Charging operation health state evaluation of charging pile based on fuzzy comprehensive evaluation method

Abstract

This invention discloses a charging pile charging operation health status assessment based on a fuzzy comprehensive evaluation method. The method involves: 1) establishing a factor set and a comment set for the charging pile charging operation health status, and using the entropy weight method to obtain an updated evaluation index weight vector; 2) establishing a health level classification table for the charging operation status evaluation indicators, and determining the membership function of the evaluation indicators to the comment levels; 3) inputting the real-time collected evaluation index values ​​online, and using a weighted average operator to perform fuzzy synthesis of the weight and membership matrix to obtain the fuzzy comprehensive membership vector, health level, and comprehensive score for the current charging operation status. This method adopts a standardized evaluation process, combining offline and online methods, resulting in traceable and comprehensive evaluation results. It improves the accuracy and speed of charging operation health status assessment, and is more helpful for subsequent safety analysis and operation and maintenance decisions.

Description

Technical Field

[0001] This invention belongs to the field of charging pile status evaluation technology, and in particular relates to the assessment of the health status of charging pile operation based on the fuzzy comprehensive evaluation method. Background Technology

[0002] In recent years, with the rapid development of the new energy vehicle industry, electric vehicles, as an important carrier of green travel, have shown a rapid growth trend. The surge in the number of electric vehicles has directly driven a large-scale demand for charging facilities. As of the end of December 2025, my country had 20.092 million electric vehicle charging facilities, forming the world's largest electric vehicle charging network, capable of supporting the charging needs of over 40 million new energy vehicles. However, charging piles are usually located outdoors, and due to multiple factors such as high-frequency charge-discharge cycles, component aging, grid voltage fluctuations, and improper human operation, their charging operation status will continue to deteriorate. This will not only lead to decreased charging efficiency and reduced service quality, but may also cause safety hazards such as overheating, short circuits, and leakage, resulting in equipment failure and shutdown, and a reduction in effective service capacity. This increases the operation and maintenance costs of operators and exacerbates users' charging anxiety, seriously hindering the high-quality development of the charging infrastructure industry. Therefore, it is urgent to conduct health status assessments of charging pile operations to ensure the reliability and safety of charging operations.

[0003] Currently, most assessments of the health status of charging piles rely on real-time data collection of parameters such as charging current, voltage, and temperature. These parameters are then used to evaluate whether the charging operation is normal based on preset thresholds or simple algorithms, thus aiding in maintenance decisions. However, simply outputting a binary judgment of "normal" or "abnormal" is insufficient for refined assessments, and the evaluation results do not provide the distribution characteristics of the health status, making them unsuitable for decision-making. The health status of charging pile operation is influenced by multiple factors, constituting a comprehensive evaluation problem with multiple evaluation indicators. Current methods for comprehensive evaluation with multiple indicators include the analytic hierarchy process (AHP), grey relational analysis, fuzzy comprehensive evaluation, and artificial neural networks. However, due to the inadequacy of continuous variable monitoring data during charging pile operation, methods such as AHP and artificial neural networks have certain limitations. Furthermore, the factors influencing the health status of charging piles exhibit a degree of "greyness," leading to a degree of "fuzziness" in the evaluation results when determining whether a charging pile is in a "healthy," "abnormal," or other state. The fuzzy comprehensive evaluation method utilizes fuzzy mathematics theory to comprehensively evaluate things, with a rigorous evaluation process. Unlike traditional methods that output a single score, the fuzzy comprehensive evaluation result is an evaluation vector, representing the degree of membership of the evaluated object to each level. It not only provides a final conclusion but also retains the distribution characteristics of the evaluation, making it more helpful for decision-making. In summary, this patent uses the fuzzy comprehensive evaluation method combined with the entropy weight method to conduct a fuzzy comprehensive evaluation of the charging operation status of charging piles, obtaining the health level and comprehensive health status score of the charging operation status, thus achieving accuracy and comprehensiveness in the assessment. Summary of the Invention

[0004] This invention aims to solve one of the technical problems existing in the prior art. To this end, this invention proposes a charging pile charging operation health status assessment based on the fuzzy comprehensive evaluation method, including the following steps:

[0005] 1) Establish a set of factors and a set of comments for the health status of charging pile operation. Based on the evaluation index value data of the factor set in historical operation, use the entropy weight method to obtain the weight vector of the evaluation index.

[0006] 2) Establish a health level classification table for the charging operation status evaluation indicators, and determine the membership function of the evaluation indicators to the evaluation level;

[0007] 3) Input the evaluation index values ​​collected in real time online, obtain the evaluation index membership matrix based on the constructed membership function, and then use the weighted average operator to perform fuzzy synthesis of the weights and membership matrix to obtain the fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status.

[0008] The establishment of the factor set and evaluation set for the healthy operation status of charging piles is based on historical factor set evaluation index values, and the corresponding weight vector is obtained using the entropy weight method, as follows:

[0009] 1) The charging operation health status of charging piles is affected by multiple operating variables. Combining the principles of high sensitivity, easy quantification, operability, and comprehensiveness of evaluation indicators, continuous quantities that can reflect the charging operation status of charging piles are selected to establish a factor set for the charging operation health status. This set includes five evaluation indicators: pile output voltage, output voltage change rate, pile output current change rate, emergency stop switch trigger count, and charging gun temperature. The factor set is X={x1, x2, x3, x4, x5}, where x1 = "pile output voltage", x2 = "output voltage change rate", x3 = "pile output current change rate", x4 = "emergency stop switch trigger count", and x5 = "charging gun temperature". The output voltage change rate and pile output current change rate are the changes in the current sampling time relative to the previous sampling time, expressed as a percentage. The emergency stop switch trigger count is the cumulative number of emergency stop switch triggers for the charging pile, which is an integer greater than or equal to 0.

[0010] 2) The charging operation health status of charging piles is divided into four levels: healthy, sub-healthy, abnormal, and faulty. A rating set V={v1, v2, v3, v4} is established, where v1, v2, v3, and v4 represent the healthy, sub-healthy, abnormal, and faulty levels, respectively. At the healthy level, the charging pile is in normal operation, and all evaluation index parameters are far from the threshold or within the standard range. At the sub-healthy level, the overall charging performance of the charging pile is reduced, but it does not currently affect its charging operation. At the abnormal level, the overall charging performance of the charging pile has significantly decreased, and the possibility of a fault has increased. At the faulty level, the charging pile can no longer charge normally, and all evaluation index parameters exceed the threshold. A corresponding rating set T=(t1, t2, t3, t4) is established. T t1, t2, t3, and t4 are the scores corresponding to the levels of health, sub-health, abnormality, and malfunction, respectively.

[0011] 3) Based on the historical factor set evaluation index value data, the entropy weight method is used to obtain the evaluation index weight vector. The specific steps are as follows:

[0012] Based on the optimal value characteristics, the output voltage change rate, pile output current change rate, and emergency stop switch trigger count evaluation indicators in the factor set are minimal indicators, while the pile output voltage and charging gun temperature indicators are interval indicators. The collected historical factor set evaluation indicator values ​​are standardized using the extreme value normalization method to obtain a standardized matrix, represented as follows:

[0013] (1)

[0014] Where X n×5 For a standardized matrix, x n1 x n2 x n3 x n4 x n5 These are the standardized values ​​of the charging pile's output voltage, output voltage change rate, output current change rate, emergency stop switch trigger count, and charging gun temperature evaluation index for the nth charging operation state, respectively, where n is the number of historical charging operation states taken.

[0015] The weights of the five evaluation indicators are obtained using the entropy weight method based on the standardized matrix. The formula for calculating the weights is as follows:

[0016] (2)

[0017] Where ω j Let x be the weight of the j-th evaluation indicator. ij Let e ​​be the standardized value of the j-th evaluation index of the charging pile's i-th charging operation status. j Information entropy;

[0018] The calculation of the entropy weight method weighting model is triggered by factors such as regular events, an increase of 1 in the number of emergency stop switch triggers, and charging pile maintenance events, and the weight vector of the evaluation indicators is updated. When the number of emergency stop switch triggers increases by 1, based on engineering experience and actual operation data, the importance of the emergency stop switch trigger count indicator is set to be β times that of other indicators, and its weight is updated. Then, the increment of the weight of the emergency stop switch trigger count indicator is reduced by the weight of the other 4 evaluation indicators.

[0019] The establishment of a health level classification table for charging operation status evaluation indicators, and the determination of the membership function of the evaluation indicators to the evaluation level, are detailed as follows:

[0020] 1) The evaluation criteria for charging operation status are the same as the evaluation criteria for the charging pile's charging operation health status, and are categorized as healthy, sub-healthy, abnormal, and faulty. Historical data of the evaluation criteria for charging operation status under normal charging and charging fault conditions are statistically analyzed, and relevant national and industry standards are combined to obtain the threshold values ​​for the healthy, sub-healthy, abnormal, and faulty levels of the evaluation criteria, forming a health level classification table for the evaluation criteria for charging operation status.

[0021] 2) The thresholds for the five evaluation indicators for health, sub-health, abnormality and fault levels are divided into single-interval type, double-interval type and single-value type. Among them, the level of emergency stop switch triggering frequency is divided into single-value type. The single-interval type and double-interval type use trapezoidal membership function.

[0022] Single-interval thresholds define a complete membership region for a given level using a single interval range, with transition zones at both ends; let the evaluation index's rating level v be... r The single-section threshold is [b] r , c r ], b r c r The rating levels are v respectively. r Completely subordinate to the starting and ending points, r=1, 2, 3, 4, this evaluation index affects the comment level v. r The trapezoidal membership function is defined by four parameters: a r b r c r d r a r d r Rating level v r The lower and upper limits of a r r <c r <d r , respectively, are the turning points where the membership degree increases from 0 to 1, remains at 1, and then decreases back to 0. The membership function of the evaluation index to the comment set under the single interval threshold is:

[0023] (3)

[0024] in For a single-interval threshold, the evaluation index value x corresponds to the comment level v. r Membership function;

[0025] The dual-interval threshold has membership in both discontinuous intervals. To maintain function continuity and computability, a single piecewise trapezoidal function is used. Let the evaluation index's rating level v be... r The dual-interval segmented threshold is [f] r , g r ]、[j r , k r ], f r g r The rating levels are v respectively. r The first point that is completely subordinate to both the start and end points, j r k r The rating levels are v respectively. r The second point is completely subordinate to the start and end points; this evaluation index affects the rating level v. r The trapezoidal membership function is defined by 8 parameters, e r f r g r h r i r j r k​r l r e r <f r <g r <h r r <j r <k r <l r e r l r h represents the inflection point between the upper and lower limits of the membership degree value of 0. r i r For the intermediate turning point where the membership degree value is 0, the membership function of the evaluation index to the comment set under the dual-interval threshold is:

[0026] (4)

[0027] in The evaluation index value x under the dual-interval threshold corresponds to the comment level v. r Membership function;

[0028] If the threshold values ​​for the emergency stop switch trigger frequency index are 0, 1, 2, and 3 for the healthy, sub-healthy, abnormal, and faulty levels, then the membership function of the emergency stop switch trigger frequency index to the healthy and sub-healthy levels is:

[0029] (5)

[0030] in , These are the emergency stop switch trigger count indicators. Membership function for health and sub-health levels, For the set membership value, ;

[0031] The membership function of the emergency stop switch trigger count index to the abnormality and fault levels is:

[0032] (6)

[0033] in , These are the emergency stop switch trigger count indicators. Membership function for anomalies and fault levels.

[0034] The evaluation index values ​​collected in real time through online input are used to obtain an evaluation index membership matrix based on the constructed membership function. Then, a weighted average operator is used to perform fuzzy synthesis of the weights and membership matrix to obtain a fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status, as detailed below:

[0035] ​The evaluation index values ​​are collected in real time online. Based on the constructed membership function, the membership matrix of the evaluation index is obtained. Then, a weighted average operator is used to fuzzily synthesize the evaluation index weights and the membership matrix to obtain the fuzzy comprehensive membership vector, which is:

[0036] (7)

[0037] Where B is the fuzzy comprehensive membership vector, and the elements of B are b1, b2, b3, and b4, which are the comprehensive membership degrees of the evaluation index to the four comment levels, respectively. W is the evaluation index weight vector, and R is the evaluation index membership matrix. For the weighted average operator, ω1, ω2, ω3, ω4, and ω5 are the weights of the pile output voltage, output voltage change rate, pile output current change rate, emergency stop switch trigger count, and charging gun temperature evaluation index, respectively. , ,…, , These are the membership values ​​of the five evaluation indicators to the four rating levels;

[0038] In equation (7), the evaluation index weight vector W is obtained by the entropy weight method based on equations (1) and (2), the evaluation index membership matrix R is obtained based on the membership function of the evaluation index to the health, sub-health, abnormal and fault levels, equations (3)-(6), and the weighted average operator is calculated as follows.

[0039] (8)

[0040] Where b k To evaluate the overall membership degree of the indicator to the rating level k, r jk Let be the membership value of the j-th evaluation indicator to the comment level k;

[0041] Furthermore, based on actual application scenarios, different state division strategies are adopted: the maximum membership principle, the weighted average principle, and the fuzzy vector single-value principle. These strategies map the fuzzy comprehensive membership vector to the charging pile's charging operation health status level, obtaining the health level assessment result. The comprehensive charging operation health status score is then calculated as follows:

[0042] (9)

[0043] F represents the overall health score of the charging operation.

[0044] The embodiments of the present invention have at least the following beneficial technical effects:

[0045] 1) This invention proposes a charging pile charging operation health status assessment strategy based on fuzzy comprehensive evaluation method. It adopts a standardized process from factor set, comment set, weight vector, membership matrix to synthesis operation. The fuzzy comprehensive evaluation results include the membership degree of the level, health level, and quantified comprehensive health status score, which makes the charging operation health status assessment results traceable and comprehensive, and more helpful for decision-making reference.

[0046] 2) This invention proposes a charging pile operation health status assessment method that combines offline and online methods. Offline, the entropy weight method is used to obtain the dynamic weights of the evaluation indicators and the membership function of the evaluation indicators to the rating level. Online, the membership matrix and fuzzy synthesis are calculated, which can improve the accuracy and speed of charging operation health status assessment. Attached Figure Description

[0047] Figure 1 This is a flowchart of the charging operation health status assessment of charging piles based on the fuzzy comprehensive evaluation method provided in the embodiments of the present invention. Detailed Implementation

[0048] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the following embodiments are by no means intended to limit the present invention.

[0049] Health status assessment of charging pile operation based on fuzzy comprehensive evaluation method, such as Figure 1 As shown, it includes the following steps:

[0050] 1) Establish a set of factors and a set of comments for the health status of charging pile operation. Based on the evaluation index value data of the factor set in historical operation, use the entropy weight method to obtain the weight vector of the evaluation index.

[0051] 2) Establish a health level classification table for the charging operation status evaluation indicators, and determine the membership function of the evaluation indicators to the evaluation level;

[0052] 3) Input the evaluation index values ​​collected in real time online, obtain the evaluation index membership matrix based on the constructed membership function, and then use the weighted average operator to perform fuzzy synthesis of the weights and membership matrix to obtain the fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status.

[0053] The establishment of the factor set and evaluation set for the healthy operation status of charging piles is based on historical factor set evaluation index values, and the corresponding weight vector is obtained using the entropy weight method, as follows:

[0054] 1) The charging operation health status of charging piles is affected by multiple operating variables. Combining the principles of high sensitivity, easy quantification, operability, and comprehensiveness of evaluation indicators, continuous quantities that can reflect the charging operation status of charging piles are selected to establish a factor set for the charging operation health status. This set includes five evaluation indicators: pile output voltage, output voltage change rate, pile output current change rate, emergency stop switch trigger count, and charging gun temperature. The factor set is X={x1, x2, x3, x4, x5}, where x1 = "pile output voltage", x2 = "output voltage change rate", x3 = "pile output current change rate", x4 = "emergency stop switch trigger count", and x5 = "charging gun temperature". The output voltage change rate and pile output current change rate are the changes in the current sampling time relative to the previous sampling time, expressed as a percentage. The emergency stop switch trigger count is the cumulative number of emergency stop switch triggers for the charging pile, which is an integer greater than or equal to 0.

[0055] 2) The charging operation health status of charging piles is divided into four levels: healthy, sub-healthy, abnormal, and faulty. A rating set V={v1, v2, v3, v4} is established, where v1, v2, v3, and v4 represent the healthy, sub-healthy, abnormal, and faulty levels, respectively. At the healthy level, the charging pile is in normal operation, and all evaluation index parameters are far from the threshold or within the standard range. At the sub-healthy level, the overall charging performance of the charging pile is reduced, but it does not currently affect its charging operation. At the abnormal level, the overall charging performance of the charging pile has significantly decreased, and the possibility of a fault has increased. At the faulty level, the charging pile can no longer charge normally, and all evaluation index parameters exceed the threshold. A corresponding rating set T=(t1, t2, t3, t4) is established. T t1, t2, t3, and t4 are the scores corresponding to the levels of health, sub-health, abnormality, and malfunction, respectively.

[0056] 3) Based on the historical factor set evaluation index value data, the entropy weight method is used to obtain the evaluation index weight vector. The specific steps are as follows:

[0057] Based on the optimal value characteristics, the output voltage change rate, pile output current change rate, and emergency stop switch trigger count evaluation indicators in the factor set are minimal indicators, while the pile output voltage and charging gun temperature indicators are interval indicators. The collected historical factor set evaluation indicator values ​​are standardized using the extreme value normalization method to obtain a standardized matrix, represented as follows:

[0058] (1)

[0059] Where X n×5 For a standardized matrix, x n1 x n2 x n3 x n4x n5 These are the standardized values ​​of the charging pile's output voltage, output voltage change rate, output current change rate, emergency stop switch trigger count, and charging gun temperature evaluation index for the nth charging operation state, respectively, where n is the number of historical charging operation states taken.

[0060] The weights of the five evaluation indicators are obtained using the entropy weight method based on the standardized matrix. The formula for calculating the weights is as follows:

[0061] (2)

[0062] Where ω j Let x be the weight of the j-th evaluation indicator. ij Let e ​​be the standardized value of the j-th evaluation index of the charging pile's i-th charging operation status. j Information entropy;

[0063] The calculation of the entropy weight method weighting model is triggered by factors such as regular events, an increase of 1 in the number of emergency stop switch triggers, and charging pile maintenance events, and the weight vector of the evaluation indicators is updated. When the number of emergency stop switch triggers increases by 1, based on engineering experience and actual operation data, the importance of the emergency stop switch trigger count indicator is set to be β times that of other indicators, and its weight is updated. Then, the increment of the weight of the emergency stop switch trigger count indicator is reduced by the weight of the other 4 evaluation indicators.

[0064] The establishment of a health level classification table for charging operation status evaluation indicators, and the determination of the membership function of the evaluation indicators to the evaluation level, are detailed as follows:

[0065] 1) The evaluation criteria for charging operation status are the same as the evaluation criteria for the charging pile's charging operation health status, and are categorized as healthy, sub-healthy, abnormal, and faulty. Historical data of the evaluation criteria for charging operation status under normal charging and charging fault conditions are statistically analyzed, and relevant national and industry standards are combined to obtain the threshold values ​​for the healthy, sub-healthy, abnormal, and faulty levels of the evaluation criteria, forming a health level classification table for the evaluation criteria for charging operation status.

[0066] 2) The thresholds for the five evaluation indicators for health, sub-health, abnormality and fault levels are divided into single-interval type, double-interval type and single-value type. Among them, the level of emergency stop switch triggering frequency is divided into single-value type. The single-interval type and double-interval type use trapezoidal membership function.

[0067] Single-interval thresholds define a complete membership region for a given level using a single interval range, with transition zones at both ends; let the evaluation index's rating level v be... r The single-section threshold is [b] r , c r ], br c r The rating levels are v respectively. r Completely subordinate to the starting and ending points, r=1, 2, 3, 4, this evaluation index affects the comment level v. r The trapezoidal membership function is defined by four parameters: a r b r c r d r a r d r Rating level v r The lower and upper limits of a r r <c r <d r , respectively, are the turning points where the membership degree increases from 0 to 1, remains at 1, and then decreases back to 0. The membership function of the evaluation index to the comment set under the single interval threshold is:

[0068] (3)

[0069] in For a single-interval threshold, the evaluation index value x corresponds to the comment level v. r Membership function;

[0070] The dual-interval threshold has membership in both discontinuous intervals. To maintain function continuity and computability, a single piecewise trapezoidal function is used. Let the evaluation index's rating level v be... r The dual-interval segmented threshold is [f] r , g r ]、[j r , k r ], f r g r The rating levels are v respectively. r The first point that is completely subordinate to both the start and end points, j r k r The rating levels are v respectively. r The second point is completely subordinate to the start and end points; this evaluation index affects the rating level v. r The trapezoidal membership function is defined by 8 parameters, e r f r g r h r i r j r k r l r e r <f r <g r <h r r <j​​r <k r <l r e r l r h represents the inflection point between the upper and lower limits of the membership degree value of 0. r i r For the intermediate turning point where the membership degree value is 0, the membership function of the evaluation index to the comment set under the dual-interval threshold is:

[0071] (4)

[0072] in The evaluation index value x under the dual-interval threshold corresponds to the comment level v. r Membership function;

[0073] If the threshold values ​​for the emergency stop switch trigger frequency index are 0, 1, 2, and 3 for the healthy, sub-healthy, abnormal, and faulty levels, then the membership function of the emergency stop switch trigger frequency index to the healthy and sub-healthy levels is:

[0074] (5)

[0075] in , These are the emergency stop switch trigger count indicators. Membership function for health and sub-health levels, For the set membership value, ;

[0076] The membership function of the emergency stop switch trigger count index to the abnormality and fault levels is:

[0077] (6)

[0078] in , These are the emergency stop switch trigger count indicators. Membership function for anomalies and fault levels.

[0079] The evaluation index values ​​collected in real time through online input are used to obtain an evaluation index membership matrix based on the constructed membership function. Then, a weighted average operator is used to perform fuzzy synthesis of the weights and membership matrix to obtain a fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status, as detailed below:

[0080] The evaluation index values ​​are collected in real time online. Based on the constructed membership function, the membership matrix of the evaluation index is obtained. Then, a weighted average operator is used to fuzzily synthesize the evaluation index weights and the membership matrix to obtain the fuzzy comprehensive membership vector, which is:

[0081] (7)

[0082] Where B is the fuzzy comprehensive membership vector, and the elements of B are b1, b2, b3, and b4, which are the comprehensive membership degrees of the evaluation index to the four comment levels, respectively. W is the evaluation index weight vector, and R is the evaluation index membership matrix. For the weighted average operator, ω1, ω2, ω3, ω4, and ω5 are the weights of the pile output voltage, output voltage change rate, pile output current change rate, emergency stop switch trigger count, and charging gun temperature evaluation index, respectively. , ,…, , These are the membership values ​​of the five evaluation indicators to the four rating levels;

[0083] In equation (7), the evaluation index weight vector W is obtained by the entropy weight method based on equations (1) and (2), the evaluation index membership matrix R is obtained based on the membership function of the evaluation index to the health, sub-health, abnormal and fault levels, equations (3)-(6), and the weighted average operator is calculated as follows.

[0084] (8)

[0085] Where b k To evaluate the overall membership degree of the indicator to the rating level k, r jk Let be the membership value of the j-th evaluation indicator to the comment level k;

[0086] Furthermore, based on actual application scenarios, different state division strategies are adopted: the maximum membership principle, the weighted average principle, and the fuzzy vector single-value principle. These strategies map the fuzzy comprehensive membership vector to the charging pile's charging operation health status level, obtaining the health level assessment result. The comprehensive charging operation health status score is then calculated as follows:

[0087] (9)

[0088] F represents the overall health score of the charging operation.

[0089] Taking the health status assessment of charging piles in a charging station in a certain area of ​​southwest my country as an example, the charging piles are of the following types: DC charging piles, voltage level of 500V, and rated power of 125kW. The health status assessment of charging piles based on the fuzzy comprehensive evaluation method proposed in this invention is used to obtain the health level and comprehensive score of the charging operation status of the charging piles in the example charging station.

[0090] 1) Establish a set of factors and a set of comments to determine the health status of the charging pile. The factor set is X = {pile output voltage, output voltage change rate, pile output current change rate, number of times the emergency stop switch is triggered, charging gun temperature}, and the comment set is V = {healthy, sub-healthy, abnormal, faulty}. Set the corresponding rating set T = (5, 4, 2, 1). T Based on the historical factor set evaluation index value data, the weight vectors of the five evaluation indicators were obtained offline using the entropy weight method and dynamically updated. The weights of the evaluation indicators under different emergency stop switch triggering times are shown in Table 1.

[0091] As shown in Table 1, the weight of the emergency stop switch increases with the number of times it is triggered, indicating that the dynamic weighting method of entropy weight proposed in this invention can better characterize the cumulative risk of emergency stop switch triggering during charging operation of charging piles, and has certain engineering application value.

[0092] Table 1. Weights of evaluation indicators for different emergency stop switch trigger counts

[0093] 2) Establish a health level classification table for the charging operation status evaluation indicators and determine the membership function of the evaluation indicators to the evaluation level; the health level classification table of the five evaluation indicators is shown in Table 2.

[0094] Table 2. Health Level Classification Table Based on 5 Evaluation Indicators

[0095] 3) Online input of real-time collected evaluation index data of a charging pile's operation status during a specific charging period on May 26, 2025, from 16:45:35 to 17:14:56. Based on the constructed membership function, the membership matrix of the evaluation index is obtained. Then, the weighted average operator is used to perform fuzzy synthesis of the weights and membership matrix to obtain the fuzzy comprehensive membership vector. Finally, the maximum membership principle is used to obtain the health level of the charging operation status and calculate the comprehensive health status score, as shown in Table 3.

[0096] Table 3 shows that: 1) The obtained fuzzy comprehensive membership vector provides the membership degree of the evaluation index to each health level, and different state division strategies can be adopted to obtain the health level of the charging operation state according to different application scenarios; 2) Under the principle of maximum membership, the health level of the charging operation state corresponding to sampling points 12, 15, and 17 is sub-healthy, which is due to the large change rate of the output current of the charging pile; 3) The quantitative comprehensive health status scores under the same health level are different, which facilitates subsequent decision analysis based on the charging operation health status. The above evaluation results and analysis verify the effectiveness, accuracy, and comprehensiveness of the method proposed in this invention.

[0097] Table 3. Fuzzy comprehensive membership vector, health level, and comprehensive health status score of charging operation status.

[0098]

Claims

1. A charging pile charging operation health status assessment based on fuzzy comprehensive evaluation method, characterized in that, The charging pile charging operation health status assessment based on the fuzzy comprehensive evaluation method includes the following steps: 1) Establish a set of factors and a set of comments for the health status of charging pile operation. Based on the evaluation index value data of the factor set in historical operation, use the entropy weight method to obtain the weight vector of the evaluation index. 2) Establish a health level classification table for the charging operation status evaluation indicators, and determine the membership function of the evaluation indicators to the evaluation level; 3) Input the evaluation index values ​​collected in real time online, obtain the evaluation index membership matrix based on the constructed membership function, and then use the weighted average operator to perform fuzzy synthesis of the weights and membership matrix to obtain the fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status.

2. The charging pile charging operation health status assessment based on the fuzzy comprehensive evaluation method according to claim 1, characterized in that, The establishment of a health level classification table for charging operation status evaluation indicators and the determination of the membership function of evaluation indicators to evaluation levels include the following steps: 1) The evaluation criteria for charging operation status are the same as the evaluation criteria for the charging pile's charging operation health status, and are categorized as healthy, sub-healthy, abnormal, and faulty. Historical data of the evaluation criteria for charging operation status under normal charging and charging fault conditions are statistically analyzed, and relevant national and industry standards are combined to obtain the threshold values ​​for the healthy, sub-healthy, abnormal, and faulty levels of the evaluation criteria, forming a health level classification table for the evaluation criteria for charging operation status. 2) The thresholds for the five evaluation indicators for health, sub-health, abnormality and fault levels are divided into single-interval type, double-interval type and single-value type. Among them, the level of emergency stop switch triggering frequency is divided into single-value type. The single-interval type and double-interval type use trapezoidal membership function. Single-interval thresholds define a complete membership region for a given level using a single interval range, with transition zones at both ends; let the evaluation index's rating level v be... r The single-section threshold is [b] r , c r ], b r c r The rating levels are v respectively. r Completely subordinate to the starting and ending points, r=1, 2, 3, 4, this evaluation index affects the comment level v. r The trapezoidal membership function is defined by four parameters: a r b r c r d r a r d r Rating level v r The lower and upper limits of a r r <c r <d r , respectively, are the turning points where the membership degree increases from 0 to 1, remains at 1, and then decreases back to 0. The membership function of the evaluation index to the comment set under the single interval threshold is:​ (1) in For a single-interval threshold, the evaluation index value x corresponds to the comment level v. r Membership function; The dual-interval threshold has membership in both discontinuous intervals. To maintain function continuity and computability, a single piecewise trapezoidal function is used. Let the evaluation index's rating level v be... r The dual-interval segmented threshold is [f] r , g r ]、[j r , k r ], f r g r The rating levels are v respectively. r The first point that is completely subordinate to both the start and end points, j r k r The rating levels are v respectively. r The second point is completely subordinate to the start and end points; this evaluation index affects the rating level v. r The trapezoidal membership function is defined by 8 parameters, e r f r g r h r i r j r k r l r e r <f r <g r <h r r <j r <k r <l r e r l r h represents the inflection point between the upper and lower limits of the membership degree value of 0. r i r For the intermediate turning point where the membership degree value is 0, the membership function of the evaluation index to the comment set under the dual-interval threshold is:​ (2) in The evaluation index value x under the dual-interval threshold corresponds to the comment level v. r Membership function; If the threshold values ​​for the emergency stop switch trigger frequency index are 0, 1, 2, and 3 for the healthy, sub-healthy, abnormal, and faulty levels, then the membership function of the emergency stop switch trigger frequency index to the healthy and sub-healthy levels is: (3) in , These are the emergency stop switch trigger count indicators. Membership function for health and sub-health levels, For the set membership value, ; The membership function of the emergency stop switch trigger count index to the abnormality and fault levels is: (4) in , These are the emergency stop switch trigger count indicators. Membership function for anomalies and fault levels.

3. The charging pile charging operation health status assessment based on the fuzzy comprehensive evaluation method according to claim 1, characterized in that, The evaluation index values ​​collected in real time through online input are used to obtain the evaluation index membership matrix based on the constructed membership function. Then, a weighted average operator is used to perform fuzzy synthesis of the weights and membership matrix to obtain the fuzzy comprehensive membership vector, thereby obtaining the health level and comprehensive score of the charging operation status, as detailed below: The evaluation index values ​​are collected in real time online. Based on the constructed membership function, the membership matrix of the evaluation index is obtained. Then, a weighted average operator is used to fuzzily synthesize the evaluation index weights and the membership matrix to obtain the fuzzy comprehensive membership vector, which is: (5) Where B is the fuzzy comprehensive membership vector, and the elements of B are b1, b2, b3, and b4, which are the comprehensive membership degrees of the evaluation index to the four comment levels, respectively. W is the evaluation index weight vector, and R is the evaluation index membership matrix. For the weighted average operator, ω1, ω2, ω3, ω4, and ω5 are the weights of the pile output voltage, output voltage change rate, pile output current change rate, emergency stop switch trigger count, and charging gun temperature evaluation index, respectively. , ,…, , These are the membership values ​​of the five evaluation indicators to the four rating levels; In equation (5), the evaluation index weight vector W is obtained by the entropy weight method, the evaluation index membership matrix R is obtained based on the membership function of the evaluation index to the health, sub-health, abnormal and fault levels, equation (3)-equation (6), and the weighted average operator is calculated as follows: (6) Where b k To evaluate the overall membership degree of the indicator to the rating level k, r jk Let be the membership value of the j-th evaluation indicator to the comment level k; Furthermore, based on actual application scenarios, different state division strategies are adopted: the maximum membership principle, the weighted average principle, and the fuzzy vector single-value principle. These strategies map the fuzzy comprehensive membership vector to the charging pile's charging operation health status level, obtaining the health level assessment result. The comprehensive charging operation health status score is then calculated as follows: (7) F represents the overall health score of the charging operation.

Images