Charging service platform performance evaluation method and system based on knowledge graph

Abstract

The invention discloses a charging service platform performance evaluation method and system based on a knowledge graph, and relates to the field of charging stations, and the method comprises the steps: building a knowledge model used for a charging service platform based on a standardized data set through employing a knowledge graph technology; extracting an implicit mode in the charging service from the knowledge model by utilizing a machine learning and data mining algorithm; constructing a charging service quality evaluation model based on the implicit mode in the charging service; performing real-time performance monitoring and evaluation on the charging service platform by using the charging service quality evaluation model, and generating a charging quality evaluation result; identifying a to-be-optimized charging performance index from the charging quality evaluation result; and adjusting the control parameters of the to-be-optimized charging performance index based on an optimization algorithm. Various kinds of information of the charging service platform are represented and inquired in a structured mode, key factors influencing the service quality can be identified, and a dynamic performance evaluation result is provided.

Description

Technical Field

[0001] The present invention relates to the field of charging stations, and more particularly, to a method and system for evaluating the performance of a charging service platform based on a knowledge graph. Background Art

[0002] A charging service platform is a platform for managing and coordinating electric vehicle charging infrastructure. The software and hardware solutions provided by this platform enable electric vehicle owners to conveniently find charging stations, perform charging operations, and conduct payment settlements. At the same time, it also helps charging station operators manage the operation and maintenance of charging facilities. Evaluating the performance of the charging service platform is to ensure its efficient and reliable service to users and operators. Through reasonable performance evaluation, bottlenecks and deficiencies in the platform operation can be identified, providing data support and decision-making basis for subsequent optimization and upgrading.

[0003] However, the existing methods for evaluating the performance of charging service platforms still have the following deficiencies:

[0004] (1) The data sources of the charging service platform are diverse and the formats are not unified, resulting in data heterogeneity problems.

[0005] (2) The charging service platform lacks in-depth analysis of charging behavior patterns and cannot identify complex implicit patterns, reducing the accuracy of the charging service platform's evaluation of service quality.

[0006] For the problems in the related art, no effective solutions have been proposed yet. Summary of the Invention

[0007] In view of the problems in the related art, the present invention proposes a method and system for evaluating the performance of a charging service platform based on a knowledge graph to overcome the above-mentioned technical problems existing in the existing related technologies.

[0008] To this end, the specific technical solutions adopted by the present invention are as follows:

[0009] According to one aspect of the present invention, there is provided a method for evaluating the performance of a charging service platform based on a knowledge graph, the method comprising:

[0010] S1. Obtain a standardized data set of the charging service platform; based on the standardized data set and using knowledge graph technology, construct a knowledge model for the charging service platform;

[0011] S2. Use machine learning and data mining algorithms to extract implicit patterns in the charging service from the knowledge model; based on the implicit patterns in the charging service, construct a charging service quality evaluation model;

[0012] S3. Use the charging service quality evaluation model to monitor and evaluate the real-time performance of the charging service platform, and generate the charging quality evaluation result; from the charging quality evaluation result, identify the charging performance indicators to be optimized;

[0013] S4. Based on the optimization algorithm, adjust the control parameters of the charging performance indicators to be optimized, and return the optimized control parameters to the charging service platform;

[0014] Among them, the standardized data set includes charging station location data, charging pile status data, user behavior data, and equipment failure data.

[0015] Furthermore, based on the standardized data set and using knowledge graph technology, the knowledge model for the charging service platform is constructed as follows:

[0016] Define the domain and scope of the knowledge graph according to the standardized data set; according to the domain and scope of the knowledge graph, use the top-down method to gradually refine the ontology structure of the charging service platform;

[0017] Fill the standardized data set into the ontology structure of the charging service platform to obtain the instantiated knowledge graph;

[0018] Perform link and disambiguation processing on the instantiated knowledge graph to obtain the knowledge model for the charging service platform.

[0019] Furthermore, use machine learning and data mining algorithms to extract the implicit patterns in the charging service from the knowledge model, including:

[0020] Extract the entities and relationships related to charging behavior from the knowledge model; use the clustering algorithm to group the data of the entities and relationships related to charging behavior;

[0021] Use the association rule mining algorithm to mine the association rules of the entities and relationships related to charging behavior;

[0022] Take the data grouping result and the mined association rules as the implicit patterns in the charging service.

[0023] Furthermore, using the clustering algorithm to group the data of the entities and relationships related to charging behavior includes:

[0024] Perform dimensionality reduction and standardization processing on the feature data composed of the entities and relationships related to charging behavior;

[0025] Configure the parameters of the clustering algorithm, and perform clustering analysis on the dimension-reduced and standardized feature data;

[0026] Obtain the clustering result, and according to each clustering result, identify the corresponding charging behavior pattern.

[0027] Further, the use of the association rule mining algorithm to mine the association rules for the entities and relationships related to the charging behavior includes:

[0028] Convert the feature data composed of the entities and relationships related to the charging behavior into a transaction form, and configure the support threshold and confidence threshold of the association rule mining algorithm;

[0029] Use the association rule mining algorithm to identify the frequent item sets that meet the support threshold from the feature data;

[0030] Generate association rules that meet the confidence threshold from the frequent item sets.

[0031] Further, based on the implicit patterns in the charging service, constructing a charging service quality evaluation model includes:

[0032] Identify the key performance indicators from the implicit patterns in the charging service, and assign weights to each key performance indicator;

[0033] Use each key performance indicator and its corresponding weight to construct a charging service quality evaluation model.

[0034] Further, use the charging service quality evaluation model to monitor and evaluate the real-time performance of the charging service platform, and generate a charging quality evaluation result; from the charging quality evaluation result, identify the charging performance indicators to be optimized, including:

[0035] Collect the standardized data set of the real-time charging service platform, and combine it with the charging service quality evaluation model to obtain the charging quality evaluation result;

[0036] Analyze the charging quality evaluation result, and take the key performance indicators lower than the preset threshold as the charging performance indicators to be optimized.

[0037] Further, based on the optimization algorithm, adjust the control parameters of the charging performance indicators to be optimized, and return the optimized control parameters to the charging service platform, including:

[0038] Configure the optimization objective and constraint conditions of the optimization algorithm; initialize the optimization algorithm, and use the current control parameters as the initial solution;

[0039] Iteratively update the control parameters within the constraint range and optimize the objective function; evaluate the result of each iteration, and take the optimal solution as the optimized control parameter;

[0040] Feed back the optimized control parameters to the charging service platform.

[0041] Further, configuring the optimization objective and constraint conditions of the optimization algorithm includes:

[0042] Define the optimization objectives, which include minimizing waiting time, maximizing charging efficiency, and reducing failure rates;

[0043] Determine the constraint conditions, which include technical constraints, economic constraints, and safety limit constraints.

[0044] According to another aspect of the present invention, there is also provided a performance evaluation system for a charging service platform based on a knowledge graph. The system includes a knowledge model construction module, an evaluation model construction module, an index to be optimized identification module, and an index control parameter optimization module, and the knowledge model construction module, the evaluation model construction module, the index to be optimized identification module, and the index control parameter optimization module are sequentially connected; wherein, the knowledge model construction module is used to obtain a standardized data set of the charging service platform; based on the standardized data set and using knowledge graph technology, construct a knowledge model for the charging service platform; the evaluation model construction module is used to extract implicit patterns in the charging service from the knowledge model by using machine learning and data mining algorithms; based on the implicit patterns in the charging service, construct a charging service quality evaluation model; the index to be optimized identification module is used to perform real-time performance monitoring and evaluation on the charging service platform by using the charging service quality evaluation model, and generate a charging quality evaluation result; from the charging quality evaluation result, identify the charging performance index to be optimized; the index control parameter optimization module is used to adjust the control parameters of the charging performance index to be optimized based on an optimization algorithm, and return the optimized control parameters to the charging service platform.

[0045] The beneficial effects of the present invention are as follows:

[0046] (1) Through the construction of the knowledge graph (knowledge model), the present invention can effectively integrate multi-source heterogeneous data, such as charging station locations, equipment status, user behavior, and failure data, making data management more efficient and systematic. It provides a structured way to represent and query various types of information on the charging service platform, supporting more flexible queries and analyses.

[0047] (2) By using machine learning and data mining algorithms, complex charging behavior patterns and implicit relationships can be extracted from the knowledge model to help identify key factors affecting service quality. The performance evaluation model constructed based on the implicit patterns can monitor the charging service quality in real time and provide dynamic performance evaluation results. After identifying the performance indicators to be optimized, the control parameters are adjusted through an optimization algorithm to achieve continuous improvement of service quality. Description of the Drawings

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for use in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0049] Figure 1 is a flowchart of a method for evaluating the performance of a charging service platform based on a knowledge graph according to an embodiment of the present invention;

[0050] Figure 2 is a block diagram of a module of a system for evaluating the performance of a charging service platform based on a knowledge graph according to an embodiment of the present invention.

[0051] In the figure:

[0052] 1. Knowledge model construction module; 2. Evaluation model construction module; 3. To-be-optimized index identification module; 4. Index control parameter optimization module. Detailed implementation manners

[0053] To further illustrate the embodiments, the present invention provides accompanying drawings. These accompanying drawings are part of the disclosure of the present invention, mainly used to illustrate the embodiments, and can cooperate with the relevant descriptions in the specification to explain the operation principle of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0054] According to an embodiment of the present invention, a method and a system for evaluating the performance of a charging service platform based on a knowledge graph are provided.

[0055] Now, the present invention will be further described in combination with the accompanying drawings and specific implementation manners. As Figure 1 shown, according to an embodiment of the present invention, a method for evaluating the performance of a charging service platform based on a knowledge graph is provided. The method includes:

[0056] S1. Obtain a standardized data set of the charging service platform; based on the standardized data set and using knowledge graph technology, construct a knowledge model for the charging service platform. The standardized data set includes charging station location data, charging pile status data, user behavior data, and equipment failure data.

[0057] In one embodiment, constructing a knowledge model for the charging service platform based on the standardized data set and using knowledge graph technology includes:

[0058] Define the domain and scope of the knowledge graph based on the standardized dataset; refine the ontology structure of the charging service platform layer by layer using a top-down approach according to the domain and scope of the knowledge graph.

[0059] Fill the ontology structure of the charging service platform with the standardized dataset to obtain an instantiated knowledge graph.

[0060] Perform link and disambiguation processing on the instantiated knowledge graph to obtain a knowledge model for the charging service platform.

[0061] Specifically, determine the domain and scope of the knowledge graph based on the information in the standardized dataset. This includes identifying the core entities (such as charging stations, charging piles, users, devices, etc.) and relationships of the charging service platform. Use a top-down approach to refine the ontology structure layer by layer, involving refining from general concepts (such as "charging equipment") to specific attributes and relationships (such as "charging pile status", "user behavior").

[0062] Fill the specific data instances in the standardized dataset into the ontology structure. This includes mapping the actual charging pile locations, statuses, user behaviors, etc. data to the corresponding entities and attributes in the ontology.

[0063] Perform entity linking on the knowledge graph to ensure that all entities are accurately corresponding in the graph. Disambiguation processing is to solve the problem of the same entity with different names or different entities with the same name, ensuring the uniqueness and accuracy of entities and relationships.

[0064] S2. Use machine learning and data mining algorithms to extract implicit patterns in the charging service from the knowledge model; build a charging service quality evaluation model based on the implicit patterns in the charging service.

[0065] In one embodiment, using machine learning and data mining algorithms to extract implicit patterns in the charging service from the knowledge model includes:

[0066] Extract entities and relationships related to charging behavior from the knowledge model; use a clustering algorithm to group the data of entities and relationships related to charging behavior.

[0067] Adopt an association rule mining algorithm to mine association rules for entities and relationships related to charging behavior.

[0068] Take the data grouping results and mined association rules as the implicit patterns in the charging service.

[0069] In one embodiment, using a clustering algorithm to group the data of entities and relationships related to charging behavior includes:

[0070] Perform dimensionality reduction and standardization processing on the feature data composed of entities and relationships related to charging behavior.

[0071] Configure the parameters of the clustering algorithm and perform clustering analysis on the feature data after dimensionality reduction and standardization.

[0072] Obtain the clustering results and identify the corresponding charging behavior patterns according to each clustering result.

[0073] It should be noted that the knowledge model includes multiple entities and relationships related to charging services. For example, entities: charging stations, charging piles, users, vehicles, time periods, geographical locations, etc. Relationships: such as "user - uses - charging pile", "charging station - located at - location", "vehicle - charges - time period", etc.

[0074] Charging station features: geographical location, service capacity, average waiting time.

[0075] Charging pile features: type (fast, slow), usage frequency, failure rate.

[0076] User behavior features: charging frequency, average charging time, preferred time period.

[0077] Environmental features: weather conditions, influence of special events, etc.

[0078] For high - dimensional feature data, use dimensionality reduction techniques (such as PCA) to simplify the data structure and retain the main information. Perform standardization processing on the feature data to eliminate the influence of different dimensions and make each feature have the same importance in clustering analysis.

[0079] Select a suitable clustering algorithm (such as K - means, DBSCAN) and configure the parameters. Perform clustering analysis on the processed feature data and divide the data into several groups. By analyzing the clustering results, identify different charging behavior patterns. For example, the preferences or rules of different user groups when using charging piles.

[0080] In one embodiment, an association rule mining algorithm is used to mine association rules for entities and relationships related to charging behavior, including:

[0081] Convert the feature data composed of entities and relationships related to charging behavior into a transaction form, and regard each charging behavior record as a transaction. For example, a transaction includes the charging station selected by the user within a specific time, the type of charging pile used, the charging duration, etc. Configure the support threshold and confidence threshold of the association rule mining algorithm.

[0082] Use the association rule mining algorithm to identify frequent item sets that meet the support threshold from the feature data.

[0083] Generate association rules that meet the confidence threshold from frequent item sets, and use the confidence threshold to filter the generated rules to ensure that only those rules with sufficient confidence support are retained.

[0084] It should be noted that the transaction form can represent a complete record of a charging behavior that occurs within a specific time. Converting the entities and relationships related to the charging behavior into transactions means integrating all the features involved in each charging behavior record (such as a charging process), such as the charging station location, charging pile type, charging duration, user ID, etc., into a set. For example: A transaction may include: ("Charging Station A", "Fast Charging Pile", "30 minutes", "User ID123").

[0085] Among them, setting the support threshold is to filter out those item sets that do not appear frequently, so as to focus on more general patterns. High confidence means that there is a strong association relationship between the antecedent and the consequent. Setting the confidence threshold is to ensure that only those rules with strong associations are retained. For example, a confidence of 80% means that if the antecedent appears, there is an 80% probability that the consequent will also appear.

[0086] Scan the data set through an algorithm (such as Apriori) to find all item sets that meet the support threshold. These item sets are the basis of potential association patterns. The process involves scanning the data set multiple times, gradually expanding the item sets, and calculating the support of each item set. Generate association rules from the frequent item sets, such as "If charging at Charging Station A, then use a fast charging pile"). Calculate the confidence of each rule and compare it with the preset confidence threshold. Only retain those rules with a confidence higher than the threshold.

[0087] In one embodiment, based on the implicit patterns in the charging service, constructing a charging service quality evaluation model includes:

[0088] Identify key performance indicators from the implicit patterns in the charging service and assign weights to each key performance indicator.

[0089] Utilize each key performance indicator and its corresponding weight to construct a charging service quality evaluation model.

[0090] For example, assign corresponding weights according to the importance of each indicator and calculate its score. The calculation formula for the score is: Score = Normalized value of the indicator × Weight; Add the weighted scores of each indicator to obtain the total score. The calculation method of the total score is as follows: Total score = ∑(Normalized value of the indicator × Weight).

[0091] S3. Use the charging service quality evaluation model to perform real-time performance monitoring and evaluation on the charging service platform and generate a charging quality evaluation result; Identify the charging performance indicators to be optimized from the charging quality evaluation result.

[0092] In one embodiment, a charging service quality evaluation model is used to monitor and evaluate the real-time performance of the charging service platform, and a charging quality evaluation result is generated; from the charging quality evaluation result, the charging performance indicators to be optimized are identified, including:

[0093] Collect a standardized data set of the real-time charging service platform, and combine it with the charging service quality evaluation model to obtain a charging quality evaluation result.

[0094] Analyze the charging quality evaluation result, and take the key performance indicators below the preset threshold as the charging performance indicators to be optimized.

[0095] Specifically, when comparing the score of each indicator with the preset threshold, the preset threshold is set according to historical data, industry standards or business goals. Identify those key performance indicators whose scores are below the preset threshold, because they may be the bottlenecks affecting the overall service quality.

[0096] Among them, historical data: Past operation data provides a baseline for judging whether the current performance deviates from the normal range.

[0097] Industry standards: Refer to the best practices and standards in the industry to ensure that the charging service maintains an advantage in the competition.

[0098] Business goals: Set according to the strategic goals of the enterprise, such as improving user satisfaction or reducing operating costs.

[0099] The indicators with scores below the preset threshold are the bottlenecks affecting the overall service quality and need special attention. Such as:

[0100] High waiting time: The main reason is the unreasonable layout of charging stations or insufficient number of charging piles.

[0101] Low user satisfaction: The main reason is untimely service or opaque charging.

[0102] High failure rate: The main reason is equipment aging or untimely maintenance.

[0103] Feed back the identified indicators to be optimized to the management or relevant teams as the basis for further analysis and decision-making. Develop targeted improvement measures, such as increasing charging piles, optimizing the layout of charging stations, and improving the response speed of service personnel. After implementing the improvement measures, continuously monitor these indicators to evaluate the improvement effect and make necessary adjustments.

[0104] S4. Based on the optimization algorithm, adjust the control parameters of the charging performance indicators to be optimized, and return the optimized control parameters to the charging service platform.

[0105] In one embodiment, based on an optimization algorithm, adjusting the control parameters of the charging performance metrics to be optimized and returning the optimized control parameters to the charging service platform includes:

[0106] Configuring the optimization objectives and constraints of the optimization algorithm; initializing the optimization algorithm and using the current control parameters as the initial solution. These control parameters are the current configurations, such as the current charging scheduling strategy, device settings, etc.

[0107] Iteratively updating the control parameters within the constraints and optimizing the objective function; evaluating the results of each iteration and using the optimal solution as the optimized control parameters.

[0108] Feeding back the optimized control parameters to the charging service platform.

[0109] In one embodiment, configuring the optimization objectives and constraints of the optimization algorithm includes:

[0110] Defining the optimization objectives, which include minimizing the waiting time, maximizing the charging efficiency, and reducing the failure rate. Determining the constraints, which include technical constraints, economic constraints, and safety limit constraints.

[0111] It should be noted that the optimization algorithms include genetic algorithms, simulated annealing, particle swarm optimization, etc. These optimization algorithms gradually approach the optimal solution by exploring and exploiting the existing solution space. In each iteration, the value of the objective function is calculated. The objective function reflects the achievement of the optimization objectives under the current parameter settings, and the direction and amplitude of parameter adjustment are judged based on the value of the objective function.

[0112] Applying the optimized control parameters to the charging service platform, which involves updating the scheduling strategy, reconfiguring the device parameters, etc. Ensure the stability and safety of the system during the implementation process. After implementing the optimized parameters, continuously monitor their effects and make necessary adjustments and re-optimizations based on real-time data.

[0113] Minimizing the waiting time: By optimizing the charging station scheduling and resource allocation, reduce the waiting time of users during charging.

[0114] Maximizing the charging efficiency: Improve the utilization efficiency of charging equipment, which involves optimizing the charging power distribution or reducing energy losses.

[0115] Reducing the failure rate: By optimizing the maintenance plan and device usage strategy, reduce the frequency of device failures.

[0116] Technical constraints: Such as the maximum power limit of the device, the physical capacity of the charging station, etc.

[0117] Economic constraints: Such as budget limitations, cost-benefit requirements, etc.

[0118] Safety restriction constraints: including user and device safety requirements, such as avoiding overcharging and ensuring grid stability, etc.

[0119] As Figure 2 As shown, according to another embodiment of the present invention, there is also provided a performance evaluation system for a charging service platform based on a knowledge graph. The system includes a knowledge model construction module 1, an evaluation model construction module 2, an index to be optimized identification module 3, and an index control parameter optimization module 4. And the knowledge model construction module 1, the evaluation model construction module 2, the index to be optimized identification module 3, and the index control parameter optimization module 4 are sequentially connected. The knowledge model construction module 1 is used to obtain a standardized data set of the charging service platform; based on the standardized data set, and using knowledge graph technology, construct a knowledge model for the charging service platform; the evaluation model construction module 2 is used to use machine learning and data mining algorithms to extract implicit patterns in the charging service from the knowledge model; based on the implicit patterns in the charging service, construct a charging service quality evaluation model; the index to be optimized identification module 3 is used to use the charging service quality evaluation model to perform real-time performance monitoring and evaluation on the charging service platform, and generate a charging quality evaluation result; from the charging quality evaluation result, identify the charging performance index to be optimized; the index control parameter optimization module 4 is used to adjust the control parameters of the charging performance index to be optimized based on an optimization algorithm, and return the optimized control parameters to the charging service platform.

[0120] In summary, through the construction of a knowledge graph (knowledge model), the present invention can effectively integrate multi-source heterogeneous data, such as charging station locations, device status, user behavior, and fault data, making data management more efficient and systematic. It provides a structured way to represent and query various types of information of the charging service platform, supporting more flexible queries and analyses. Using machine learning and data mining algorithms, complex charging behavior patterns and implicit relationships can be extracted from the knowledge model to help identify key factors affecting service quality. The performance evaluation model constructed based on the implicit patterns can monitor the charging service quality in real time and provide dynamic performance evaluation results. After identifying the performance indicators to be optimized, the control parameters are adjusted through an optimization algorithm to achieve continuous improvement of service quality.

[0121] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the performance of a charging service platform based on a knowledge graph, characterized in that The method includes: S1. Obtain the standardized data set of the charging service platform; based on the standardized data set and using the knowledge graph technology, construct a knowledge model for the charging service platform; S2. Use machine learning and data mining algorithms to extract implicit patterns in the charging service from the knowledge model; based on the implicit patterns in the charging service, construct a charging service quality evaluation model; S3. Use the charging service quality evaluation model to monitor and evaluate the real-time performance of the charging service platform, and generate a charging quality evaluation result; from the charging quality evaluation result, identify the charging performance indicators to be optimized; S4. Based on the optimization algorithm, adjust the control parameters of the charging performance indicators to be optimized, and return the optimized control parameters to the charging service platform; Among them, the standardized data set includes charging station location data, charging pile status data, user behavior data, and equipment failure data.

2. The performance evaluation method of a charging service platform based on a knowledge graph according to claim 1, wherein The constructing a knowledge model for the charging service platform based on the standardized data set and using the knowledge graph technology includes: Define the domain and scope of the knowledge graph according to the standardized data set; according to the domain and scope of the knowledge graph, adopt a top-down method to gradually refine the ontology structure of the charging service platform; Fill the standardized data set into the ontology structure of the charging service platform to obtain an instantiated knowledge graph; Perform link and disambiguation processing on the instantiated knowledge graph to obtain a knowledge model for the charging service platform.

3. The performance evaluation method of a charging service platform based on a knowledge graph according to claim 1, characterized in that The extracting implicit patterns in the charging service from the knowledge model using machine learning and data mining algorithms includes: Extract entities and relationships related to charging behavior from the knowledge model; use the clustering algorithm to group the data of entities and relationships related to charging behavior; Adopt the association rule mining algorithm to mine the association rules of entities and relationships related to charging behavior; Take the data grouping result and the mined association rules as the implicit patterns in the charging service.

4. The performance evaluation method of a charging service platform based on a knowledge graph according to claim 3, characterized in that, The using the clustering algorithm to group the data of entities and relationships related to charging behavior includes: Perform dimensionality reduction and standardization processing on the feature data composed of entities and relationships related to charging behavior; Configure the parameters of the clustering algorithm, and perform clustering analysis on the dimensionally reduced and standardized feature data; Obtain the clustering result, and according to each clustering result, identify the corresponding charging behavior pattern.

5. The performance evaluation method of a charging service platform based on a knowledge graph according to claim 3, characterized in that The adopting the association rule mining algorithm to mine the association rules of entities and relationships related to charging behavior includes: Convert the feature data composed of entities and relationships related to charging behavior into a transaction form, and configure the support threshold and confidence threshold of the association rule mining algorithm; Use the association rule mining algorithm to identify frequent item sets that meet the support threshold from the feature data; Generate association rules that meet the confidence threshold from the frequent item sets.

6. The performance evaluation method of a charging service platform based on a knowledge graph according to claim 1, characterized in that, The constructing a charging service quality evaluation model based on the implicit patterns in the charging service includes: Identify key performance indicators from the implicit patterns in the charging service, and assign weights to each key performance indicator; Use each key performance indicator and the corresponding weight to construct a charging service quality evaluation model.

7. A performance evaluation method for a charging service platform based on a knowledge graph according to claim 1, characterized in that The charging service platform is monitored and evaluated in real time using the charging service quality evaluation model, and a charging quality evaluation result is generated; From the charging quality evaluation result, the charging performance indicators to be optimized are identified, including: Collect a standardized data set of the real-time charging service platform, and combine it with the charging service quality evaluation model to obtain a charging quality evaluation result; Analyze the charging quality evaluation result, and use the key performance indicators below the preset threshold as the charging performance indicators to be optimized.

8. A performance evaluation method for a charging service platform based on a knowledge graph according to claim 1, characterized in that, Based on the optimization algorithm, adjusting the control parameters of the charging performance indicators to be optimized and returning the optimized control parameters to the charging service platform includes: Configure the optimization objectives and constraints of the optimization algorithm; initialize the optimization algorithm and use the current control parameters as the initial solution; Iteratively update the control parameters within the constraints and optimize the objective function; evaluate the results of each iteration and use the optimal solution as the optimized control parameters; Feed back the optimized control parameters to the charging service platform.

9. A method for evaluating the performance of a charging service platform based on a knowledge graph according to claim 8, wherein, Configuring the optimization objectives and constraints of the optimization algorithm includes: Define the optimization objectives, which include minimizing the waiting time, maximizing the charging efficiency, and reducing the failure rate; Determine the constraints, which include technical constraints, economic constraints, and safety limit constraints.

10. A performance evaluation system for a charging service platform based on a knowledge graph, which is used to implement the performance evaluation method for the charging service platform based on a knowledge graph according to any one of claims 1-9, characterized in that, The system includes a knowledge model construction module, an evaluation model construction module, an indicator to be optimized identification module, and an indicator control parameter optimization module, and the knowledge model construction module, the evaluation model construction module, the indicator to be optimized identification module, and the indicator control parameter optimization module are connected in sequence; Among them, the knowledge model construction module is used to obtain the standardized data set of the charging service platform; based on the standardized data set, and using the knowledge graph technology, construct a knowledge model for the charging service platform; The evaluation model construction module is used to extract the implicit patterns in the charging service from the knowledge model using machine learning and data mining algorithms; based on the implicit patterns in the charging service, construct a charging service quality evaluation model; The indicator to be optimized identification module is used to monitor and evaluate the real-time performance of the charging service platform using the charging service quality evaluation model, and generate a charging quality evaluation result; from the charging quality evaluation result, identify the charging performance indicators to be optimized; The indicator control parameter optimization module is used to adjust the control parameters of the charging performance indicators to be optimized based on the optimization algorithm, and return the optimized control parameters to the charging service platform.

Images