A quality evaluation method of intelligent circuit breaker

Abstract

The application discloses a kind of quality evaluation methods of intelligent circuit breaker, it is related to circuit breaker evaluation technical field, the method includes: connecting circuit breaker detection system, call target circuit breaker's factory detection dataset;Determine the initial quality coefficient of circuit breaker;Obtain M circuit breaker use record group and Q circuit breaker maintenance record group;Generate circuit breaker first quality coefficient;Judge whether circuit breaker first quality coefficient is less than / equal to predetermined circuit breaker quality threshold;If circuit breaker first quality coefficient is less than / equal to predetermined circuit breaker quality threshold, determine circuit breaker second quality coefficient;Correctly obtain circuit breaker third quality coefficient.Solve the existing circuit breaker evaluation only relies on factory detection data, cannot be dynamically adjusted according to the use and maintenance record of circuit breaker, in turn lead to the technical problem that the accuracy of circuit breaker quality evaluation result is insufficient, reliability is poor, reaches the technical effect that the accuracy and reliability of circuit breaker quality evaluation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of circuit breaker evaluation, and particularly relates to a quality evaluation method of an intelligent circuit breaker. BACKGROUND

[0002] With the rapid development of science and technology, the safety and reliability requirements of the power system for equipment are increasingly improved. As an indispensable key component in the power system, the quality evaluation and monitoring of the intelligent circuit breaker become an important link to ensure the stable operation of the power system. The intelligent circuit breaker integrates advanced sensor technology, data processing technology and communication technology, can monitor the key parameters such as current, voltage, power, temperature and humidity in the circuit in real time, and quickly cut off the circuit in abnormal conditions to protect the equipment from damage. However, the highly integrated design brings complexity to the quality evaluation. The traditional circuit breaker evaluation method often relies on static test data at the factory, which cannot fully reflect the performance changes and potential problems of the circuit breaker in actual operation, and it is difficult to dynamically adjust the quality coefficient of the circuit breaker and maintain it in time, which reduces the service life of the circuit breaker and further affects the operation stability and safety of the entire power system.

[0003] In the existing circuit breaker quality evaluation technology, the circuit breaker quality evaluation only relies on factory detection data and cannot be dynamically adjusted according to the use and maintenance records of the circuit breaker, which leads to the technical problems of insufficient accuracy and poor reliability of the circuit breaker quality evaluation results. SUMMARY

[0004] The present application provides a quality evaluation method of an intelligent circuit breaker, which establishes a circuit breaker quality reward and punishment adjustment formula, quality analysis, quality coefficient correction and other technical means, solves the technical problems of insufficient accuracy and poor reliability of the circuit breaker quality evaluation results caused by the existing circuit breaker evaluation only relying on factory detection data and being unable to dynamically adjust according to the use and maintenance records of the circuit breaker, and achieves the technical effect of improving the accuracy and reliability of the circuit breaker quality evaluation.

[0005] The application provides a quality evaluation method of an intelligent circuit breaker, comprising: connecting a circuit breaker detection system, and calling factory detection data set of a target circuit breaker; performing quality evaluation on the target circuit breaker according to the factory detection data set, and determining an initial quality coefficient of the circuit breaker; connecting a circuit breaker management module, collecting use record information and maintenance record information of the target circuit breaker, obtaining M circuit breaker use record groups and Q circuit breaker maintenance record groups, wherein M and Q are positive integers; performing reward and punishment adjustment on the initial quality coefficient of the circuit breaker according to a circuit breaker quality reward and punishment adjustment formula, combining the M circuit breaker use record groups and the Q circuit breaker maintenance record groups, generating a first quality coefficient of the circuit breaker; judging whether the first quality coefficient of the circuit breaker is less than / equal to a predetermined circuit breaker quality threshold; if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, performing quality analysis on the target circuit breaker according to the circuit breaker detection system, and determining a second quality coefficient of the circuit breaker; correcting the first quality coefficient of the circuit breaker according to the second quality coefficient of the circuit breaker, and obtaining a third quality coefficient of the circuit breaker.

[0006] In a possible implementation, the quality evaluation on the target circuit breaker according to the factory detection data set to determine the initial quality coefficient of the circuit breaker further performs the following processing: a circuit breaker quality evaluation channel is built, wherein the circuit breaker quality evaluation channel comprises P circuit breaker quality evaluation models, and P is a positive integer greater than 1; data cleaning is performed according to the factory detection data set to generate an initial detection matrix of the circuit breaker; the initial detection matrix of the circuit breaker is input into the P circuit breaker quality evaluation models to obtain P initial quality evaluation coefficients of the circuit breaker; proportion calculation is performed based on P quality evaluation accuracy parameters corresponding to the P circuit breaker quality evaluation models to generate P quality evaluation output incentive coefficients; and the P initial quality evaluation coefficients of the circuit breaker are weighted calculated according to the P quality evaluation output incentive coefficients to generate the initial quality coefficient of the circuit breaker.

[0007] In a possible implementation, the initial quality coefficient of the circuit breaker is adjusted for rewards and penalties according to a circuit breaker quality reward and penalty adjustment formula in combination with the M circuit breaker use record groups and the Q circuit breaker maintenance record groups, to generate a first quality coefficient of the circuit breaker, and the following processing is further performed: performing circuit breaker quality attenuation influence analysis according to the M circuit breaker use record groups to determine M circuit breaker quality penalty coefficients; performing circuit breaker quality enhancement incentive analysis according to the Q circuit breaker maintenance record groups to generate Q circuit breaker quality reward coefficients; performing time-sensitive analysis on the M circuit breaker use record groups and the Q circuit breaker maintenance record groups respectively to generate M quality penalty sensitive factors and Q quality reward sensitive factors; inputting the M circuit breaker quality penalty coefficients, the Q circuit breaker quality reward coefficients, the M quality penalty sensitive factors, the Q quality reward sensitive factors, and the initial quality coefficient of the circuit breaker into the circuit breaker quality reward and penalty adjustment formula, to obtain the first quality coefficient of the circuit breaker.

[0008] In a possible implementation, the initial quality coefficient of the circuit breaker is adjusted for rewards and penalties according to a circuit breaker quality reward and penalty adjustment formula in combination with the M circuit breaker use record groups and the Q circuit breaker maintenance record groups, and the following processing is further performed: the circuit breaker quality reward and penalty adjustment formula is as follows:

[0009] ;

[0010] wherein CQY1 represents the first quality coefficient of the circuit breaker, CQY0 represents the initial quality coefficient of the circuit breaker, QPX represents the mth circuit breaker quality penalty coefficient, PXW represents a quality penalty sensitive factor corresponding to the mth circuit breaker quality penalty coefficient, QRZ represents the qth circuit breaker quality reward coefficient, and RZW represents a quality reward sensitive factor corresponding to the qth circuit breaker quality reward coefficient, m and q are positive integers, 1≤m≤M, and 1≤q≤Q. m represents the mth circuit breaker quality penalty coefficient, PXW m represents the mth circuit breaker quality penalty coefficient, PXW q represents the qth circuit breaker quality reward coefficient, RZW q represents the mth circuit breaker quality penalty coefficient, PXW

[0011] In a possible implementation, the M circuit breaker usage record sets are used to analyze the influence of circuit breaker quality degradation, and the M circuit breaker quality penalty coefficients are determined, and the following processing is further performed: according to the circuit breaker management module, a sample circuit breaker usage record set and a sample circuit breaker quality degradation influence coefficient record set are loaded; a quality degradation influence learning model is supervised trained according to the sample circuit breaker usage record set and the sample circuit breaker quality degradation influence coefficient record set, and a quality degradation influence prediction loss coefficient is obtained every time the quality degradation influence learning model is trained for a predetermined number of times; it is judged whether the quality degradation influence prediction loss coefficient is less than a quality degradation influence prediction loss threshold; if the quality degradation influence prediction loss coefficient is less than the quality degradation influence prediction loss threshold, a quality degradation influence analysis channel is generated; and based on the M circuit breaker usage record sets, the M circuit breaker quality penalty coefficients are obtained according to the quality degradation influence analysis channel.

[0012] In a possible implementation, the M circuit breaker usage record sets and the Q circuit breaker maintenance record sets are respectively subjected to time-sensitive analysis, and M quality penalty sensitive factors and Q quality reward sensitive factors are generated, and the following processing is further performed: time information corresponding to the M circuit breaker usage record sets is called, and M usage time nodes are generated; the M usage time nodes are sorted according to the circuit breaker usage time node table, and the M quality penalty sensitive factors are obtained according to the time-sensitive evaluation of the circuit breaker usage time node table; time information corresponding to the Q circuit breaker maintenance record sets is called, and Q maintenance time nodes are generated; the Q maintenance time nodes are sorted, a circuit breaker maintenance time node table is constructed, and the Q quality reward sensitive factors are generated according to the time-sensitive evaluation of the circuit breaker maintenance time node table.

[0013] In a possible implementation, if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, the target circuit breaker is subjected to quality analysis according to the circuit breaker detection system, and a second quality coefficient of the circuit breaker is determined, and the following processing is further performed: if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, a real-time detection data set of the target circuit breaker is obtained according to the circuit breaker detection system; the real-time detection data set is subjected to data cleaning, and a circuit breaker real-time detection matrix is generated; the circuit breaker real-time detection matrix is input into the P circuit breaker quality evaluation models, and P circuit breaker real-time quality evaluation coefficients are obtained; the P circuit breaker real-time quality evaluation coefficients are weighted calculated according to the P quality evaluation output excitation coefficients, and the second quality coefficient of the circuit breaker is generated.

[0014] In a possible implementation, the first quality coefficient of the circuit breaker is corrected according to the second quality coefficient of the circuit breaker to obtain a third quality coefficient of the circuit breaker, and the following processing is further performed: standardization processing is performed according to the first quality coefficient of the circuit breaker and the second quality coefficient of the circuit breaker to generate a first quality standard coefficient of the circuit breaker and a second quality standard coefficient of the circuit breaker; deviation analysis is performed according to the first quality standard coefficient of the circuit breaker and the second quality standard coefficient of the circuit breaker to obtain a circuit breaker quality standard deviation; it is judged whether the circuit breaker quality standard deviation is less than a circuit breaker quality standard deviation threshold; if the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold, mean value calculation is performed according to the first quality standard coefficient of the circuit breaker and the second quality standard coefficient of the circuit breaker to generate the third quality coefficient of the circuit breaker.

[0015] In a possible implementation, it is further judged whether the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold, and the following processing is further performed: if the circuit breaker quality standard deviation is greater than / equal to the circuit breaker quality standard deviation threshold, proportion calculation is performed according to the first quality standard coefficient of the circuit breaker and the second quality standard coefficient of the circuit breaker to obtain a first quality proportion coefficient and a second quality proportion coefficient; the first quality standard coefficient of the circuit breaker and the second quality standard coefficient of the circuit breaker are weighted and calculated according to the first quality proportion coefficient and the second quality proportion coefficient to obtain the third quality coefficient of the circuit breaker.

[0016] The application provides a quality evaluation method of an intelligent circuit breaker. First, a circuit breaker detection system is connected, and factory detection data sets of a target circuit breaker are called. An initial quality coefficient of the circuit breaker is determined. M circuit breaker use record groups and Q circuit breaker maintenance record groups are obtained. A first quality coefficient of the circuit breaker is generated. It is judged whether the first quality coefficient of the circuit breaker is less than / equal to a predetermined circuit breaker quality threshold. If the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, a second quality coefficient of the circuit breaker is determined. The first quality coefficient of the circuit breaker is corrected to obtain a third quality coefficient of the circuit breaker. The method solves the technical problems that existing circuit breaker evaluation only depends on factory detection data and cannot be dynamically adjusted according to circuit breaker use and maintenance records, and thus the accuracy of circuit breaker quality evaluation is insufficient and the reliability is poor. The technical effect of improving the accuracy and reliability of circuit breaker quality evaluation is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application will be briefly introduced below. In the present application, flowcharts are used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously according to needs. Meanwhile, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0018] Figure 1 A flowchart of a quality evaluation method of an intelligent circuit breaker provided by an embodiment of the present application;

[0019] Figure 2 A flowchart of determining an initial quality coefficient of a circuit breaker in a quality evaluation method of an intelligent circuit breaker provided by an embodiment of the present application. DETAILED DESCRIPTION

[0020] The foregoing description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0021] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application in combination with the drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application.

[0022] In the following description, "some embodiments" are related to a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The term "first\second" is only to distinguish similar objects, and does not represent a specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art of the technology to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiments of the present application provide a quality evaluation method of an intelligent circuit breaker, such as Figure 1As shown, the method comprises:

[0024] In step S100, the circuit breaker detection system is connected to retrieve the factory detection data set of the target circuit breaker. The circuit breaker is connected to the detection system through a physical interface (such as RS-485, CAN bus, etc.) or a wireless communication method (such as Wi-Fi, Bluetooth, etc.), to ensure that the circuit breaker detection system can correctly identify and adapt to the target circuit breaker, and to ensure that the circuit breaker and the detection system comply with a unified communication protocol to correctly, stably and in real time transmit data. The circuit breaker detection system may include a hardware detection device and a data processing platform for real-time monitoring and recording of various parameters and states of the circuit breaker. The factory detection data set contains data of various performance tests undergone by the target circuit breaker before leaving the factory, such as performance detection of the target circuit breaker (including breaking and closing time, breaking and closing speed, contact pressure, etc.), environmental adaptability test (such as high and low temperature resistance test, combustion performance test, dielectric test, etc.), mechanical property test (such as mechanical life test, aging test, etc.), and other tests (such as corrosion detection, explosion-proof detection, flammability detection, etc.). Specifically, by connecting the circuit breaker detection system, the unique identification (such as the production serial number) of the target circuit breaker is input, and the system will retrieve and display the factory detection data set of the target circuit breaker. For example, the factory detection data set of a certain circuit breaker obtained from the circuit breaker detection system contains rated current 100A (standard value 100A), rated voltage 400V (standard value 400V), mechanical life 10000 times (standard value 10000 times), closing time 45ms (standard value ≤50ms), opening time 28ms (standard value ≤30ms), and contact resistance 93Ω (standard value ≤100Ω).

[0025] In step S200, the quality of the target circuit breaker is evaluated according to the factory detection data set, and the initial quality coefficient of the circuit breaker is determined. The quality evaluation according to the factory detection data set of the target circuit breaker specifically refers to identifying the key parameters that affect the quality of the target circuit breaker, which may include opening and closing performance (such as breaking and closing time, operating voltage), mechanical properties (such as mechanical life, mechanical operating force), dielectric properties (such as insulation strength), contact resistance, etc. Then, the identified parameter values of the target circuit breaker are compared with the evaluation standard to evaluate the quality performance of the circuit breaker in each parameter. Then, according to the importance of each parameter to the quality of the target circuit breaker, different weights are assigned, and finally the initial quality coefficient of the target circuit breaker is calculated by using the weighted average method. The initial quality coefficient is used to reflect the overall quality level of the circuit breaker at the time of leaving the factory.

[0026] In one possible implementation, as Figure 2As shown, step S200 further includes step S210 of establishing a circuit breaker quality evaluation channel, wherein the circuit breaker quality evaluation channel includes P circuit breaker quality evaluation models, P being a positive integer greater than 1. Specifically, the evaluation models are constructed based on machine learning algorithms (such as neural networks, decision trees, random forests, etc.), the models are automatically trained to learn the relationship between circuit breaker quality and various indicators, P circuit breaker quality evaluation models are trained, and then the P circuit breaker quality evaluation models are integrated to form a circuit breaker quality evaluation channel, thereby providing more comprehensive and accurate circuit breaker quality evaluation results.

[0027] Step S200 further includes step S220 of performing data cleaning according to the factory detection data set to generate a circuit breaker initial detection matrix. Data related to circuit breaker quality evaluation is extracted from the factory detection data set and cleaned, including steps of removing duplicate data, handling missing values, correcting erroneous data, etc., to ensure the accuracy and reliability of subsequent analysis. The cleaned data is arranged in matrix form, i.e., a circuit breaker initial detection matrix, wherein each row of the matrix represents the detection data of a circuit breaker, and each column represents a detection indicator or feature.

[0028] Step S200 further includes step S230 of inputting the circuit breaker initial detection matrix into the P circuit breaker quality evaluation models to obtain P circuit breaker initial quality evaluation coefficients. The circuit breaker initial detection matrix is input as input data into the P circuit breaker quality evaluation models. Each circuit breaker quality evaluation model analyzes and processes the input circuit breaker initial detection matrix to obtain an evaluation result reflecting the quality of the circuit breaker, and finally P circuit breaker initial quality evaluation coefficients are obtained, each representing the evaluation result of the same set of circuit breaker detection data by different models.

[0029] Step S200 further includes step S240 of performing proportion calculation based on P quality evaluation accuracy rate parameters corresponding to the P circuit breaker quality evaluation models to generate P quality evaluation output incentive coefficients. Each circuit breaker quality evaluation model corresponds to a parameter for evaluating its accuracy rate. This quality evaluation accuracy rate parameter is obtained through historical data and reflects the accuracy and reliability of the model in evaluating circuit breaker quality. Specifically, based on the quality evaluation accuracy rate parameter of each model, proportion calculation is performed to generate P quality evaluation output incentive coefficients, which are used for subsequent weighted calculation of the initial quality evaluation coefficients to reflect the reliability and importance of the evaluation results of different models.

[0030] The step S200 further comprises a step S250 of calculating the initial quality evaluation coefficient of the circuit breaker by weighting the P initial quality evaluation coefficients of the circuit breaker according to the P output excitation coefficients of the quality evaluation. Specifically, the initial quality evaluation coefficient of the circuit breaker is calculated by multiplying each initial quality evaluation coefficient of the model by the corresponding output excitation coefficient and then adding all the weighted evaluation coefficients. The initial quality evaluation coefficient of the circuit breaker thus obtained is a comprehensive evaluation of the quality of the target circuit breaker.

[0031] In step S300, the circuit breaker management module is connected to collect the use record information and the maintenance record information of the target circuit breaker to obtain M circuit breaker use record groups and Q circuit breaker maintenance record groups, where M and Q are positive integers. The circuit breaker management module is a system integrating data collection, processing, storage and analysis functions. The use record information and the maintenance record information of the target circuit breaker are obtained by connecting the circuit breaker management module through a communication interface. Specifically, during the use and operation of the circuit breaker, the circuit breaker management module collects the use record information of the circuit breaker in real time, which may include but is not limited to the number of switchings, the load current, the operation time, the ambient temperature, the fault record, etc. These information is sorted into M circuit breaker use record groups, each of which contains the use record information of the circuit breaker in a specific time period. M is a positive integer representing the number of use record groups collected, which depends on the frequency and time span of the collection. In addition to the use record information, the management module also collects the maintenance record information of the circuit breaker regularly. These maintenance record information records the maintenance and repair of the circuit breaker, including but not limited to the maintenance date, the maintenance content (such as replacement of parts, cleaning of contacts, etc.), the maintenance result, etc. These maintenance record information is sorted into Q circuit breaker maintenance record groups, each of which contains the detailed record of the circuit breaker in a specific maintenance event. Q is also a positive integer representing the number of maintenance record groups collected, which depends on the maintenance frequency and the record detail level of the circuit breaker.

[0032] Step S400, according to the circuit breaker quality reward and punishment adjustment formula, combining the M circuit breaker use record groups and the Q circuit breaker maintenance record groups, reward and punishment adjustment is carried out on the initial quality coefficient of the circuit breaker, and a first quality coefficient of the circuit breaker is generated. Wherein, the circuit breaker quality reward and punishment adjustment formula includes the initial quality coefficient of the target circuit breaker, a positive reward factor (calculated from the use record and the maintenance record) and a negative punishment factor (calculated from the use record and the maintenance record), and the reward and punishment adjustment of the initial quality coefficient of the circuit breaker specifically refers to that, by analyzing the switch times and load current information in the M use record groups, the use frequency and load condition of the target circuit breaker are evaluated, if the target circuit breaker can still operate stably under high frequency or high load, and no abnormal situation occurs, the quality coefficient of the target circuit breaker is rewarded positively, otherwise, if an abnormality or failure occurs, the quality coefficient of the target circuit breaker is punished negatively; if the target circuit breaker can still maintain good performance for a long time or in a high temperature environment, the quality coefficient of the target circuit breaker is rewarded positively, otherwise, the quality coefficient of the target circuit breaker is punished negatively. By analyzing the maintenance effect information in the Q maintenance record groups, the influence of different maintenance strategies on the quality of the target circuit breaker is evaluated, if the performance of the target circuit breaker is significantly improved after a certain maintenance, and the subsequent operation is stable, the quality coefficient of the target circuit breaker is rewarded positively, otherwise, if the maintenance effect is poor or leads to new problems, the quality coefficient of the target circuit breaker is punished negatively; regular and timely maintenance helps to maintain the good performance of the circuit breaker, and the corresponding quality coefficient is rewarded positively; while neglect or delayed maintenance will lead to the decrease of the quality coefficient, and the corresponding quality coefficient is punished negatively. The reward and punishment adjustment of the initial quality coefficient of the circuit breaker by combining the M circuit breaker use record groups and the Q circuit breaker maintenance record groups finally generates the first quality coefficient of the circuit breaker, which is a process of comprehensively evaluating the performance of the circuit breaker in actual operation, and more accurately reflects the actual quality level of the circuit breaker.

[0033] In a possible implementation, step S400 further includes step S410 of performing circuit breaker quality degradation influence analysis according to the M circuit breaker use record groups to determine M circuit breaker quality penalty coefficients. The M circuit breaker use record groups are analyzed, and according to the use record of each circuit breaker, including use time, use condition, load condition and the like, the influence of these factors on the quality of the circuit breaker is analyzed. The circuit breaker use record information indicates the conditions that cause the quality of the target circuit breaker to decrease in the use process, such as overload, overheating, frequent operation, high temperature environment, lack of maintenance and the like. Each identified influence factor is quantitatively evaluated to determine the specific influence degree of the influence factor on the quality of the circuit breaker. Based on the quantitative results of the influence factors, a quality penalty coefficient is calculated for each circuit breaker to reflect the degree of quality decrease of the circuit breaker due to various adverse factors in the use process. Finally, the quality penalty coefficients of the M circuit breakers are output.

[0034] Step S400 further includes step S420 of performing a circuit breaker quality enhancement incentive analysis based on the Q circuit breaker maintenance record sets to generate Q circuit breaker quality reward coefficients. Relevant data is collected from the Q circuit breaker maintenance record sets, including but not limited to maintenance time, maintenance content, performance comparison before and after maintenance, maintenance effect evaluation, etc. Each item of data in the maintenance record is analyzed to evaluate the specific influence of each maintenance activity on the quality of the circuit breaker, which can include performance parameter comparison, failure rate reduction, service life extension, etc. Based on the evaluation results, it is determined which maintenance activities have a significant effect on the improvement of the quality of the circuit breaker. Based on the evaluation results of the maintenance effect, a quality reward coefficient is calculated for each circuit breaker to reflect the degree of improvement in the quality of the circuit breaker due to maintenance activities. Finally, the quality reward coefficients of the Q circuit breakers are output.

[0035] Step S400 further includes step S430 of performing time-sensitive analysis on the M circuit breaker usage record sets and the Q circuit breaker maintenance record sets, respectively, to generate M quality penalty sensitivity factors and Q quality reward sensitivity factors. The time-sensitive analysis on the M circuit breaker usage record sets and the Q circuit breaker maintenance record sets specifically refers to considering the influence of time factors on quality penalty and quality reward. For example, recent usage or maintenance activities can have a greater influence on the quality of the circuit breaker, while distant records can have a smaller influence. Finally, M quality penalty sensitivity factors and Q quality reward sensitivity factors are generated to reflect the sensitivity of time to quality penalty and quality reward.

[0036] Step S400 further includes step S440 of inputting the M circuit breaker quality penalty coefficients, the Q circuit breaker quality reward coefficients, the M quality penalty sensitivity factors, the Q quality reward sensitivity factors, and the initial quality coefficient of the circuit breaker into the circuit breaker quality reward and penalty adjustment formula to obtain the first quality coefficient of the circuit breaker. The circuit breaker quality reward and penalty adjustment formula takes the M circuit breaker quality penalty coefficients, the Q circuit breaker quality reward coefficients, the M quality penalty sensitivity factors, the Q quality reward sensitivity factors, and the initial quality coefficient of the circuit breaker as input data, comprehensively considers the initial quality of the circuit breaker, quality degradation during use, quality improvement due to maintenance, and the influence of time on these factors, and thus obtains a more comprehensive and accurate circuit breaker quality evaluation result. Finally, the first quality coefficient of the circuit breaker is obtained.

[0037] Step S400 further includes step S450, and the circuit breaker quality reward and penalty adjustment formula is:

[0038] ;

[0039] wherein CQY1 represents the first quality coefficient of the circuit breaker, CQY0 represents the initial quality coefficient of the circuit breaker, QPX represents the M quality penalty coefficients, QPY represents the Q quality reward coefficients, QPS represents the M quality penalty sensitivity factors, QPS represents the Q quality reward sensitivity factors, and CQY0 represents the initial quality coefficient of the circuit breaker. mcharacterizing the mth circuit breaker quality penalty coefficient, PXW m characterizing a quality penalty sensitive factor corresponding to the mth circuit breaker quality penalty coefficient, QRZ q characterizing the qth circuit breaker quality reward coefficient, RZW q characterizing a quality reward sensitive factor corresponding to the qth circuit breaker quality reward coefficient, m, q are positive integers, 1≤m≤M, 1≤q≤Q.

[0040] In a possible implementation, the step S410 further includes a step S411 of loading, according to the circuit breaker management module, a sample circuit breaker use record set and a sample circuit breaker quality degradation influence coefficient record set. The sample circuit breaker use record set contains use records of a plurality of sample circuit breakers, including but not limited to use time, load condition, environmental condition, operation frequency and other factors that may affect quality degradation; and the sample circuit breaker quality degradation influence coefficient record set records quality degradation conditions of known sample circuit breakers under different use conditions, usually in the form of a coefficient or an index.

[0041] The step S410 further includes a step S412 of supervising training of a quality degradation influence learning model according to the sample circuit breaker use record set and the sample circuit breaker quality degradation influence coefficient record set, and obtaining a quality degradation influence prediction loss coefficient every predetermined number of times of training. A prediction model is constructed based on machine learning (such as a decision tree, a neural network, etc.), the quality degradation influence learning model is trained using the loaded sample circuit breaker use record set and the sample circuit breaker quality degradation influence coefficient record set, and the quality degradation conditions of the circuit breaker are learned and predicted from the use records. During the training process, a quality degradation influence prediction loss coefficient is calculated and output every predetermined number of times of training, for measuring the difference between the prediction result of the model and the actual quality degradation condition, that is, the prediction accuracy of the model.

[0042] The step S410 further includes a step S413 of judging whether the quality degradation influence prediction loss coefficient is less than a quality degradation influence prediction loss threshold. The quality degradation influence prediction loss coefficient is compared with a preset quality degradation influence prediction loss threshold, which is a critical value set according to business requirements and model performance, for judging whether the model has reached sufficient prediction accuracy.

[0043] The step S410 further includes a step S414 of generating a quality degradation influence analysis channel if the quality degradation influence prediction loss coefficient is less than the quality degradation influence prediction loss threshold. If the quality degradation influence prediction loss coefficient is less than the quality degradation influence prediction loss threshold, it indicates that the prediction accuracy of the model is already high enough, and a quality degradation influence analysis channel is generated, which can predict the quality degradation condition of a new circuit breaker based on the use record of the new circuit breaker.

[0044] Step S410 further includes step S415, based on the M circuit breaker usage record sets, obtaining the M circuit breaker quality penalty coefficients according to the quality degradation influence analysis channel. Based on the M circuit breaker usage record sets, the quality degradation of each circuit breaker is predicted using the generated quality degradation influence analysis channel, and according to the prediction result, a quality penalty coefficient is generated for each circuit breaker, which reflects the degree of quality degradation of the circuit breaker due to various factors in the use process.

[0045] In a possible implementation, step S430 further includes step S431, calling the time information corresponding to the M circuit breaker usage record sets to generate M usage time nodes. Key time information related to quality degradation is extracted from the usage record of each circuit breaker, such as start time, end time, important operation time, etc., and the time information of each circuit breaker is converted into specific M usage time nodes, which represent the key time points in the use process of the circuit breaker.

[0046] Step S430 further includes step S432, sorting according to the M usage time nodes to generate a circuit breaker usage time node table. All circuit breaker time nodes are sorted in time order to reflect the order and time span of circuit breaker use, and the sorted time nodes are arranged in table form, i.e. a circuit breaker usage time node table is generated, which is convenient for subsequent analysis and processing.

[0047] Step S430 further includes step S433, time-sensitive evaluation according to the circuit breaker usage time node table to obtain the M quality penalty sensitivity factors. Time-sensitive evaluation according to the circuit breaker usage time node table means that through analysis of the circuit breaker usage time node table, the sensitivity of time factors to circuit breaker quality degradation is evaluated, which usually includes considering the cumulative effect of time, the change of use frequency, the time dependence of environmental conditions, etc. According to the result of time-sensitive evaluation, a quality penalty sensitivity factor is generated for each circuit breaker, i.e. M quality penalty sensitivity factors are obtained, which reflect the degree of influence of time factors on circuit breaker quality degradation, i.e. the closer the time, the more frequent the use or the more severe the environmental conditions, the higher the quality penalty sensitivity factor may be.

[0048] Step S430 further comprises step S434, calling the time information corresponding to the Q circuit breaker maintenance record groups to generate Q maintenance time nodes. Extract the key time information related to quality improvement from the maintenance record of each circuit breaker, such as maintenance start time, maintenance end time, important maintenance operation time, etc., and convert the maintenance time information of each circuit breaker into specific time nodes, i.e. generate Q maintenance time nodes, which represent the key time points of the circuit breaker in the maintenance process.

[0049] Step S430 further comprises step S435, constructing a circuit breaker maintenance time node table according to the Q maintenance time nodes, and performing time-sensitive evaluation according to the circuit breaker maintenance time node table to generate the Q quality reward sensitive factors. Organize the maintenance time nodes of all circuit breakers into a table form and sort them in time sequence to construct a circuit breaker maintenance time node table. Time-sensitive evaluation according to the circuit breaker maintenance time node table means analyzing the maintenance time node table to assess the sensitivity of time factors to the quality improvement of the circuit breaker, such as considering the timeliness of maintenance, the durability of maintenance effect, the comparison of performance changes before and after maintenance, etc. According to the results of time-sensitive evaluation, a quality reward sensitive factor is generated for each circuit breaker, i.e. Q quality reward sensitive factors are generated, reflecting the influence degree of time factors on the quality improvement of the circuit breaker, i.e. the more timely the maintenance, the more significant the effect or the more obvious the performance improvement after maintenance, the higher the quality reward sensitive factor may be.

[0050] Step S500, determine whether the first quality coefficient of the circuit breaker is less than / equal to a predetermined circuit breaker quality threshold. The predetermined circuit breaker quality threshold is a factory standard value or range for a circuit breaker, which is used to measure whether the quality of the circuit breaker meets the standard. Compare the first quality coefficient of the circuit breaker with the predetermined quality threshold. If the first quality coefficient of the circuit breaker is greater than the predetermined circuit breaker quality threshold, it means that the current quality of the circuit breaker is better than the predetermined standard, and the quality is qualified. If the first quality coefficient of the circuit breaker is less than or equal to the predetermined circuit breaker quality threshold, it means that the current quality of the circuit breaker does not meet the predetermined standard, and the quality may have problems.

[0051] Step S600, if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, performing quality analysis on the target circuit breaker according to the circuit breaker detection system to determine a second quality coefficient of the circuit breaker. When the first quality coefficient of the circuit breaker does not reach the predetermined circuit breaker quality threshold, it means that the circuit breaker may have some problems in performance or reliability. In-depth quality analysis is performed using the circuit breaker detection system. Specifically, the test equipment in the circuit breaker detection system is used to test the key performance parameters of the circuit breaker, such as insulation resistance, voltage withstand, opening and closing performance, etc. The test results are compared with the performance parameters when the equipment is shipped, to evaluate whether the performance of the circuit breaker meets the standard, and accordingly determine the second quality coefficient of the circuit breaker.

[0052] In one possible implementation, step S600 further includes step S610, if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold, obtaining a real-time detection data set of the target circuit breaker according to the circuit breaker detection system. If the first quality coefficient of the circuit breaker is less than or equal to the predetermined circuit breaker quality threshold, the circuit breaker detection system is triggered to obtain a real-time detection data set of the target circuit breaker. For example, the real-time detection data set of a certain circuit breaker obtained from the circuit breaker detection system includes rated current 99A (standard value 100A), rated voltage 398V (standard value 400V), mechanical life 9500 times (standard value 10000 times), closing time 51ms (standard value ≤50ms), opening time 32ms (standard value ≤30ms), contact resistance 87Ω (standard value ≤100Ω).

[0053] Step S600 further includes step S620, data cleaning is performed according to the real-time detection data set to generate a circuit breaker real-time detection matrix. The real-time detection data set is cleaned to remove noise, outliers, etc. contained therein, so as to improve the data quality and ensure the accuracy and reliability of subsequent analysis. The cleaned real-time detection data is arranged in the form of a matrix, i.e., a circuit breaker real-time detection matrix. Each row of the matrix represents the detection data at a time point, and each column represents a type of detection parameter.

[0054] Step S600 further includes step S630, inputting the circuit breaker real-time detection matrix into the P circuit breaker quality evaluation models to obtain P circuit breaker real-time quality evaluation coefficients. The circuit breaker real-time detection matrix is input into the P quality evaluation models. Each circuit breaker quality evaluation model receives the circuit breaker real-time detection matrix as input and outputs a corresponding real-time quality evaluation coefficient, i.e., obtains P circuit breaker real-time quality evaluation coefficients, which reflect the quality state of the circuit breaker from different angles and dimensions.

[0055] The step S600 further includes a step S640 of calculating the real-time quality evaluation coefficients of the P circuit breakers according to the P quality evaluation output excitation coefficients to generate second quality coefficients of the circuit breakers. The P quality evaluation output excitation coefficients are used to calculate the real-time quality evaluation coefficients of the P circuit breakers, and finally the second quality coefficients of the circuit breakers are generated. The second quality coefficients of the circuit breakers comprehensively reflect the evaluation results of multiple models and multiple dimensions, and more comprehensively and accurately reflect the quality state of the circuit breakers.

[0056] The step S700 includes a step S720 of performing deviation analysis on the first quality standard coefficient and the second quality standard coefficient of the circuit breaker to obtain a quality standard deviation of the circuit breaker. The deviation analysis refers to evaluating the difference between two or more data sets. Specifically, the deviation between the first quality standard coefficient and the second quality standard coefficient, i.e., the quality standard deviation of the circuit breaker, is calculated. The deviation reflects the consistency of the quality evaluation results of the circuit breaker in two evaluations.

[0057] In a possible implementation, the step S700 further includes a step S710 of performing standardization processing on the first quality coefficient and the second quality coefficient of the circuit breaker to generate a first quality standard coefficient and a second quality standard coefficient of the circuit breaker. The standardization processing is a data preprocessing method, which is used to convert data of different dimensions or ranges into a unified form that can be directly compared or calculated. The first quality coefficient and the second quality coefficient of the circuit breaker are standardized to eliminate the dimension influence of the original data, and the first quality standard coefficient and the second quality standard coefficient of the circuit breaker are obtained, which facilitates subsequent comparison and calculation of the first quality standard coefficient and the second quality standard coefficient of the circuit breaker.

[0058] The step S700 further includes a step S720 of performing deviation analysis on the first quality standard coefficient and the second quality standard coefficient of the circuit breaker to obtain a quality standard deviation of the circuit breaker. The deviation analysis refers to evaluating the difference between two or more data sets. Specifically, the deviation between the first quality standard coefficient and the second quality standard coefficient, i.e., the quality standard deviation of the circuit breaker, is calculated. The deviation reflects the consistency of the quality evaluation results of the circuit breaker in two evaluations.

[0059] Step S700 further includes step S730 of judging whether the circuit breaker quality standard deviation is less than a circuit breaker quality standard deviation threshold value. The calculated circuit breaker quality standard deviation is compared with a preset circuit breaker quality standard deviation threshold value, which is set according to the circuit breaker standard, to judge whether the circuit breaker quality standard deviation is within an acceptable range, i.e., whether the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold value. If the deviation is less than the threshold value, it indicates that the two evaluation results are relatively consistent, and the quality status of the circuit breaker is relatively stable.

[0060] Step S700 further includes step S740 of, if the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold value, performing mean calculation according to the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient to generate the circuit breaker third quality coefficient. If the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold value, it indicates that the two quality evaluation results are close, and thus the comparison is reliable. Therefore, the comprehensive quality coefficient of the circuit breaker is further calculated. Specifically, the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient are averaged to obtain the circuit breaker third quality coefficient. The circuit breaker third quality coefficient comprehensively reflects the quality status of the circuit breaker and is more comprehensive and accurate.

[0061] In a possible implementation, step S730 further includes step S731 of, if the circuit breaker quality standard deviation is greater than / equal to the circuit breaker quality standard deviation threshold value, performing proportion calculation according to the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient to obtain a first quality proportion coefficient and a second quality proportion coefficient. If the circuit breaker quality standard deviation is greater than or equal to the circuit breaker quality standard deviation threshold value, the first quality proportion coefficient and the second quality proportion coefficient are calculated according to the proportion of the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient.

[0062] Step S730 further includes step S732 of performing weighted calculation on the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient according to the first quality proportion coefficient and the second quality proportion coefficient to obtain the circuit breaker third quality coefficient. The first quality proportion coefficient and the second quality proportion coefficient are respectively used as the weights of the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient for weighted calculation, so as to obtain a comprehensive circuit breaker quality evaluation result, i.e., the circuit breaker third quality coefficient, which is more comprehensive and accurate than the circuit breaker first quality coefficient and the circuit breaker second quality coefficient.

[0063] The foregoing DETAILED DESCRIPTION, including the above section titled "Detailed Description," is not to be taken as limiting the scope of the application. Various modifications, combinations, and equivalents can be apparent to those skilled in the art and can be made once the nature of the application is understood. Any modification, combination, or equivalent, which falls within the principles and the scope of the present application, is intended to be included in the present application. In some instances, the actions or steps can be performed in different order from those described herein, and still achieve desirable results. Additionally, the process depicted in the figures can not necessarily require the particular order shown or sequential order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

Claims

1. A method of quality assessment of an intelligent circuit breaker, characterized in that, The method comprises: connecting a circuit breaker detection system to retrieve factory detection data set of a target circuit breaker; conducting quality evaluation on the target circuit breaker according to the factory detection data set to determine initial quality coefficient of the circuit breaker; connecting a circuit breaker management module to collect usage record information and maintenance record information of the target circuit breaker to obtain M circuit breaker usage record groups and Q circuit breaker maintenance record groups, wherein M and Q are positive integers; adjusting the initial quality coefficient of the circuit breaker according to a circuit breaker quality reward and punishment adjustment formula combined with the M circuit breaker usage record groups and the Q circuit breaker maintenance record groups to generate first quality coefficient of the circuit breaker; determining whether the first quality coefficient of the circuit breaker is less than / equal to a predetermined circuit breaker quality threshold value; if the first quality coefficient of the circuit breaker is less than / equal to the predetermined circuit breaker quality threshold value, conducting quality analysis on the target circuit breaker according to the circuit breaker detection system to determine second quality coefficient of the circuit breaker; correcting the first quality coefficient of the circuit breaker according to the second quality coefficient of the circuit breaker to obtain third quality coefficient of the circuit breaker; wherein the quality evaluation on the target circuit breaker according to the factory detection data set to determine the initial quality coefficient of the circuit breaker comprises: building a circuit breaker quality evaluation channel, wherein the circuit breaker quality evaluation channel comprises P circuit breaker quality evaluation models, and P is a positive integer greater than 1; performing data cleaning according to the factory detection data set to generate an initial detection matrix of the circuit breaker; inputting the initial detection matrix of the circuit breaker into the P circuit breaker quality evaluation models to obtain P initial quality evaluation coefficients of the circuit breaker; performing proportion calculation based on P quality evaluation accuracy parameters corresponding to the P circuit breaker quality evaluation models to generate P quality evaluation output incentive coefficients; performing weighted calculation on the P initial quality evaluation coefficients of the circuit breaker according to the P quality evaluation output incentive coefficients to generate the initial quality coefficient of the circuit breaker; wherein the adjusting of the initial quality coefficient of the circuit breaker according to the circuit breaker quality reward and punishment adjustment formula combined with the M circuit breaker usage record groups and the Q circuit breaker maintenance record groups to generate the first quality coefficient of the circuit breaker comprises: conducting circuit breaker quality attenuation influence analysis according to the M circuit breaker usage record groups to determine M circuit breaker quality penalty coefficients; conducting circuit breaker quality enhancement incentive analysis according to the Q circuit breaker maintenance record groups to generate Q circuit breaker quality reward coefficients; respectively performing time-sensitive analysis on the M circuit breaker usage record groups and the Q circuit breaker maintenance record groups to generate M quality penalty sensitive factors and Q quality reward sensitive factors; inputting the M circuit breaker quality penalty coefficients, the Q circuit breaker quality reward coefficients, the M quality penalty sensitive factors, the Q quality reward sensitive factors and the initial quality coefficient of the circuit breaker into the circuit breaker quality reward and punishment adjustment formula to obtain the first quality coefficient of the circuit breaker; wherein the circuit breaker quality reward and punishment adjustment formula is: ； Wherein, CQY1 represents the first quality coefficient of the circuit breaker, CQY0 represents the initial quality coefficient of the circuit breaker, QPX m represents the mth circuit breaker quality penalty coefficient, PXW m represents the quality penalty sensitive factor corresponding to the mth circuit breaker quality penalty coefficient, QRZ q represents the qth circuit breaker quality reward coefficient, RZW q represents the quality reward sensitive factor corresponding to the qth circuit breaker quality reward coefficient, m, q are positive integers, 1≤m≤M, 1≤q≤Q; The M circuit breaker usage record groups are used to analyze the quality degradation influence of the circuit breakers, and M circuit breaker quality penalty coefficients are determined, including: According to the circuit breaker management module, load a sample circuit breaker usage record set and a sample circuit breaker quality degradation influence coefficient record set; According to the sample circuit breaker usage record set and the sample circuit breaker quality degradation influence coefficient record set, supervise the training of the quality degradation influence learning model, and obtain a quality degradation influence prediction loss coefficient every predetermined number of times of training; Determine whether the quality degradation influence prediction loss coefficient is less than a quality degradation influence prediction loss threshold value; If the quality degradation influence prediction loss coefficient is less than the quality degradation influence prediction loss threshold value, generate a quality degradation influence analysis channel; Based on the M circuit breaker usage record groups, the M circuit breaker quality penalty coefficients are obtained according to the quality degradation influence analysis channel.

2. The method of claim 1, wherein the quality of the intelligent circuit breaker is evaluated based on the at least one of the plurality of parameters. The M circuit breaker usage record groups and the Q circuit breaker maintenance record groups are respectively subjected to time-sensitive analysis, and M quality penalty sensitive factors and Q quality reward sensitive factors are generated, including: Retrieve the time information corresponding to the M circuit breaker usage record groups to generate M usage time nodes; According to the M usage time nodes, generate a circuit breaker usage time node table; According to the circuit breaker usage time node table, perform time-sensitive evaluation to obtain the M quality penalty sensitive factors; Retrieve the time information corresponding to the Q circuit breaker maintenance record groups to generate Q maintenance time nodes; According to the Q maintenance time nodes, construct a circuit breaker maintenance time node table, and according to the circuit breaker maintenance time node table, perform time-sensitive evaluation to generate the Q quality reward sensitive factors.

3. The method of claim 1, wherein the quality of the intelligent circuit breaker is evaluated based on the at least one of the plurality of parameters. If the first circuit breaker quality coefficient is less than / equal to the predetermined circuit breaker quality threshold value, perform quality analysis on the target circuit breaker according to the circuit breaker detection system to determine a second circuit breaker quality coefficient, including: If the first circuit breaker quality coefficient is less than / equal to the predetermined circuit breaker quality threshold value, obtain a real-time detection data set of the target circuit breaker according to the circuit breaker detection system; According to the real-time detection data set, perform data cleaning to generate a circuit breaker real-time detection matrix; Input the circuit breaker real-time detection matrix into the P circuit breaker quality evaluation models to obtain P circuit breaker real-time quality evaluation coefficients; According to the P quality evaluation output excitation coefficients, perform weighted calculation on the P circuit breaker real-time quality evaluation coefficients to generate the second circuit breaker quality coefficient.

4. The method of claim 1, wherein the quality of the intelligent circuit breaker is evaluated based on the at least one of the plurality of parameters. According to the second circuit breaker quality coefficient, correct the first circuit breaker quality coefficient to obtain a third circuit breaker quality coefficient, including: According to the first circuit breaker quality coefficient and the second circuit breaker quality coefficient, perform standardization processing to generate a first circuit breaker quality standard coefficient and a second circuit breaker quality standard coefficient; According to the first circuit breaker quality standard coefficient and the second circuit breaker quality standard coefficient, perform bias analysis to obtain a circuit breaker quality standard deviation; Determine whether the circuit breaker quality standard deviation is less than a circuit breaker quality standard deviation threshold value; If the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold value, mean calculation is performed according to the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient to generate the circuit breaker third quality coefficient.

5. The method of claim 4, wherein the quality of the intelligent circuit breaker is evaluated based on the comparison of the first and second values of the at least one parameter. The judgment of whether the circuit breaker quality standard deviation is less than the circuit breaker quality standard deviation threshold value comprises: If the circuit breaker quality standard deviation is greater than / equal to the circuit breaker quality standard deviation threshold value, proportion calculation is performed according to the circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient to obtain a first quality proportion coefficient and a second quality proportion coefficient; The circuit breaker first quality standard coefficient and the circuit breaker second quality standard coefficient are weighted calculated according to the first quality proportion coefficient and the second quality proportion coefficient to obtain the circuit breaker third quality coefficient.

Images