A capacitor state evaluation method and system based on an agent and related equipment

Abstract

The application provides a capacitor state evaluation method and system based on an agent, and related equipment, and relates to the technical field of capacitor evaluation. The method comprises the following steps: collecting capacitor data by driving a data collection device by a single agent, and constructing a supervised learning database based on the capacitor data; constructing a group agent and a corresponding group memory bank; driving the single agent to perform state evaluation on the capacitor according to a preset state evaluation task, generating an initial state evaluation method, and storing the initial state evaluation method in a single memory bank corresponding to the single agent; iteratively screening the initial state evaluation method based on a preset index, and updating the group memory bank based on the screened result; outputting a final state evaluation method through the group memory bank based on the updated result, and deleting the initial state evaluation method of the single memory bank.

Description

Technical Field

[0001] This application relates to the field of capacitor evaluation technology, and in particular to a capacitor state evaluation method, system and related equipment based on intelligent agents. Background Technology

[0002] With the transformation of the national energy landscape in recent years, clean and low-carbon energy has become the core of current development, and flexible DC transmission technology, with its unique advantages, plays a significant role in transmitting clean energy. As one of the largest components in the converter valve of flexible DC transmission, the capacitor's operating state directly affects the reliability of the equipment. Therefore, it is necessary to conduct state assessments of the capacitor to understand its condition and ensure stable system operation. Currently, many scholars and capacitor manufacturers have proposed several state assessment methods based on their research or experience. However, these methods have the following problems: 1. Experience-based state assessment methods often require remodeling after changes in operating conditions, making them unsuitable for uniform application; 2. State assessment methods based on mathematical or physical models can generally only evaluate a limited number of capacitor state parameters, failing to provide a comprehensive assessment of the capacitor's operating state; 3. Existing data-driven assessment methods only address the state assessment of capacitors under a single operating model, resulting in significant limitations. Summary of the Invention

[0003] The main objective of this application is to propose a capacitor state evaluation method, system, and related equipment based on intelligent agents, aiming to solve the problems in the background art.

[0004] To achieve the above objectives, one aspect of this application proposes a capacitor state evaluation method based on an intelligent agent, the method comprising: Capacitor data is collected by a data acquisition device driven by a single intelligent agent, and a supervised learning database is constructed based on the capacitor data. Construct a swarm intelligence agent and a swarm memory corresponding to the swarm intelligence agent; According to the preset state assessment task, a single intelligent agent is driven to perform state assessment on the capacitor, generate an initial state assessment method, and store the initial state assessment method in the single agent's corresponding single-entity memory bank. Based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results. Based on the updated results, the final state evaluation method is output through the group memory, and the initial state evaluation method of the individual memory is deleted.

[0005] In some embodiments, the iterative filtering of the initial state evaluation method based on preset indicators, and the updating of the group memory based on the filtered initial state evaluation method, specifically includes: Construct a voting agent, wherein the voting agent selects an initial state evaluation method from the single-entity memory bank; Based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results. When the final state evaluation method of the group memory reaches the maximum number of memory entries generated, the iterative screening stops.

[0006] In some embodiments, the method for driving a single intelligent agent to perform state evaluation on the capacitor and generate an initial state evaluation method specifically includes: Each individual agent performs a code generation operation on the supervised learning database according to the state evaluation task, generating code corresponding to the state evaluation function. The code is used to characterize the initial state evaluation method proposed by the individual agent. After the code is run on the local computing platform, the state evaluation value corresponding to the state evaluation task is output.

[0007] In some embodiments, after the code is sent to the local computing platform for execution and the state evaluation value corresponding to the state evaluation task is output, the method further includes: Calculate the deviation between the assessed state value and the actual state value; Calculate the standard deviation corresponding to the deviation value, and calculate the weighted average standard deviation of the initial state evaluation method generated by each individual agent based on the standard deviation; The weighted average standard deviation is stored in the individual memory corresponding to the single agent.

[0008] In some embodiments, the iterative screening of the initial state evaluation method based on preset indicators, and the updating of the group memory based on the screening results, specifically includes: Calculate the comprehensive evaluation index of the initial state evaluation method, wherein the comprehensive evaluation index includes an error penalty term and an algorithm complexity penalty term; The voting agent ranks all initial state evaluation methods based on the comparison results between the comprehensive evaluation index and the preset index. The voting-based intelligent agent selects an initial state evaluation method based on the ranking results, with a preset threshold number of agents. The group memory is updated based on the initial state evaluation method.

[0009] In some embodiments, the single agent includes a state evaluation function, the expression of which is: ; in For state evaluation function, For input parameters, For input parameters, This is a transpose.

[0010] In some embodiments, the expression for the weighted average standard deviation is: ; in The weighted average standard deviation The number of output parameters, For the first The weighting coefficients of each output parameter. For the first The standard deviation of each output parameter.

[0011] To achieve the above objectives, another aspect of this application proposes a capacitor state evaluation system based on an intelligent agent, the system comprising: A database construction module is used to collect capacitor data through a data acquisition device driven by a single intelligent agent, and to construct a supervised learning database based on the capacitor data. A swarm intelligence agent construction module is used to construct swarm intelligence agents and the swarm memory corresponding to the swarm intelligence agents; The initial state evaluation module is used to drive a single agent to evaluate the state of the capacitor according to a preset state evaluation task, generate an initial state evaluation method, and store the initial state evaluation method in the single agent's corresponding single-entity memory. The database update module is used to iteratively filter the initial state evaluation method based on preset indicators, and update the group memory based on the filtered results. The final state evaluation module is used to output the final state evaluation method through the group memory based on the updated results, and to delete the initial state evaluation method of the individual memory.

[0012] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0013] To achieve the above objectives, another aspect of the embodiments of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0014] The embodiments of this application include at least the following beneficial effects: This application provides a capacitor state assessment method, system, and related equipment based on intelligent agents. This application achieves fully automated operation and leverages the advantages of swarm intelligence, covering the entire process of data acquisition, processing, model prediction, and result output. During execution, everything is automatically completed by the application programming interface, requiring no additional settings or confirmation operations. Once the entire system is deployed, it can efficiently output a state assessment model, and it can automatically output different adaptive methods for different operating conditions, demonstrating strong generalization ability.

[0015] Furthermore, the database constructed using this method can evaluate various capacitor operating parameters and comprehensively reflect the operating status of the capacitor. Attached Figure Description

[0016] Figure 1 A flowchart of the agent-based capacitor state evaluation method provided in this application embodiment; Figure 2 A block diagram of an agent-based capacitor state assessment system provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0018] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0019] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] refer to Figure 1 As shown, one embodiment of this application proposes a capacitor state evaluation method based on an intelligent agent, including: S101: Capacitor data is collected by a data acquisition device driven by a single intelligent agent, and a supervised learning database is built based on the capacitor data; S102: Construct swarm intelligence agents and the corresponding swarm memory library for swarm intelligence agents; S103: Based on the preset state assessment task, drive a single agent to assess the state of the capacitor, generate an initial state assessment method, and store the initial state assessment method in the single agent's corresponding memory bank. S104: Based on preset indicators, iteratively filter the initial state evaluation methods and update the group memory based on the filtered results; S105: Based on the updated results, output the final state evaluation method through the group memory and delete the initial state evaluation method of the individual memory.

[0022] Steps S101-S105 of this application, with single intelligent agents and swarm intelligent agents as the core, realize fully automated operation and give full play to the advantages of swarm intelligence, covering the entire process of data collection, processing, model prediction and result output.

[0023] The model search and execution process is completed automatically by the application programming interface without the need for additional settings or confirmation operations. Once the entire method is deployed, it can efficiently output a state evaluation model. It can also automatically output different adaptive methods for different working conditions, and has strong generalization ability.

[0024] Furthermore, the initial database constructed using this method can evaluate various capacitor operating state parameters, comprehensively reflecting the operating state of the capacitor.

[0025] Specifically, the data acquisition equipment in step S101 includes at least a voltage and current acquisition device, an impedance analyzer (for measuring capacitor impedance), a dielectric spectrometer (for measuring the correlation between capacitor capacitance, frequency, and dielectric loss), a system source meter (for measuring leakage current and leakage conductance, etc.), and a temperature transceiver (for receiving thermocouple or infrared temperature signals). Communication between these devices and the intelligent agents is accomplished through application programming interfaces (APIs) provided by the device manufacturers. Each intelligent agent communicates with the others through the API to exchange information and achieve collaborative cooperation.

[0026] Specifically, in step S102, the swarm intelligence can assign different specialized tasks to different intelligence agents to cooperate in completing the capacitor data acquisition task. The swarm intelligence consists of a series of transformer architecture large language models, which should include the mainstream generative models at present.

[0027] The swarm intelligence agent automatically collects the time series data required for capacitor state assessment, such as capacitance, frequency, and voltage, based on the data prediction accuracy. After collection, it automatically saves the data to form an initial database to facilitate subsequent two-stage iterative search of capacitor state. Different individual intelligence agents drive different data acquisition devices.

[0028] Furthermore, in some embodiments, given a capacitor state evaluation task, a supervised learning database, a swarm agent decision direction, and a single agent search strategy, the swarm agent can provide a series of evaluation functions by updating the swarm memory.

[0029] Specifically, in step S103: the single intelligent agent is driven to perform a state assessment of the capacitor, generating an initial state assessment method, which specifically includes: Each individual agent performs code generation operations on the supervised learning database according to the state evaluation task, generating code corresponding to the state evaluation function. The code is used to represent the initial state evaluation method proposed by the individual agent. After the code is run on the local computing platform, the corresponding state evaluation value of the state evaluation task is output. Calculate the deviation between the assessed state value and the actual state value; Calculate the standard deviation corresponding to the deviation value, and calculate the weighted average standard deviation of the initial state evaluation method generated by each individual agent based on the standard deviation; The weighted average standard deviation is stored in the individual memory corresponding to each agent.

[0030] Each agent includes a state evaluation function, the expression of which is: ; in For state evaluation function, For input parameters, For input parameters, This is a transpose.

[0031] The expression for the weighted average standard deviation is: ; in The weighted average standard deviation The number of output parameters, For the first The weighting coefficients of each output parameter. For the first The standard deviation of each output parameter.

[0032] Specifically, in this application, the input parameters include voltage, current, temperature, etc., and the output parameters include capacitance, loss, and impedance.

[0033] Specifically, in step S104, the initial state evaluation method is iteratively screened based on preset indicators, and the group memory is updated based on the screened initial state evaluation method. This includes: Construct a voting agent, and select an initial state evaluation method from a single-entity memory bank for the voting agent; Based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results. When the final state evaluation method of the swarm memory reaches the maximum number of memory entries generated, the iterative screening stops.

[0034] When the maximum number of memory entries generated is reached, this application updates the search strategy of a single agent to an enhanced or optimized single method, while keeping the number of memory entries of the swarm agents unchanged, and repeatedly generates the optimal state evaluation method under the current computing resources.

[0035] After obtaining the initial swarm memory, the search strategies of individual agents are updated to optimize the parameters or adjust the steps of existing methods to reduce prediction errors. Once all individual agents have completed their generation tasks, a voting process is used to select methods, keeping the total number of methods in the swarm memory constant. Better methods replace existing methods to complete one swarm generation task. This process iterates until the maximum number of iterations is reached or the method error in the swarm memory reaches the expected value. The method with the smallest error is selected from the group memory as the optimal state assessment method for the capacitor.

[0036] Furthermore, based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results, including: Calculate the comprehensive evaluation index of the initial state evaluation method. The comprehensive evaluation index includes an error penalty term and an algorithm complexity penalty term. The voting agent ranks all initial state evaluation methods based on the comparison results of comprehensive evaluation indicators and preset indicators. The voting-based intelligent agent selects an initial state evaluation method based on the ranking results, with a preset threshold number of agents. The group memory is updated based on the initial state evaluation method.

[0037] The preset indicators include preset thresholds for error penalty items and algorithm complexity penalty items.

[0038] The following examples illustrate this application: This embodiment assumes that the capacitance C and equivalent series resistance (ESR) of the capacitor in the converter valve are evaluated, and the data acquisition equipment is a voltage and current sensor and an impedance analyzer.

[0039] First, the single agent collects the voltage and current I of the capacitor during its operation, as well as the capacitance value and equivalent series resistance (ESR) measured by the impedance analyzer controlled by the single agent, based on the existing knowledge it has acquired during training (reinforcement learning of the local database according to the product manual and existing APIs).

[0040] Information transmission between individual agents is accomplished through Python's data communication interface, and the execution process follows a specific order. The voltage and current during capacitor operation are fluctuating values ​​over a period of time. Current and voltage can be used as inputs, while the ideal capacitance and equivalent series resistance for that capacitor state are used as outputs, thus forming a single data point. The entire swarm of agents collects multiple data points from different capacitor operation processes to form an initial database. During the data collection process, the control of all devices and the transfer of samples are handled by different agents paired with different interface devices.

[0041] The current task is to predict the ideal capacitance and equivalent series resistance of a capacitor simultaneously. The criterion is the weighted average standard deviation of the two (if the ideal capacitance is on the order of μF and the equivalent series resistance fluctuates on the order of mΩ, then the error weighting coefficients can be set to 1μF and 1mΩ respectively, representing that the two are of equal importance). Furthermore, prompts can be set to require generating as many different methods as possible based on the weighted average standard deviation obtained from the feedback, and the weighted average standard deviation must be output.

[0042] The maximum capacity of the swarm memory is set to 30 methods, meaning that iteration stops when 30 different or similar methods are added to the swarm agents. The number of methods integrated in each voting agent is 3, and the number of individual agents is 10. The current search strategy is to try as many different methods as possible and to provide the prediction error.

[0043] After a single agent generates an algorithm code, the system automatically calls the software to execute it. For example, if the requirement is to generate Python code, the system can directly call the Python API to execute the program and return the result. The algorithm code and the error of the returned result are then concatenated into a method class and sent to the voting system to select the agent.

[0044] The voting agent calculates a comprehensive evaluation index for the methods proposed by 10 different individual agents. Based on the comparison between the comprehensive evaluation index and the preset index, the voting agent ranks all initial state evaluation methods. After comprehensively considering the algorithm complexity penalty term and the error penalty term, the voting agent selects 3 methods that are as different as possible to enter the swarm memory.

[0045] At this point, the individual agent's memory is deleted and updated to the group memory. Because the individual agent has a context window memory function throughout the iteration process, it can know the execution status and historical methods of the group agents, which facilitates its optimization of strategies in new generation. This iteration continues until the group memory size reaches 30.

[0046] At this point, the search strategy for a single agent is updated by optimizing the parameters of a single method based on historical memory or adjusting a small number of execution steps to optimize the computational results. The task of the single agent and the supervised learning library remain unchanged, thus completing the generation task. After all single agents are generated, they are also sent to a voting selection process. The best three methods are selected to replace the three worst methods in the swarm memory, which is then updated and the single agent memory is replaced. This process is iterated until the maximum number of updates is reached.

[0047] The state assessment method with the smallest weighted average standard deviation in the current group memory is selected as the final capacitor state assessment method, which can calculate the ideal capacitance and equivalent series resistance based on the input current and voltage data.

[0048] refer to Figure 2 As shown, another aspect of this application embodiment also proposes a capacitor state evaluation system based on an intelligent agent, the system comprising: The database construction module is used to collect capacitor data through a data acquisition device driven by a single intelligent agent, and to build a supervised learning database based on the capacitor data. The swarm intelligence agent construction module is used to build swarm intelligence agents and the corresponding swarm memory library for each swarm intelligence agent. The initial state evaluation module is used to drive a single agent to evaluate the state of the capacitor according to a preset state evaluation task, generate an initial state evaluation method, and store the initial state evaluation method in the single agent's corresponding memory bank. The database update module is used to iteratively filter the initial state evaluation method based on preset indicators, and update the group memory based on the filtered results. The final state evaluation module is used to output the final state evaluation method through the group memory based on the updated results, and to delete the initial state evaluation method of the individual memory.

[0049] Another aspect of this application embodiment also proposes an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0050] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0051] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0052] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0053] The methods provided in this application relate to the field of information technology. The methods provided in this application can be applied to terminals, servers, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the method, but is not limited to the above forms.

[0054] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0055] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0056] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0057] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0058] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0059] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0060] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0061] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0062] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0063] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0064] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0065] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A capacitor state evaluation method based on an intelligent agent, characterized in that, The method includes: Capacitor data is collected by a data acquisition device driven by a single intelligent agent, and a supervised learning database is constructed based on the capacitor data. Construct a swarm intelligence agent and a swarm memory corresponding to the swarm intelligence agent; According to the preset state assessment task, a single intelligent agent is driven to perform state assessment on the capacitor, generate an initial state assessment method, and store the initial state assessment method in the single agent's corresponding single-entity memory bank. Based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results. Based on the updated results, the final state evaluation method is output through the group memory, and the initial state evaluation method of the individual memory is deleted.

2. The capacitor state evaluation method based on an intelligent agent according to claim 1, characterized in that, The step of iteratively filtering the initial state evaluation method based on preset indicators, and updating the group memory based on the filtered initial state evaluation methods, specifically includes: Construct a voting agent, wherein the voting agent selects an initial state evaluation method from the single-entity memory bank; Based on preset indicators, the initial state evaluation method is iteratively screened, and the group memory is updated based on the screening results. When the final state evaluation method of the group memory reaches the maximum number of memory entries generated, the iterative screening stops.

3. The capacitor state evaluation method based on intelligent agents according to claim 2, characterized in that, The method for driving a single intelligent agent to perform state evaluation on the capacitor and generate an initial state evaluation method specifically includes: Each individual agent performs a code generation operation on the supervised learning database according to the state evaluation task, generating code corresponding to the state evaluation function. The code is used to characterize the initial state evaluation method proposed by the individual agent. After the code is run on the local computing platform, the state evaluation value corresponding to the state evaluation task is output.

4. The capacitor state evaluation method based on intelligent agents according to claim 3, characterized in that, After the code is sent to the local computing platform for execution and the state evaluation value corresponding to the state evaluation task is output, the process further includes: Calculate the deviation between the assessed state value and the actual state value; Calculate the standard deviation corresponding to the deviation value, and calculate the weighted average standard deviation of the initial state evaluation method generated by each individual agent based on the standard deviation; The weighted average standard deviation is stored in the individual memory corresponding to the single agent.

5. The capacitor state evaluation method based on intelligent agents according to claim 4, characterized in that, The step of iteratively filtering the initial state evaluation method based on preset indicators and updating the group memory based on the filtered results specifically includes: Calculate the comprehensive evaluation index of the initial state evaluation method, wherein the comprehensive evaluation index includes an error penalty term and an algorithm complexity penalty term; The voting agent ranks all initial state evaluation methods based on the comparison results between the comprehensive evaluation index and the preset index. The voting-based intelligent agent selects an initial state evaluation method based on the ranking results, with a preset threshold number of agents. The group memory is updated based on the initial state evaluation method.

6. The capacitor state evaluation method based on an intelligent agent according to claim 3, characterized in that, The single agent includes a state evaluation function, the expression of which is: ； in For state evaluation function, For input parameters, For input parameters, This is a transpose.

7. The capacitor state evaluation method based on intelligent agents according to claim 4, characterized in that, The expression for the weighted average standard deviation is: ； in The weighted average standard deviation The number of output parameters, For the first The weighting coefficients of each output parameter. For the first The standard deviation of each output parameter.

8. A capacitor state assessment system based on an intelligent agent, characterized in that, The system includes: A database construction module is used to collect capacitor data through a data acquisition device driven by a single intelligent agent, and to construct a supervised learning database based on the capacitor data. A swarm intelligence agent construction module is used to construct swarm intelligence agents and the swarm memory corresponding to the swarm intelligence agents; The initial state evaluation module is used to drive a single agent to evaluate the state of the capacitor according to a preset state evaluation task, generate an initial state evaluation method, and store the initial state evaluation method in the single agent's corresponding single-entity memory. The database update module is used to iteratively filter the initial state evaluation method based on preset indicators, and update the group memory based on the filtered results. The final state evaluation module is used to output the final state evaluation method through the group memory based on the updated results, and to delete the initial state evaluation method of the individual memory.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images