational method, minimum distance method, and correlation entropy method, respectively, where m is the number of distribution functions.

[0101] Step S2: Optimize the scoring process of each failure mode in the FMEA method according to rough set theory, and calculate the hazard of failure modes. The hazard of failure modes includes the roughness number of the scoring corresponding to the failure mode and the ranking of the risk level of the failure mode.

[0102] The FMEA dynamic scoring method improvement focuses on failure severity (S), occurrence rate (O), and detectability (D). Among these, S and D are still evaluated using traditional methods because the scoring criteria are stable. Since the occurrence rate varies the most during system operation, the failure rate of each failure mode is estimated using a distribution function to obtain its probability density function, calculated as follows:

[0103]

[0104] This step also includes steps S21 to S23.

[0105] Step S21: Obtain the scoring matrix of each failure mode and the weight vector corresponding to each risk factor, and convert the risk factor scores corresponding to all scoring matrices into rough scores based on the scoring transformation function.

[0106] Rough set theory is used to improve the scoring process for each failure mode. Lower and upper approximations are used to describe the imprecision and fuzziness of the information. Let the k-th expert's score for each failure mode be calculated using the following formula:

[0107]

[0108] Among them, EM k Let m be the rating matrix for the k-th expert, and m be the number of failure modes. For the k-th expert's score on the j-th risk factor of the i-th failure mode, EV k Let be the weight matrix of the k-th expert. Let be the weight of the k-th expert on the j-th risk factor.

[0109] After obtaining the score matrix for each failure mode and the weight vector for each risk factor, the scores are converted into rough numbers, calculated using the following formula:

[0110]

[0111] in, This represents the lower bound of the rough number of the risk factor score. This represents the upper bound of the rough number of the risk factor score. This represents the lower bound of the rough number of risk factor weights. This represents the upper bound of the rough number of risk factor weights.

[0112] The rough number of the scores given by each expert for the j-th risk factor of each failure mode is calculated using the following formula:

[0113]

[0114] According to the rough number calculation rules, the rough numbers for each failure mode and the weighted rough numbers for each risk factor are fused together, and the calculation formula is as follows:

[0115]

[0116] in, This represents the coarse score after fusing all fault modes. This represents the rough weighted number after integrating all risk factors, and K represents the number of experts.

[0117] Step S22: Calculate the distance from the rough number of each failure mode score to the ideal solution under different risk factors, so as to obtain the group utility maximization ranking, individual regret minimization ranking, and compromise ranking of each failure mode.

[0118] Furthermore, based on the distance from the rough score to the ideal solution for each failure mode under different risk factors, the group utility maximization ranking S, the individual regret minimization ranking R, and the compromise ranking Q for each failure mode are calculated. The calculation formulas are as follows:

[0119]

[0120] Among them, S - =minS i S + =maxS i R - =minR i R + =maxR i v = 0.5, where v is the maximum group utility weight.

[0121] Step S23: Sort the failure modes according to the compromise sorting to obtain the failure mode risk level sequence.

[0122] Failure modes are ranked according to the compromise ranking Q. The smaller the Q value, the higher the risk level of the failure mode. The calculation formula is as follows:

[0123]

[0124] Where Q(fm) (2) ) is the Q value corresponding to the second-ranked fault mode, and m is the number of fault modes.

[0125] By constructing a scoring matrix and introducing a weight vector, the risk factor scores are transformed into rough scores. At the same time, the ideal solution distance calculation is combined to achieve dual optimization of maximizing group utility and minimizing individual regret, ultimately obtaining an objective failure mode risk level sequence. This effectively improves the scientificity and reliability of the FMEA scoring process, overcomes the shortcomings of strong subjectivity in traditional scoring methods, and provides a more accurate and reliable quantitative basis for risk assessment of wind power equipment failure modes.

[0126] Step S3: Based on the reliability data of the wind power equipment system, construct a dynamic reliability network model, and obtain the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model. The dynamic reliability network model includes multiple risk nodes. The dynamic reliability network model has 5 layers, from the outside to the inside: reliability drivers, failure causes, system components, failure modes, and central nodes.

[0127] By collecting comprehensive reliability data, including system components, failure modes, failure causes, and reliability drivers, and combining the roughness score to calculate the ideal and negative ideal solutions to obtain an objective compromise ranking, the reliability hazard is transformed into node weights. At the same time, the failure rate time function is considered as the edge weight, and finally a dynamic reliability network model is constructed, realizing the improvement from static assessment to dynamic analysis. This not only improves the accuracy and comprehensiveness of risk assessment, but also reflects the changes in system reliability over time, providing a more timely and targeted scientific basis for equipment maintenance decisions.

[0128] This step also includes steps S31 to S34.

[0129] Step S31: Collect reliability data of the wind power equipment system and use various types of data in the reliability data as risk nodes. The reliability data includes system components, failure modes, failure causes and reliability drivers data.

[0130] In this step, data on typical components, failure modes, failure causes, reliability drivers, and corresponding fault data of wind turbine equipment are collected; and the N-type data of wind turbine equipment is organized. p Each system component, N F A typical failure mode and N f Failure cause data, collect and organize failure data F that led to the occurrence of failure modes due to failure causes. fi→FjFault data F that caused the failure of system components due to the occurrence of a failure mode. Fi→Pj Fault data F caused by system component failure leading to system power shutdown Pi→C Among them, f i Let F represent the i-th cause of failure. j Let P represent the j-th fault mode. j This represents the j-th system component.

[0131] Step S32: Based on the rough score number corresponding to each risk node, calculate the ideal solution and negative ideal solution of each risk node under different risk factors, so as to obtain the distance and compromise ranking of the rough score number to the ideal solution of each risk node.

[0132] Step S33: Using the compromise ranking as the node hazard degree, obtain the reliability hazard degree corresponding to the wind power equipment system components, failure modes and failure causes respectively, and use each reliability hazard degree as the node weight of the corresponding risk node.

[0133] Furthermore, based on reliability data, an improved FMEA method is used to perform failure mode and impact analysis. First, the scoring weights of different risk factors for each node are obtained, and data and modeling errors are introduced to form rough scores for different risk factors for each node. Then, based on the rough scores of each node, the ideal and negative ideal solutions for each node under different risk factors are calculated, and the distance from the rough score to the ideal solution is calculated. Based on this, calculations are performed on nodes at each level to obtain the group utility maximization ranking S, the individual regret minimization ranking R, and the compromise ranking Q. The compromise ranking Q value is used as the node hazard degree to obtain the hazard degree of wind power system components, failure modes, and failure causes to the reliability of wind power equipment, and this is used as the node weight.

[0134] Step S34: Based on the failure data in the failure modes, calculate the time function of the failure rate under different propagation relationships, and use the time function as the edge weight of the corresponding risk node to construct a dynamic reliability network model.

[0135] In this step, historical fault data is used to fit various distribution functions through parametric and nonparametric estimation methods. Different methods are then used to perform goodness analysis on the fitted distribution functions. Subsequently, the DS evidence theory method is used to optimize the distribution function to obtain the best distribution type of the fault data. The distribution function is determined under time-varying uncertainty conditions, and the error is calculated. Finally, the time function of the edge weights of inter-layer nodes is determined. The influence relationship of intra-layer nodes is scored using the experience and knowledge of domain experts to obtain the edge weights of intra-layer nodes.

[0136] like Figure 2 and Figure 3As shown, the reliability network model can be defined as a triple, which can be represented as:

[0137] RM = <V,R,W>

[0138] Where: V = {V C V P V F V f V I} represents the set of all nodes in the reliability network model. V C V is the central node of the model, and V represents the comprehensive reliability index of the wind power equipment. P V F V f V I These represent the system components, typical failure modes, failure causes, and reliability drivers in the reliability model, respectively.

[0139] R = {R I→f ,R f→f ,R f→F ,R F→F ,R F→P ,R P→P ,R P→C R represents the set of relationship matrices between nodes within and between layers in a reliability network model. I→f R is the relationship matrix between reliability driver nodes and failure cause nodes. f→F R is the relationship matrix between failure cause nodes and typical system failure mode nodes. F→P R is the relationship matrix between typical failure mode nodes and system component nodes. P→C R is the relationship matrix between system component nodes and the central node; f→f R is the relationship matrix between the nodes causing the failure. F→F R is the relationship matrix between nodes in typical system failure modes. P→P This is the relationship matrix between system component nodes.

[0140] B is the set of all weight matrices in the reliability network model. f B is the weight matrix of the failure cause node. F B is the node weight matrix for typical failure modes of the system. P This is the system component node weight matrix; This is the influence weight matrix between failure causes and typical system failure modes. This is a matrix showing the influence weights between typical system failure modes and system components. This is the influence weight matrix between system components and the central node; The weight matrix represents the influence between failure causes. This is the influence weight matrix among typical failure modes of the system. This forms the influence weight matrix between system components. Ultimately, it results in a dynamic reliability model for wind power equipment where nodes have sizes and edges have weights.

[0141] Based on a dynamic reliability network model, this study analyzes the coupling relationships and fault correlations of components in wind power equipment systems, providing reliability assessment and maintenance guidance, specifically including:

[0142] Collect all monitoring indicators affecting system reliability under all failure causes and their normal threshold ranges, i.e., reliability indicators, and use the S-type membership function to calculate their degree of influence on the failure causes. The calculation formula is as follows:

[0143]

[0144] Where q1 represents the lower limit of the monitoring indicator threshold and q2 represents the upper limit of the monitoring indicator threshold.

[0145] To obtain the reliability impact of the system under the worst-case scenario, the influence of all reliability indicators on the cause of failure is set to the maximum value of 1, thus obtaining the maximum value matrix I of the reliability indicator influence. max And calculate the maximum value C that the system reliability level is affected under this condition. max The weights of each node and edge are functions of time, but the time axes of each function are not the same. The time variable in each function is the time since the last failure of each node and edge.

[0146] Based on the first system reliability level influence function and the second system reliability level influence function, the reliability level of the wind power equipment system under the current state is evaluated. The expressions for the first system reliability level influence function and the second system reliability level influence function are as follows:

[0147] C max =I max (B f E)W f→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0148] The impact of system reliability on actual values: In reality, most monitoring indicators are in normal condition, with only a few deviating from their normal threshold ranges. Based on the real-time monitoring values ​​and normal threshold ranges of each indicator, the impact of each indicator on the causes of failure is assessed, forming the monitoring indicator impact matrix I. tAnd calculate the degree to which the system reliability level, C, is affected under the current state. t The weights of each node and edge are the same as above, and the time variable in each function is the time since the last failure of each node and edge.

[0149] C t =I t (B f E)W f→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0150] Among them, C max I represents the function that influences the reliability level of the first system. max B represents the matrix of maximum values ​​affecting reliability indicators. f The weight matrix represents the nodes causing the failure, E represents the expert rating matrix, and W represents the weight matrix of the nodes causing the failure. f→f W represents the influence weight matrix among failure causes. f→F B represents the weighting matrix of the influence between failure causes and typical system failure modes. F W represents the node weight matrix for typical failure modes of the system. F→F W represents the influence weight matrix among typical failure modes of the system. F→P B represents the influence weight matrix between typical failure modes of the system and system components. P W represents the system component node weight matrix. P→P W represents the influence weight matrix between system components. P→C C represents the influence weight matrix between system components and the central node. t I represents the function that influences the reliability level of the second system. t This represents the impact matrix of the monitoring indicators.

[0151] In practice, most monitoring indicators are within normal limits, with only a few deviating from their normal threshold ranges. Based on the real-time monitoring values ​​and normal threshold ranges of each indicator, the impact of each indicator on the cause of failure is assessed, forming a monitoring indicator impact matrix I. t And calculate the degree to which the system reliability level, C, is affected under the current state. t The weights of each node and edge are the same as above, and the time variable in each function is the time since the last failure of each node and edge.

[0152] Based on sensitivity analysis, this study investigates the importance of wind power equipment system nodes to reliability fluctuations, traces the influencing factors, evaluates and optimizes system performance, and thus achieves a comprehensive analysis of wind power equipment operation reliability under time-varying conditions.

[0153] The expression for the reliability index function is:

[0154]

[0155] Where C represents the influence of the central node, k represents the number of groups, P represents the influence of system component nodes, i represents the i-th risk node, j represents the j-th risk node, F represents the influence of typical system failure mode nodes, f represents the influence of failure cause nodes, and B f B represents the weight matrix of the node causing the failure. F B represents the node weight matrix for typical failure modes of the system. P This represents the system component node weight matrix. Represents system component node P j To system component node P i Influence weight, Represents system component node P i The influence weight of the central node Indicates fault mode F i To Fault Mode F j Influence weight, Indicates fault mode F i To system component P k Influence weight, Node f represents the cause of failure. j With the failure cause node f i Influence weight matrix Node f represents the cause of failure. j With the system's typical failure mode node F k The influence weight matrix, where dist() represents the distance function. This represents the impact matrix of the monitoring indicators. Represents the interval number.

[0156] The reliability sensitivity of each node is calculated using the following formula:

[0157]

[0158] The node reliability sensitivity error analysis is performed, and the calculation formula is as follows:

[0159]

[0160] Where C represents the influence of the central node, k represents the number of groups, P represents the influence of system component nodes, F represents the influence of typical system failure mode nodes, f represents the influence of failure cause nodes, and σ x This represents the error of each weight.

[0161] Dynamic identification of key risk points in the system is performed. Since the sensitivity range of each node in the system is an interval, it is necessary to sort the sensitivity intervals of each node to analyze the relative importance of each node. The relative dominance ranking method based on the number of intervals is used to sort the sensitivity of each node. The specific calculation process of this method is as follows:

[0162] 1) Compare the numbers in each interval pairwise to construct a relative dominance matrix;

[0163]

[0164] Among them, A - Let A denote the left limit of the interval number A. + Let B represent the right limit of the interval number A. - Let B represent the left limit of the interval number B. + Let B represent the right limit of the interval number B.

[0165] 2) Calculate the sorting vector for each interval;

[0166]

[0167] Where P represents the relative dominance matrix of interval numbers, and n represents the number of interval numbers.

[0168] 3) Sort the interval numbers using the sorting vector of each interval number.

[0169] By combining maximum likelihood estimation and improved DS evidence theory, the accuracy of fault data distribution characteristics identification is enhanced. The obtained probability density function more accurately reflects the fault characteristics of wind power equipment systems. Simultaneously, rough set theory is used to optimize the FMEA scoring process, achieving objective fault mode hazard assessment. Furthermore, a dynamic reliability network model is employed to analyze the reliability indicators and sensitivity of each risk node. This constructs a complete technical chain from data processing and risk assessment to reliability analysis, improving not only the accuracy of fault prediction and the objectivity of risk assessment but also enhancing reliability analysis capabilities.

[0170] Based on the above method, this application discloses a wind power equipment operation reliability analysis system under time-varying conditions, with reference to... Figure 4 The wind power equipment operation reliability analysis system 1 includes a function fitting module 11, a risk ranking module 12, and a reliability analysis module 13, wherein...

[0171] The function fitting module 11 is used to obtain the maximum likelihood estimator and distribution function of each distribution type in the wind power equipment system based on the maximum likelihood estimation method and the fault data of the wind power equipment system, and to perform fitting processing on the distribution function of each distribution type according to the improved DS evidence theory to obtain the probability density function of the fault data.

[0172] The risk ranking module 12 is used to optimize the scoring process of each failure mode in the FMEA method according to rough set theory and calculate the hazard of failure mode. The hazard of failure mode includes the rough number of the scoring corresponding to the failure mode and the ranking of the risk level of failure mode.

[0173] The reliability analysis module 13 is used to construct a dynamic reliability network model based on the reliability data of the wind power equipment system, and to obtain the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model. The dynamic reliability network model includes multiple risk nodes.

[0174] In one example, the function fitting module 11 is used to determine the number of groups based on parameter estimation and empirical formulas, calculate the unknown parameters in the probability density function based on the maximum likelihood estimation method and the fault data of the wind power equipment system, and calculate the maximum likelihood estimate of each distribution type based on the maximum likelihood estimation method and the fault data of the wind power equipment system.

[0175] In one example, the function fitting module 11 is used to calculate the slope between two adjacent points of the probability density function corresponding to the fault data, the median of two adjacent points of the probability density function corresponding to the fault data, and the median difference of the probability density function corresponding to the fault data; the maximum value image processing is performed on the slope and the median difference respectively, and the gray correlation degree between the probability density functions corresponding to the fault data is calculated to obtain the probability density function of the fault data.

[0176] In one example, the risk ranking module 12 is used to obtain the rating matrix of each failure mode and the weight vector corresponding to each risk factor. Based on the rating transformation function, the risk factor ratings corresponding to all rating matrices are converted into rough rating numbers. The distance from the rough rating number of each failure mode to the ideal solution is calculated under different risk factors to obtain the group utility maximization ranking, individual regret minimization ranking, and compromise ranking of each failure mode. The failure modes are ranked according to the compromise ranking to obtain the failure mode risk level sequence.

[0177] In one example, the reliability analysis module 13 is used to collect reliability data of the wind power equipment system and use various types of data in the reliability data as risk nodes. The reliability data includes system components, failure modes, failure causes, and reliability drivers. Based on the rough score corresponding to each risk node, the ideal solution and negative ideal solution of each risk node under different risk factors are calculated to obtain the distance and compromise ranking from the rough score to the ideal solution of each risk node. The compromise ranking is used as the node hazard, and the reliability hazard corresponding to the wind power equipment system components, failure modes, and failure causes are obtained respectively. Each reliability hazard is used as the node weight of the corresponding risk node. Based on the failure data in the failure modes, the time function of the failure rate under different propagation relationships is calculated, and the time function is used as the edge weight of the corresponding risk node to construct a dynamic reliability network model.

[0178] In one example, the reliability level of the wind power system under the current state is evaluated based on the first system reliability level influence function and the second system reliability level influence function, wherein the expressions for the first system reliability level influence function and the second system reliability level influence function are:

[0179] C max =I max (B f E)W f→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0180] C t =I t (B f E)W f→f W f→F (B F E)W F→F W F→P (B P E)W P→P W P→C

[0181] Among them, C max I represents the function that influences the reliability level of the first system. max B represents the matrix of maximum values ​​affecting reliability indicators. f The weight matrix represents the nodes causing the failure, E represents the expert rating matrix, and W represents the weight matrix of the nodes causing the failure. f→f W represents the influence weight matrix among failure causes. f→F B represents the weighting matrix of the influence between failure causes and typical system failure modes. FW represents the node weight matrix for typical failure modes of the system. F→F W represents the influence weight matrix among typical failure modes of the system. F→P B represents the influence weight matrix between typical failure modes of the system and system components. P W represents the system component node weight matrix. P→P W represents the influence weight matrix between system components. P→C C represents the influence weight matrix between system components and the central node. t I represents the function that influences the reliability level of the second system. t This represents the impact matrix of the monitoring indicators.

[0182] In one example, the expression for the reliability metric function is:

[0183]

[0184] Where C represents the influence of the central node, k represents the number of groups, P represents the influence of system component nodes, i represents the i-th risk node, j represents the j-th risk node, F represents the influence of typical system failure mode nodes, f represents the influence of failure cause nodes, and B f B represents the weight matrix of the node causing the failure. F B represents the node weight matrix for typical failure modes of the system. P This represents the system component node weight matrix. Represents system component node P j To system component node P i Influence weight, Represents system component node P i The influence weight of the central node Indicates fault mode F i To Fault Mode F j Influence weight, Indicates fault mode F i To system component P k Influence weight, Node f represents the cause of failure. j With the failure cause node f i Influence weight matrix Node f represents the cause of failure. j With the system's typical failure mode node F k The influence weight matrix, where dist() represents the distance function. This represents the impact matrix of the monitoring indicators. Represents the interval number.

[0185] Please see Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 2 may include: at least one processor 21, at least one network interface 24, user interface 23, memory 25, and at least one communication bus 22.

[0186] The communication bus 22 is used to enable communication between these components.

[0187] The user interface 23 may include a display screen and a camera. Optionally, the user interface 23 may also include a standard wired interface and a wireless interface.

[0188] The network interface 24 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0189] The processor 21 may include one or more processing cores. The processor 21 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 25, and by calling data stored in the memory 25. Optionally, the processor 21 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 21.

[0190] The memory 25 may include random access memory (RAM) or read-only memory. Optionally, the memory 25 may include non-transitory computer-readable storage medium. The memory 25 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 25 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 25 may also be at least one storage device located remotely from the aforementioned processor 21. Figure 5 As shown, the memory 25, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a method for analyzing the operational reliability of wind power equipment under time-varying conditions.

[0191] exist Figure 5 In the electronic device 2 shown, the user interface 23 is mainly used to provide an input interface for the user and obtain the user input data; while the processor 21 can be used to call the wind power equipment operation reliability analysis method based on time-varying conditions stored in the memory 25. When executed by one or more processors, the electronic device executes one or more methods as described in the above embodiments.

[0192] A computer-readable storage medium storing instructions that, when executed by one or more processors, cause a computer to perform one or more methods as described in the embodiments above.

[0193] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0194] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0195] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0196] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0197] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0198] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0199] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure.

[0200] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for analyzing the operational reliability of wind power equipment under time-varying conditions, characterized in that, The method includes: Based on the maximum likelihood estimation method and wind power equipment system fault data, the maximum likelihood estimators and distribution functions of each distribution type in the wind power equipment system are obtained. The distribution functions of each distribution type are fitted according to the improved DS evidence theory to obtain the probability density function of the fault data. The process of fitting the distribution functions of each distribution type according to the improved DS evidence theory to obtain the probability density function of fault data specifically includes: Calculate the slope between two adjacent points of the probability density function corresponding to the fault data, the median of two adjacent points of the probability density function corresponding to the fault data, and the difference of the median of the probability density function corresponding to the fault data, respectively. The slope and the median difference are processed by maximum value image processing respectively, and the gray correlation degree between the probability density functions corresponding to the fault data is calculated to obtain the probability density function of the fault data. The scoring process for each failure mode in the FMEA method is optimized based on rough set theory, and the hazard of failure modes is calculated. The hazard of failure modes includes the roughness number of the scoring corresponding to the failure mode and the ranking of the risk level of the failure mode. The optimization of the failure mode scoring process in the FMEA method based on rough set theory, and the calculation of the failure mode severity, specifically include: Obtain the scoring matrix for each failure mode and the weight vector corresponding to each risk factor. Based on the scoring transformation function, convert the risk factor scores corresponding to all scoring matrices into rough scores. Under different risk factors, the distance from the rough number of each failure mode score to the ideal solution is calculated to obtain the group utility maximization ranking, individual regret minimization ranking, and compromise ranking of each failure mode. The failure modes are sorted according to the compromise ranking to obtain a failure mode risk level sequence; Based on the reliability data of the wind power equipment system, a dynamic reliability network model is constructed, and the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model are obtained. The dynamic reliability network model includes multiple risk nodes.

2. The method as described in claim 1, characterized in that, The method based on maximum likelihood estimation and wind power system fault data is used to obtain maximum likelihood estimates and distribution functions for each distribution type in the wind power system. Specifically, this includes: The number of groups is determined based on parameter estimation and empirical formulas. The unknown parameters in the probability density function are calculated based on the maximum likelihood estimation method and the fault data of the wind power equipment system. Based on the maximum likelihood estimation method and the distribution types in the fault data of the wind power equipment system, the maximum likelihood estimate of each distribution type is calculated.

3. The method as described in claim 1, characterized in that, The step of constructing a dynamic reliability network model based on the reliability data of the wind power equipment system specifically includes: Reliability data of wind power equipment systems are collected, and various types of data in the reliability data are used as risk nodes. The reliability data includes data on system components, failure modes, failure causes, and reliability drivers. Based on the rough score number corresponding to each risk node, calculate the ideal solution and negative ideal solution of each risk node under different risk factors, so as to obtain the distance and compromise ranking from the rough score number corresponding to each risk node to the ideal solution. Using the compromise sorting as the node hazard degree, the reliability hazard degree corresponding to the wind power equipment system components, failure modes and failure causes is obtained respectively, and each reliability hazard degree is used as the node weight of the corresponding risk node; Based on the failure data in the failure modes, the time function of the failure rate under different propagation relationships is calculated, and the time function is used as the edge weight of the corresponding risk node to construct a dynamic reliability network model.

4. The method as described in claim 1, characterized in that, It also includes evaluating the reliability level of the wind power equipment system under the current state based on the first system reliability level influence function and the second system reliability level influence function, wherein the expressions for the first system reliability level influence function and the second system reliability level influence function are: ； ； in, This represents the function that influences the reliability level of the first system. This represents the matrix of maximum values ​​affecting reliability indicators. The weight matrix representing the nodes causing the failure. Represents the expert rating matrix. This represents the influence weight matrix among the causes of failure. This represents the weighting matrix indicating the influence between failure causes and typical system failure modes. This represents the node weight matrix for typical failure modes of the system. This represents the influence weight matrix among typical failure modes of the system. This represents the influence weight matrix between typical failure modes of the system and system components. This represents the system component node weight matrix. This represents the influence weight matrix between system components. This represents the influence weight matrix between system components and the central node. This represents the function that influences the reliability level of the second system. This represents the impact matrix of the monitoring indicators.

5. The method as described in claim 1, characterized in that, The expression for the reliability index function is: ； ； ； ； in, Indicates the influence of the central node. Indicates the number of groups. Indicates the influence of system component nodes. i Indicates the first i Risk nodes j Indicates the first j Risk nodes F Indicates the impact of nodes in typical system failure modes. f Indicates the degree of influence of the node causing the failure. The weight matrix representing the nodes causing the failure. This represents the node weight matrix for typical failure modes of the system. This represents the system component node weight matrix. Represents system component nodes To system component nodes Influence weight, Represents system component nodes The influence weight of the central node Indicates the fault mode To Fail Mode Influence weight, Indicates the fault mode To system components Influence weight, Node indicating the cause of failure Failure Cause Node Influence weight matrix Node indicating the cause of failure With typical system failure mode nodes Influence weight matrix dist ( ) represents the distance function. This represents the impact matrix of the monitoring indicators. Represents the interval number.

6. A wind power equipment operation reliability analysis system based on time-varying conditions, characterized in that, The wind power equipment operation reliability analysis system (1) includes a function fitting module (11), a risk ranking module (12), and a reliability analysis module (13), wherein, The function fitting module (11) is used to obtain the maximum likelihood estimate and distribution function of each distribution type in the wind power equipment system based on the maximum likelihood estimation method and the fault data of the wind power equipment system, and to perform fitting processing on the distribution function of each distribution type according to the improved DS evidence theory to obtain the probability density function of the fault data. The process of fitting the distribution functions of each distribution type according to the improved DS evidence theory to obtain the probability density function of fault data specifically includes: Calculate the slope between two adjacent points of the probability density function corresponding to the fault data, the median of two adjacent points of the probability density function corresponding to the fault data, and the difference of the median of the probability density function corresponding to the fault data, respectively. The slope and the median difference are processed by maximum value image processing respectively, and the gray correlation degree between the probability density functions corresponding to the fault data is calculated to obtain the probability density function of the fault data. The risk ranking module (12) is used to optimize the scoring process of each failure mode in the FMEA method according to the rough set theory, and to calculate the hazard of the failure mode, wherein the hazard of the failure mode includes the scoring rough number corresponding to the failure mode and the failure mode risk level ranking. The optimization of the failure mode scoring process in the FMEA method based on rough set theory, and the calculation of the failure mode severity, specifically include: Obtain the scoring matrix for each failure mode and the weight vector corresponding to each risk factor. Based on the scoring transformation function, convert the risk factor scores corresponding to all scoring matrices into rough scores. Under different risk factors, the distance from the rough number of each failure mode score to the ideal solution is calculated to obtain the group utility maximization ranking, individual regret minimization ranking, and compromise ranking of each failure mode. The failure modes are sorted according to the compromise ranking to obtain a failure mode risk level sequence; The reliability analysis module (13) is used to construct a dynamic reliability network model based on the reliability data of the wind power equipment system, and to obtain the reliability index function and reliability sensitivity of each risk node in the dynamic reliability network model, wherein the dynamic reliability network model includes multiple risk nodes.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-5.

Images