A method and system for multi-source information fusion in a smart charging network

Abstract

The present invention provides a multi-source information fusion method and system in a smart charging network, and relates to the technical field of data processing. The method comprises: acquiring original data; preprocessing the acquired original data to obtain preprocessed data; classifying the preprocessed data according to the information source, and performing unified timestamp synchronization on the same type of data to generate a preliminary fusion data set; performing association analysis on the same type of data in the preliminary fusion data set, removing redundant information and screening to obtain multiple streamlined data sets; performing cross-class data association on the streamlined data set to generate a multi-dimensional fusion data set; grading the information in the multi-dimensional fusion data set to form charging requests of different levels; allocating charging power and time periods to charging requests of different priorities to obtain an optimized scheduling data set; performing conflict detection and adjustment on the optimized scheduling data set to generate an optimized control data set. The present invention can provide good fusion data support for power grid control.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for fusing multi-source information in a smart charging network. Background Art

[0002] Existing smart charging networks rely mainly on information from a single source, such as vehicle battery level and charging pile status, to optimize operation and control. Traditional technologies usually use these single information sources to make system decisions and operational controls, such as determining charging duration and adjusting charging power. However, with the increase in charging requests and the diversification of charging network application scenarios, relying solely on a single information source can no longer fully reflect the complex dynamics in the charging network. This model lacks effective fusion and analysis of information from multiple different sources, making it difficult to respond comprehensively and in real time to complex charging environments. The intelligent and efficient operation of charging networks urgently requires accurate decision-making through the fusion of multi-source information.

[0003] For example, in actual application scenarios, in addition to obtaining vehicle battery status information, the smart charging network also needs to integrate other key data, such as the real-time load of the power grid, the dynamic changes in renewable energy generation, traffic flow fluctuations, and user charging priority requirements. Existing technologies often lack a unified fusion mechanism when processing these heterogeneous data, and may not be able to effectively screen and extract information useful for system optimization. Taking the change in power grid load as an example, relying solely on vehicle battery power information for charging scheduling may lead to unreasonable allocation of charging power during peak power grid load periods, causing local power grid overload or excessive charging time. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for multi-source information fusion in a smart charging network, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a method for fusing multi-source information in a smart charging network is provided, the method comprising:

[0007] S100, acquiring raw data related to the charging network from multiple information sources, wherein the information sources include vehicle battery status information, charging pile status information, grid load information, traffic flow information, and user charging request information;

[0008] S200, formatting, denoising and data integrity checking the acquired raw data to obtain pre-processed data;

[0009] S300, classifying the preprocessed data according to the information source, and performing unified timestamp synchronization on the same type of data to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set;

[0010] S400, performing association analysis on similar data in the preliminary fused data set, removing redundant information and screening, and obtaining multiple streamlined data sets;

[0011] S500, by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and power grid load, cross-class data association is performed on the streamlined data set to generate a multi-dimensional fusion data set;

[0012] S600, scoring the user's charging time requirements according to the multi-dimensional fusion data set to form charging requests of different priorities, including a high priority charging request, a medium priority charging request, and a low priority charging request;

[0013] S700, allocating charging power and time periods for charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme;

[0014] S800: Conflict detection is performed on the optimized scheduling data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load, and the optimized scheduling data set is adjusted according to the potential conflicts to generate an optimized control data set.

[0015] Preferably, the preprocessed data is classified according to the information source, and the same type of data is synchronized with a unified timestamp to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set, including:

[0016] Standardize the data format of raw data from different information sources to obtain formatted data;

[0017] The noise data in the formatted data is cleared according to a preset noise threshold to obtain denoised data;

[0018] The denoised data is verified for data integrity, and abnormal data is removed to obtain preprocessed data.

[0019] Preferably, the preprocessed data is classified according to the information source, and the same type of data is synchronized with a unified timestamp to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set, including:

[0020] Classify the preprocessed data according to the type of information source to obtain a preliminary classified data set;

[0021] Perform unified timestamp synchronization processing on data from the same information source in the preliminary classification data set to generate a synchronized data set;

[0022] According to the category of information source, the synchronized data set is divided into vehicle battery status data set, charging pile status data set, power grid load data set, traffic flow data set and user demand data set to form a preliminary fusion data set.

[0023] Preferably, similar data in the preliminary fusion data set are subjected to association analysis, redundant information is removed and screened, and multiple streamlined data sets are obtained, including:

[0024] Perform association analysis on similar data in the preliminary fusion data set to determine the degree of association between similar data and obtain a preliminary association data set. The formula is as follows:

[0025] ,

[0026] in, For similar data and The correlation between For the The weight of the category, For data In the The standard value on the category, For data In the The standard value on the category, is the number of categories;

[0027] According to a preset correlation threshold, redundant data is removed from the preliminary correlation data set to obtain a de-redundant data set;

[0028] The key data of the de-redundant data set is screened according to the preset timestamp threshold to generate a streamlined data set.

[0029] Preferably, by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and grid load, cross-class data association is performed on the streamlined data set to generate a multi-dimensional fusion data set, including:

[0030] Perform cross-class correlation analysis on different data sets in the streamlined data set, analyze the correlation between vehicle battery status, charging pile utilization, traffic flow and power grid load, and obtain a correlation information data set;

[0031] According to the associated information in the associated information dataset, low-value cross-category data in the streamlined dataset is eliminated to obtain a cross-category screening dataset;

[0032] The cross-class screening dataset and the associated information dataset are fused in multiple dimensions to generate a multi-dimensional fused dataset.

[0033] Preferably, cross-category association analysis is performed on different data sets in the simplified data set to analyze the association between vehicle battery status, charging pile utilization, traffic flow and power grid load to obtain an associated information data set, including:

[0034] Extract vehicle battery status data, charging station utilization data, traffic flow data, and grid load data from the streamlined data set;

[0035] According to the remaining battery power of the vehicle and the current utilization rate of the charging pile, the urgency of the demand for charging resources for different charging requests is calculated to obtain the battery-charging pile association information;

[0036] Match the grid load data with the charging pile utilization data, compare the grid load fluctuation and the charging pile power consumption trend, and generate grid-charging pile load correlation information;

[0037] Analyze the impact of traffic flow on each charging request based on traffic flow data and vehicle battery status data, and generate traffic-charging priority association information;

[0038] The battery-charging pile correlation information, the power grid-charging pile load correlation information, and the traffic-charging priority correlation information are merged to generate a correlation information data set.

[0039] Preferably, the user's charging time requirements are scored according to the multi-dimensional fusion data set to form charging requests of different priorities, including a high priority charging request, a medium priority charging request and a low priority charging request, including:

[0040] Score the user's charging time requirements based on the multi-dimensional fusion data set to obtain a priority score. The formula is as follows:

[0041] ,

[0042] in, Score the priority of charging request i, is the adjustment factor, The time requirement for charging request, is the time-dependent attenuation factor, is the current remaining power of the vehicle battery, is the maximum capacity of the vehicle battery, is the weight of the grid load, is the value of the current grid load, is the current utilization rate of the charging pile, is the current traffic flow, is the adjustment coefficient, is a regulation parameter related to the grid load;

[0043] The priority scores of different charging requests are divided according to a preset scoring threshold range to obtain a high priority charging request, a medium priority charging request and a low priority charging request;

[0044] Data with different priorities are labeled and classified into corresponding charging request types to generate a hierarchical dataset.

[0045] Preferably, charging power and time periods are allocated to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme, including:

[0046] According to the priorities of different charging requests, the corresponding allocated power is determined, and a power allocation data set is formed according to the allocated power; the calculation formula of the allocated power is:

[0047] ,

[0048] in, is the allocated power for charging request i, is the maximum charging power available in the current power grid, For the The priority score of each charging request, is the nonlinear coefficient, To prevent division by zero, The charging time requested by the user, is the maximum charging time for all requests in the system, is the weight coefficient, is a nonlinear parameter.

[0049] According to the priorities of different charging requests, the corresponding time period allocation length is determined, and according to the time period allocation length, a time period allocation data set is formed; the calculation formula of the time period allocation length is:

[0050] ,

[0051] in, Assign a length to the time period of charging request i, Score the priority of the current charging request i, is the influence coefficient, For the The power requirement of a charging request, Nonlinear coefficient;

[0052] According to the power allocation data set and the time period allocation data set, an optimized scheduling data set is generated, including a charging power allocation plan and a charging time period allocation plan.

[0053] Preferably, conflict detection is performed on the optimized scheduling data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load, and the optimized scheduling data set is adjusted according to the potential conflicts to generate an optimized control data set, including:

[0054] Conflict detection on the optimized dispatch data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load;

[0055] Perform data analysis on potential conflicts to obtain conflict severity scores, and adjust the allocation scheme in the corresponding optimized scheduling data set according to the conflict severity scores to generate an optimized control data set;

[0056] The conflict severity score is calculated using the following formula:

[0057] ,

[0058] in, Score the conflict severity of the charging request, which is used to indicate the total amount of load overload. is the total number of time periods for evaluation, For time period The total charging power requirement in is the maximum load power allowed by the power grid in each time period, For calculating time period If the limit is not exceeded, it will be counted as 0.

[0059] In a second aspect, a multi-source information fusion system in a smart charging network includes:

[0060] A data acquisition module, used to acquire raw data related to the charging network from multiple information sources, the information sources including vehicle battery status information, charging pile status information, grid load information, traffic flow information and user charging request information;

[0061] A preprocessing module is used to format, remove noise and verify data integrity of the acquired raw data to obtain preprocessed data;

[0062] The preliminary fusion module is used to classify the pre-processed data according to the information source, synchronize the timestamps of the same type of data, and generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set;

[0063] The data reduction module is used to perform correlation analysis on similar data in the preliminary fusion data set, remove redundant information and perform screening to obtain multiple reduced data sets;

[0064] The data fusion module is used to generate a multi-dimensional fused data set by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and grid load, and performing cross-class data association on the streamlined data set;

[0065] The data screening module is used to remove redundant and useless data from the multi-dimensional fusion data set and generate a key data set based on the real-time status of the charging network and user charging requests;

[0066] The request classification module is used to classify the information in the key data set according to the user's charging time requirements, the remaining battery power of the vehicle, and the current grid load, and form different levels of charging requests, including high priority charging requests, medium priority charging requests, and low priority charging requests;

[0067] A scheme allocation module is used to allocate charging power and time periods to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme;

[0068] The conflict optimization module is used to perform conflict detection on the optimized scheduling data set, identify potential conflicts that lead to excessive concentration of charging resources or excessive load, and adjust the optimized scheduling data set according to the potential conflicts to generate an optimized control data set.

[0069] The above solution of the present invention includes at least the following beneficial effects:

[0070] The present invention achieves accurate response and efficient control of complex dynamics in the smart charging network by integrating multi-source information. Unlike traditional smart charging systems that rely on only a single information source, this method obtains raw data from multiple information sources, including vehicle battery status, charging pile status, grid load, traffic flow, and user demand, formats, denoises, performs integrity verification, and classifies it, and generates a multi-dimensional fusion data set, providing a comprehensive and accurate decision-making basis for the intelligent operation of the charging network.

[0071] In the process of information fusion, this method screens and extracts the most critical data for system optimization through cross-category correlation analysis of multi-source information and the potential relationship between different information sources, and further classifies charging requests through a priority scoring system to ensure that high-priority charging demands can be met first during peak load periods. This multi-level data processing method solves the problem of one-sided data caused by a single information source in the existing technology, and can dynamically adjust the allocation of charging power and charging time periods based on real-time data, and can also respond flexibly when the load changes greatly.

[0072] In addition, by performing conflict detection on the optimized scheduling data set, identifying potential conflicts that lead to excessive concentration of charging resources or overload, and adjusting the optimized scheduling data set according to the potential conflicts to generate an optimized control data set, this method significantly improves the efficiency of allocating charging resources among different demands. And this allocation strategy avoids the local overload and charging delay caused by unreasonable allocation in traditional technologies. Through multi-source information fusion and intelligent scheduling, this method ensures that the charging network has higher reliability and flexibility in complex environments and realizes the rational allocation of global resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flowchart of a multi-source information fusion method in a smart charging network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0074] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0075] like Figure 1 As shown, an embodiment of the present invention proposes a multi-source information fusion method in a smart charging network, and the method includes the following steps:

[0076] S100, acquiring raw data related to the charging network from multiple information sources, wherein the information sources include vehicle battery status information, charging pile status information, grid load information, traffic flow information, and user charging request information;

[0077] S200, formatting, denoising and data integrity checking the acquired raw data to obtain pre-processed data;

[0078] S300, classifying the preprocessed data according to the information source, and performing unified timestamp synchronization on the same type of data to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set;

[0079] S400, performing association analysis on similar data in the preliminary fused data set, removing redundant information and screening, and obtaining multiple streamlined data sets;

[0080] S500, by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and power grid load, cross-class data association is performed on the streamlined data set to generate a multi-dimensional fusion data set;

[0081] S600, scoring the user's charging time requirements according to the multi-dimensional fusion data set to form charging requests of different priorities, including a high priority charging request, a medium priority charging request, and a low priority charging request;

[0082] S700, allocating charging power and time periods for charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme;

[0083] S800: Conflict detection is performed on the optimized scheduling data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load, and the optimized scheduling data set is adjusted according to the potential conflicts to generate an optimized control data set.

[0084] In the embodiment of the present invention, the present invention realizes accurate response and efficient control of complex dynamics in the smart charging network by integrating multi-source information. Unlike the traditional smart charging system that relies on only a single information source, the present method obtains raw data from multiple information sources, including vehicle battery status, charging pile status, grid load, traffic flow and user demand, formats, denoises, performs integrity verification and classification processing on it, and generates a multi-dimensional fusion data set, which provides a comprehensive and accurate decision-making basis for the intelligent operation of the charging network.

[0085] In the process of information fusion, this method screens and extracts the most critical data for system optimization through cross-category correlation analysis of multi-source information and the potential relationship between different information sources, and further classifies charging requests through a priority scoring system to ensure that high-priority charging demands can be met first during peak load periods. This multi-level data processing method solves the problem of one-sided data caused by a single information source in the existing technology, and can dynamically adjust the allocation of charging power and charging time periods based on real-time data, and can also respond flexibly when the load changes greatly.

[0086] In addition, by performing conflict detection on the optimized scheduling data set, identifying potential conflicts that lead to excessive concentration of charging resources or overload, and adjusting the optimized scheduling data set according to the potential conflicts to generate an optimized control data set, this method significantly improves the efficiency of allocating charging resources among different demands. And this allocation strategy avoids the local overload and charging delay caused by unreasonable allocation in traditional technologies. Through multi-source information fusion and intelligent scheduling, this method ensures that the charging network has higher reliability and flexibility in complex environments and realizes the rational allocation of global resources.

[0087] In actual application scenarios, such as in public charging stations with large traffic flow fluctuations, the system can monitor the charging needs of vehicles, the utilization rate of charging piles, the grid load, and changes in traffic flow in real time, estimate the load pressure during peak hours in advance, and dynamically adjust the charging time period and power of low-priority requests to ensure that high-priority requests receive timely charging support during peak hours. This multi-source information fusion mechanism solves the problems of resource allocation delays and low charging efficiency caused by traditional single information sources, enabling the smart charging network to have higher response speed and stability when dealing with complex and changing needs, greatly improving the operating efficiency of the charging network.

[0088] Among them, in step S200, the role of these preprocessing steps is to improve the accuracy, completeness and compatibility of the data, thereby providing a reliable data basis for subsequent data analysis and decision-making. Specifically:

[0089] Formatting: Converting raw data from different information sources into a unified data format. Since the data received by the system may come from a variety of different sources such as vehicle battery status, charging pile status, power grid load, traffic flow, etc., the data formats of each information source may be different (such as units, precision, time stamp, etc.). Through formatting, the numerical unit, data type and time format of the data are standardized so that all data meet the unified format requirements, which is convenient for subsequent processing and analysis. This process ensures data compatibility, enables data from different information sources to be recognized and accurately read by the system, and reduces the possibility of data confusion and processing errors.

[0090] The preset format conversion module converts units and standardizes the format of data from different sources. The data format of each information source is predefined. For example, the vehicle battery status data may contain parameters such as power percentage, temperature, and health status, while the grid load may be in units of power (kW).

[0091] Use the format conversion algorithm to match the units of inconsistent data and convert all values ​​into standard units. In terms of time format, through timestamp conversion, the time stamps of each data are converted into a unified UTC format to ensure that all types of data are processed according to a unified time clue.

[0092] Denoising: used to remove random noise or abnormal data from the data. Noise is inevitable in the data collection process and may be caused by equipment failure, signal interference, etc. For example, the grid load data in a certain period may have abnormal peaks due to collection errors, which will affect the accuracy of the system's decision-making. Through denoising, abnormal data can be screened out based on the preset noise threshold and removed or corrected. Denoising improves the reliability of the data and can be analyzed and judged based on more accurate data information, thereby improving the accuracy of charging scheduling.

[0093] Use filtering algorithms to denoise the data, such as using a low-pass filter (such as Kalman filtering or mean filtering) to smooth random fluctuations in the data. For obviously abnormal data points, detection is performed based on a preset threshold. When a data value exceeds the set range, it is considered abnormal noise. For abnormal values, correction is performed through median interpolation or the average of the previous and next values ​​to eliminate isolated abnormal peaks.

[0094] Integrity check: Check the integrity of pre-processed data to ensure that there are no omissions or anomalies in the data. Data collection may be affected by factors such as network fluctuations and equipment failures, resulting in incomplete or missing data. Through integrity check, incomplete data can be identified and eliminated, or missing data can be filled in through interpolation and supplementation. For example, after detecting that traffic flow data in certain time periods are missing, it can be supplemented based on the data of the previous and next time periods to ensure the consistency and integrity of the entire data set.

[0095] By setting up the integrity check module, each data set is checked item by item to ensure that the data fields are complete and not missing. Specifically, it includes checking whether each data contains all necessary fields, such as timestamp, data source, data value, etc., and whether the data collection frequency meets the target requirements.

[0096] For missing data, interpolation methods such as linear interpolation and spline interpolation are used to complete the data. For example, if there is a lack of grid load data in a certain period of time, interpolation prediction can be performed based on the load values ​​at the previous and next time points.

[0097] In a preferred embodiment of the present invention, S300, the pre-processed data is classified according to the information source, and the same type of data is synchronized with a unified timestamp to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set, including:

[0098] Classify the preprocessed data according to the type of information source to obtain a preliminary classified data set;

[0099] Perform unified timestamp synchronization processing on data from the same information source in the preliminary classification data set to generate a synchronized data set;

[0100] According to the category of information source, the synchronized data set is divided into vehicle battery status data set, charging pile status data set, power grid load data set, traffic flow data set and user demand data set to form a preliminary fusion data set.

[0101] In an embodiment of the present invention, after preprocessing the acquired raw data, the system classifies the data according to different information sources, and synchronizes the timestamps of similar data, thereby generating a preliminary fused data set. The preliminary fused data set includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set. This step not only reduces the computational burden caused by data mixing by classifying different types of data sources, but also allows targeted analysis of specific categories of data. For example, in the combined analysis of the traffic flow data set and the user demand data set, the system can effectively analyze the urgency of the current charging request.

[0102] During the timestamp synchronization process, the timestamps of the same type of data are aligned, so that data from different sources are consistent in time, ensuring that the system can analyze and process data from multiple sources in real time at the same time. Through this classification and synchronization method, the system can respond to charging needs more flexibly and quickly, making the entire charging network have higher real-time and data response capabilities, and can quickly obtain current multi-source information when facing sudden charging peaks, ensuring the reasonable allocation of charging resources.

[0103] In a preferred embodiment of the present invention, S400, performing association analysis on similar data in the preliminary fused data set, removing redundant information and screening, and obtaining multiple streamlined data sets, including:

[0104] Perform association analysis on similar data in the preliminary fusion data set to determine the degree of association between similar data and obtain a preliminary association data set. The formula is as follows:

[0105] ,

[0106] in, For similar data and The correlation between For the The weight of the category, For data In the Normalized values ​​on categories, For data In the Normalized values ​​on categories, is the number of categories;

[0107] According to a preset correlation threshold, redundant data is removed from the preliminary correlation data set to obtain a de-redundant data set;

[0108] The key data of the de-redundant data set is screened according to the preset timestamp threshold, and the data information valuable for charging network optimization is extracted to generate a streamlined data set.

[0109] In an embodiment of the present invention, based on the preliminary fusion data set, the same type of data in the data set is deduplicated through association analysis, and the streamlined data is screened. Association analysis can identify the correlation between data through a specific algorithm, remove duplicate content of the same type of data, retain only data with strong correlation, and generate a preliminary correlation data set. This step not only improves the data processing efficiency, but also reduces the resources required to store redundant data, and improves the quality and accuracy of data processing. For example, in the process of deduplication of the vehicle battery status data set and the charging pile status data set, the system can delete those data that are less correlated with the current battery status based on the changing trend of the battery status, and retain information that is more valuable for decision-making of the current charging network.

[0110] The preliminary association data set generated by association analysis further improves the accuracy of the data. On this basis, the useless information in the redundant data set is removed through threshold screening to ensure that only useful information is left for optimizing the allocation of charging network resources. This streamlined data processing step not only reduces the computational burden, but also enables the system to respond faster when facing changing charging needs, ensuring the optimal configuration of data resources and laying a good foundation for subsequent cross-class data association.

[0111] Among them, the preset correlation threshold and the preset timestamp threshold are both for improving data utilization, and their values ​​are generally set based on historical data.

[0112] In a preferred embodiment of the present invention, S500, by analyzing the potential relationship between the vehicle battery status, the charging pile utilization rate, the traffic flow and the grid load, the simplified data set is cross-classified and associated to generate a multi-dimensional fusion data set, including:

[0113] Perform cross-class correlation analysis on different data sets in the streamlined data set, analyze the correlation between vehicle battery status, charging pile utilization, traffic flow and power grid load, and obtain a correlation information data set;

[0114] According to the associated information in the associated information dataset, low-value cross-category data in the streamlined dataset is eliminated to obtain a cross-category screening dataset;

[0115] The cross-class screening dataset and the associated information dataset are fused in multiple dimensions to generate a multi-dimensional fused dataset.

[0116] In an embodiment of the present invention, the streamlined data generates a multi-dimensional fusion data set through cross-class association analysis. This step analyzes cross-class data of vehicle battery status, charging pile utilization, traffic flow, and grid load to identify the potential relationship between various types of data, thereby improving the ability in data mining. For example, in the association analysis process, by identifying the relationship between the remaining battery power and the charging pile utilization, the trend of charging demand can be estimated and resource scheduling can be done in advance. In addition, by identifying the fluctuation relationship between grid load and traffic flow, future charging demand peaks can be predicted, and scheduling strategies can be formulated based on these data.

[0117] In practical applications, this cross-class analysis method can effectively reduce the power allocation deviation of the charging network under high load. For example, during peak traffic flow, the charging power allocation of low-priority requests can be reduced in advance by identifying potential correlations between related data, ensuring that the charging needs of high-priority requests are met first. This multi-dimensional data fusion method breaks through the limitations of traditional data and can achieve accurate data analysis and scheduling in complex charging environments.

[0118] Among them, according to the associated information in the associated information data set, the low-value cross-class data in the streamlined data set is eliminated to obtain the cross-class screening data set. The correlation threshold can be set. When the correlation of the cross-class data is lower than the threshold, it is regarded as data with little impact on system optimization, and these low-correlation data are eliminated. Through screening, redundant cross-class data can be removed, and only those data with significant impact in the current environment are retained. For example, if the correlation between traffic flow and charging pile status data is low during a certain period of time, the system can ignore the traffic flow data during that period to reduce the amount of data processing.

[0119] After filtering out irrelevant data, the remaining highly relevant data is reintegrated to form a cross-category screening data set. This data set retains data that is closely related to charging decisions, providing more targeted data support for subsequent multi-dimensional data fusion. Through this process, we can focus on data that is truly valuable for charging network optimization, reduce data redundancy, and improve data processing efficiency and decision accuracy.

[0120] In a preferred embodiment of the present invention, cross-category association analysis is performed on different data sets in the simplified data set to analyze the association between vehicle battery status, charging pile utilization, traffic flow and power grid load, and obtain an associated information data set, including:

[0121] Extract vehicle battery status data, charging station utilization data, traffic flow data, and grid load data from the streamlined data set;

[0122] According to the remaining battery power of the vehicle and the current utilization rate of the charging pile, the urgency of the demand for charging resources for different charging requests is calculated to obtain the battery-charging pile association information;

[0123] Match the grid load data with the charging pile utilization data, compare the grid load fluctuation and the charging pile power consumption trend, and generate grid-charging pile load correlation information;

[0124] Analyze the impact of traffic flow on each charging request based on traffic flow data and vehicle battery status data, and generate traffic-charging priority association information;

[0125] The battery-charging pile correlation information, the power grid-charging pile load correlation information, and the traffic-charging priority correlation information are merged to generate a correlation information data set.

[0126] In the embodiment of the present invention, by performing cross-class correlation analysis on the streamlined data set, and comprehensively considering multiple key factors such as vehicle battery status, charging pile utilization, traffic flow and grid load, a comprehensive evaluation of the urgency, load pressure and priority of charging requests in the smart charging network is achieved. Through these correlation analyses, the system can dynamically perceive the potential relationship between different data, more accurately grasp the current load situation of the charging network and the urgency of charging requests, thereby providing a more accurate basis for resource scheduling and priority allocation.

[0127] In traditional charging networks, due to the limitations of a single data source or independent data processing, the system is often unable to fully consider the interaction of various factors, resulting in inefficient resource allocation during peak loads. The present invention achieves deep data fusion through multi-dimensional cross-class data analysis, so that the system can give priority to emergency charging requests when the load is high, and can flexibly arrange other requests when the load is low, thereby optimizing the overall resource allocation strategy. This correlation analysis not only improves the response speed of the system, but also makes resource allocation more reasonable and efficient.

[0128] Specifically, according to the remaining battery power of the vehicle and the current utilization rate of the charging pile, the urgency of the demand for charging resources of different charging requests is calculated to obtain the battery-charging pile association information. The urgency calculation formula can be applied to obtain the association through the weighted calculation of the remaining battery power Er of the vehicle and the utilization rate of the charging pile Uc. Use the formula:

[0129] ,

[0130] in, Indicates the urgency of the battery-charging pile association information, and are weight coefficients, which respectively control the impact of battery status and charging pile utilization on urgency. and The calculation is performed for each charging request and the associated information is generated, providing urgency data basis for subsequent priority allocation.

[0131] Specifically, the grid load data is matched with the charging pile utilization data, and the grid load fluctuation is compared with the charging pile power consumption trend to generate grid-charging pile load correlation information. The grid load L and charging pile utilization rate Uc data can be synchronized in time, and the sliding window method can be used to analyze load changes. Use the correlation formula:

[0132] ,

[0133] in, Indicates the grid-charging pile load association information, and To adjust the parameters to correlate load fluctuations with utilization trends, is the current grid load, is the current utilization rate of the charging pile. The generated load association information is stored in the association data set to provide a reference for subsequent load balancing and power allocation strategies.

[0134] Specifically, based on the traffic flow data and the vehicle battery status data, the impact of traffic flow on each charging request is analyzed to generate traffic-charging priority association information. The priority impact formula can be used:

[0135] ,

[0136] in, Indicates the traffic-charging priority association information, and To adjust the parameters, The system calculates the priority association information of each charging request under the current traffic conditions to ensure that charging services are provided to emergency requests first during high-demand periods.

[0137] In a preferred embodiment of the present invention, S600, scoring the user's charging time requirements according to the multi-dimensional fusion data set to form charging requests of different priorities, including a high priority charging request, a medium priority charging request and a low priority charging request, including:

[0138] Score the user's charging time requirements based on the multi-dimensional fusion data set to obtain a priority score. The formula is as follows:

[0139] ,

[0140] in, is the priority score of charging request i, indicating the relative urgency of the charging request. is a constant, which acts as a baseline adjustment factor and affects the scaling of the entire priority score. It can be regarded as a fixed multiplier used to enlarge or reduce the score. Indicates the urgency impact of the charging request, The time requirement for charging request. The smaller the value, the more urgent the request. It usually refers to the time when the user expects charging to be completed. is a time-related attenuation factor that controls the degree of impact of urgency on priority. A value of 0 means that the longer the request time, the smaller the impact of urgency. The item affected by insufficient power. The current remaining power of the vehicle battery, indicating the urgency of the charging status. The maximum capacity of the vehicle battery, indicating the maximum amount of electricity that the battery can store. Indicates the battery charging status ratio. The larger the value, the more sufficient the battery is, and the smaller the insufficient battery is. is the weight of the grid load, which controls the influence of the grid load on the priority. is the current grid load status. The higher the value, the heavier the grid load. The square of this factor increases the nonlinear effect. is the interaction term between charging pile utilization and traffic flow, The current utilization rate of the charging pile indicates the usage of the charging pile. The higher the value, the higher the usage frequency. is the current traffic flow, indicating the traffic conditions around the charging station. A higher value indicates heavier traffic. is the adjustment coefficient, which is used to balance the impact of the interaction term on the priority score and the grid load The higher it is, the more this term affects the contribution of the interaction term. is the time period adjustment item for traffic flow, It is a regulation parameter related to the grid load, which affects the contribution of traffic flow to the priority score. This item indicates that when the grid load is high, the impact of traffic flow on the priority score will be weakened;

[0141] The priority scores of different charging requests are divided according to a preset scoring threshold range to obtain a high priority charging request, a medium priority charging request and a low priority charging request;

[0142] Data with different priorities are labeled and classified into corresponding charging request types to generate a hierarchical dataset.

[0143] In an embodiment of the present invention, based on a multi-dimensional fusion data set, priorities are set for user charging requests through scoring, forming three types of charging requests: high, medium, and low. The priority scoring formula comprehensively considers the user's charging time requirements, the remaining battery power of the vehicle, the grid load, the current utilization rate of the charging pile, and the traffic flow. Through the calculation of these parameters, the system can accurately determine the urgency and resource occupancy of the charging request. During periods of high grid load, the priority scoring formula can automatically increase the weight of high-priority requests, so that urgent charging needs have priority in obtaining charging resources.

[0144] In a preferred embodiment of the present invention, S700, charging power and time periods are allocated to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme, including:

[0145] According to the priorities of different charging requests, the corresponding allocated power is determined, and a power allocation data set is formed according to the allocated power; the calculation formula of the allocated power is:

[0146] ,

[0147] in, is the allocated power for charging request i, is the maximum charging power available in the current power grid, For the The priority score of each charging request, is the nonlinear coefficient, To prevent division by zero, The charging time requested by the user, is the maximum charging time for all requests in the system, is the weight coefficient, is a nonlinear parameter.

[0148] According to the priorities of different charging requests, the corresponding time period allocation length is determined, and according to the time period allocation length, a time period allocation data set is formed; the calculation formula of the time period allocation length is:

[0149] ,

[0150] in, Assign a length to the time period of charging request i, Score the priority of the current charging request i, is the influence coefficient, For the The power requirement of a charging request, Nonlinear coefficient;

[0151] According to the power allocation data set and the time period allocation data set, an optimized scheduling data set is generated, including a charging power allocation plan and a charging time period allocation plan.

[0152] In the embodiment of the present invention, after obtaining charging requests of different priorities, charging power and time period are allocated to each priority request, thereby forming an optimized scheduling data set. Power allocation is based on priority score and grid load status, ensuring that high-priority requests receive higher power allocation during high-load periods.

[0153] When calculating the allocated power for each charging request, a nonlinear allocation formula is used to dynamically adjust the allocated power as the priority score changes. Through this formula, high-priority requests can obtain higher allocated power under high load conditions, while low-priority requests are moderately reduced in power allocation when the load increases. For example, during peak charging hours, vehicles with high-priority requests can obtain higher power in a short period of time to achieve fast charging, while the power of low-priority requests is automatically adjusted to adapt to the grid load to avoid affecting the stability of the grid due to excessive load. This power allocation method is more flexible in resource allocation, and can give priority to emergency charging needs under high load conditions, while reducing system instability caused by unreasonable power allocation.

[0154] When allocating the time period for each charging request, the nonlinear coefficient Control makes it possible to more flexibly meet requests of different priorities during low load periods. This time period allocation based on dynamic load ensures the rational use of charging resources and improves the load adaptability of the system.

[0155] In a preferred embodiment of the present invention, S800, conflict detection is performed on the optimized scheduling data set, potential conflicts that lead to excessive concentration of charging resources or excessive load are identified, and the optimized scheduling data set is adjusted according to the potential conflicts to generate an optimized control data set, including:

[0156] Conflict detection on the optimized dispatch data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load;

[0157] Perform data analysis on potential conflicts to obtain conflict severity scores, and adjust the allocation scheme in the corresponding optimized scheduling data set according to the conflict severity scores to generate an optimized control data set;

[0158] The conflict severity score is calculated using the following formula:

[0159] ,

[0160] in, Score the conflict severity of the charging request, which is used to indicate the total amount of load overload. is the total number of time periods for evaluation, For time period The total charging power requirement in is the maximum load power allowed by the power grid in each time period, For calculating time period If the limit is not exceeded, it will be counted as 0.

[0161] In an embodiment of the present invention, in order to ensure the rational use of charging resources and avoid grid pressure caused by excessive concentration of resources, conflict detection is performed on the basis of the optimized scheduling data set to identify potential resource conflicts. The conflict detection analyzes the total amount of charging power demand in the time period to determine whether there is a situation that exceeds the load upper limit, and scores the identified over-limit part. After completing the conflict detection, the system further adjusts the optimized scheduling data set according to the conflict detection results and generates an optimized control data set to avoid resource conflicts during high-load periods. The conflict severity score of the high-load period is analyzed, and the power and time period allocation of the charging request are adjusted according to the over-limit score. For conflict time periods with higher scores, the system gives priority to high-priority requests, reduces the charging power of low-priority requests and appropriately postpones them, thereby reducing the concentration of resource allocation.

[0162] In practical applications, for example, in a charging network during a peak period, if it is detected that the charging power allocation exceeds the system load limit, the system will reallocate low-priority requests to extend their charging time period while reducing their instantaneous power to ensure the stability of system operation under load conditions. This adjustment mechanism can intelligently allocate the power and time of requests of different priorities, dynamically adapt to real-time load changes, achieve efficient resource allocation, and provide more comprehensive guarantees for the efficient operation of the smart charging network.

[0163] Through the adjustment and control of the above steps, resource conflicts and overloads can be effectively avoided in complex scenarios with high loads, so that the entire charging network can achieve optimal resource allocation while meeting user needs, thereby improving the reliability and response speed of the system.

[0164] An embodiment of the present invention further provides a multi-source information fusion system in a smart charging network, the system comprising:

[0165] A data acquisition module, used to acquire raw data related to the charging network from multiple information sources, the information sources including vehicle battery status information, charging pile status information, grid load information, traffic flow information and user charging request information;

[0166] A preprocessing module is used to format, remove noise and verify data integrity of the acquired raw data to obtain preprocessed data;

[0167] The preliminary fusion module is used to classify the pre-processed data according to the information source, synchronize the timestamps of the same type of data, and generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set;

[0168] The data reduction module is used to perform correlation analysis on similar data in the preliminary fusion data set, remove redundant information and perform screening to obtain multiple reduced data sets;

[0169] The data fusion module is used to generate a multi-dimensional fused data set by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and grid load, and performing cross-class data association on the streamlined data set;

[0170] The data screening module is used to remove redundant and useless data from the multi-dimensional fusion data set and generate a key data set based on the real-time status of the charging network and user charging requests;

[0171] The request classification module is used to classify the information in the key data set according to the user's charging time requirements, the remaining battery power of the vehicle, and the current grid load, and form different levels of charging requests, including high priority charging requests, medium priority charging requests, and low priority charging requests;

[0172] A scheme allocation module is used to allocate charging power and time periods to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme;

[0173] The conflict optimization module is used to perform conflict detection on the optimized scheduling data set, identify potential conflicts that lead to excessive concentration of charging resources or excessive load, and adjust the optimized scheduling data set according to the potential conflicts to generate an optimized control data set.

[0174] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0175] The embodiment of the present invention further provides a computing device, including: a processor, a memory storing a computer program, and when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0176] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the above method. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0177] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-source information fusion method in a smart charging network, characterized in that: The method comprises: S100, acquiring raw data related to the charging network from multiple information sources, wherein the information sources include vehicle battery status information, charging pile status information, grid load information, traffic flow information, and user charging request information; S200, formatting, denoising and data integrity checking the acquired raw data to obtain pre-processed data; S300, classifying the preprocessed data according to the information source, and performing unified timestamp synchronization on the same type of data to generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set; S400, performing association analysis on similar data in the preliminary fused data set, removing redundant information and screening, and obtaining multiple streamlined data sets; S500, by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and power grid load, cross-class data association is performed on the streamlined data set to generate a multi-dimensional fusion data set; S600, scoring the user's charging time requirements according to the multi-dimensional fusion data set to form charging requests of different priorities, including a high priority charging request, a medium priority charging request, and a low priority charging request; S700, allocating charging power and time periods for charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme; S800, performing conflict detection on the optimized scheduling data set, identifying potential conflicts that may lead to excessive concentration of charging resources or excessive load, and adjusting the optimized scheduling data set according to the potential conflicts to generate an optimized control data set; Perform association analysis on similar data in the preliminary fusion data set, remove redundant information and filter, and obtain multiple streamlined data sets, including: Perform association analysis on similar data in the preliminary fusion data set to determine the degree of association between similar data and obtain a preliminary association data set. The formula is as follows: ， in, For similar data and The correlation between For the The weight of the category, For data In the Normalized values ​​on categories, For data In the Normalized values ​​on categories, is the number of categories; According to a preset correlation threshold, redundant data is removed from the preliminary correlation data set to obtain a de-redundant data set; Screen the key data of the de-redundant data set according to the preset timestamp threshold to generate a streamlined data set; Conflict detection is performed on the optimized scheduling data set to identify potential conflicts that may lead to excessive concentration of charging resources or excessive load. The optimized scheduling data set is adjusted according to the potential conflicts to generate an optimized control data set, including: Conflict detection on the optimized dispatch data set to identify potential conflicts that lead to excessive concentration of charging resources or excessive load; Perform data analysis on potential conflicts to obtain conflict severity scores, and adjust the allocation scheme in the corresponding optimized scheduling data set according to the conflict severity scores to generate an optimized control data set; The conflict severity score is calculated using the following formula: ， in, Score the conflict severity of the charging request, which is used to indicate the total amount of load overload. is the total number of time periods for evaluation, For time period The total charging power requirement in is the maximum load power allowed by the power grid in each time period, For calculating time period If the limit is not exceeded, it will be counted as 0.

2. The method for multi-source information fusion in a smart charging network according to claim 1, characterized in that: The pre-processed data is classified according to the information source, and the same type of data is synchronized with a unified timestamp to generate a preliminary fusion data set. The preliminary fusion data set includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set, including: Standardize the data format of raw data from different information sources to obtain formatted data; The noise data in the formatted data is cleared according to a preset noise threshold to obtain denoised data; The denoised data is verified for data integrity, and abnormal data is removed to obtain preprocessed data.

3. The multi-source information fusion method in a smart charging network according to claim 2, characterized in that: The pre-processed data is classified according to the information source, and the same type of data is synchronized with a unified timestamp to generate a preliminary fusion data set. The preliminary fusion data set includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set, including: Classify the preprocessed data according to the type of information source to obtain a preliminary classified data set; Perform unified timestamp synchronization processing on data from the same information source in the preliminary classification data set to generate a synchronized data set; According to the category of information source, the synchronized data set is divided into vehicle battery status data set, charging pile status data set, power grid load data set, traffic flow data set and user demand data set to form a preliminary fusion data set.

4. The multi-source information fusion method in a smart charging network according to claim 3, characterized in that: By analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and grid load, the simplified data set is cross-classified and associated to generate a multi-dimensional fusion data set, including: Perform cross-class correlation analysis on different data sets in the streamlined data set, analyze the correlation between vehicle battery status, charging pile utilization, traffic flow and power grid load, and obtain a correlation information data set; According to the associated information in the associated information dataset, low-value cross-category data in the streamlined dataset is eliminated to obtain a cross-category screening dataset; The cross-class screening dataset and the associated information dataset are fused in multiple dimensions to generate a multi-dimensional fused dataset.

5. The method for multi-source information fusion in a smart charging network according to claim 4, characterized in that: Perform cross-class correlation analysis on different data sets in the streamlined data set, analyze the correlation between vehicle battery status, charging pile utilization, traffic flow and power grid load, and obtain the correlation information data set, including: Extract vehicle battery status data, charging station utilization data, traffic flow data, and grid load data from the streamlined data set; According to the remaining battery power of the vehicle and the current utilization rate of the charging pile, the urgency of the demand for charging resources for different charging requests is calculated to obtain the battery-charging pile association information; Match the grid load data with the charging pile utilization data, compare the grid load fluctuation and the charging pile power consumption trend, and generate grid-charging pile load correlation information; Analyze the impact of traffic flow on each charging request based on traffic flow data and vehicle battery status data, and generate traffic-charging priority association information; The battery-charging pile correlation information, the power grid-charging pile load correlation information, and the traffic-charging priority correlation information are merged to generate a correlation information data set.

6. The method for multi-source information fusion in a smart charging network according to claim 5, characterized in that: The user's charging time requirements are scored based on the multi-dimensional fusion data set to form charging requests of different priorities, including high priority charging requests, medium priority charging requests and low priority charging requests, including: Score the user's charging time requirements based on the multi-dimensional fusion data set to obtain a priority score. The formula is as follows: ， in, Score the priority of charging request i, is the adjustment factor, The time requirement for charging request, is the time-dependent attenuation factor, is the current remaining power of the vehicle battery, is the maximum capacity of the vehicle battery, is the weight of the grid load, is the value of the current grid load, is the current utilization rate of the charging pile, is the current traffic flow, is the adjustment coefficient, is a regulation parameter related to the grid load; The priority scores of different charging requests are divided according to a preset scoring threshold range to obtain a high priority charging request, a medium priority charging request and a low priority charging request; Data with different priorities are labeled and classified into corresponding charging request types to generate a hierarchical dataset.

7. The method for multi-source information fusion in a smart charging network according to claim 6, characterized in that: Charging power and time periods are allocated to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme, including: According to the priorities of different charging requests, the corresponding allocated power is determined, and a power allocation data set is formed according to the allocated power; the calculation formula of the allocated power is: ， in, is the allocated power for charging request i, is the maximum charging power available in the current power grid, For the The priority score of each charging request, is the nonlinear coefficient, To prevent division by zero, The charging time requested by the user, is the maximum charging time for all requests in the system, is the weight coefficient, is a nonlinear parameter; According to the priorities of different charging requests, the corresponding time period allocation length is determined, and according to the time period allocation length, a time period allocation data set is formed; the calculation formula of the time period allocation length is: ， in, Assign a length to the time period of charging request i, Score the priority of the current charging request i, is the influence coefficient, For the The power requirement of a charging request, Nonlinear coefficient; According to the power allocation data set and the time period allocation data set, an optimized scheduling data set is generated, including a charging power allocation plan and a charging time period allocation plan.

8. A multi-source information fusion system in a smart charging network, characterized in that: Applied to the method according to any one of claims 1 to 7, the system comprises: A data acquisition module, used to acquire raw data related to the charging network from multiple information sources, the information sources including vehicle battery status information, charging pile status information, grid load information, traffic flow information and user charging request information; A preprocessing module is used to format, remove noise and verify data integrity of the acquired raw data to obtain preprocessed data; The preliminary fusion module is used to classify the pre-processed data according to the information source, synchronize the timestamps of the same type of data, and generate a preliminary fusion data set, which includes a vehicle battery status data set, a charging pile status data set, a power grid load data set, a traffic flow data set, and a user demand data set; The data reduction module is used to perform correlation analysis on similar data in the preliminary fusion data set, remove redundant information and perform screening to obtain multiple reduced data sets; The data fusion module is used to generate a multi-dimensional fused data set by analyzing the potential relationship between vehicle battery status, charging pile utilization, traffic flow and grid load, and performing cross-class data association on the streamlined data set; The data screening module is used to remove redundant and useless data from the multi-dimensional fusion data set and generate a key data set based on the real-time status of the charging network and user charging requests; The request classification module is used to classify the information in the key data set according to the user's charging time requirements, the remaining battery power of the vehicle, and the current grid load to form different levels of charging requests, including high priority charging requests, medium priority charging requests, and low priority charging requests; A scheme allocation module is used to allocate charging power and time periods to charging requests of different priorities to obtain an optimized scheduling data set, wherein the optimized scheduling data set includes a charging power allocation scheme and a charging time period allocation scheme; The conflict optimization module is used to perform conflict detection on the optimized scheduling data set, identify potential conflicts that lead to excessive concentration of charging resources or excessive load, and adjust the optimized scheduling data set according to the potential conflicts to generate an optimized control data set.

Images