Adaptive control method for electric vehicle groups participating in power grid peak load regulation

Abstract

The present invention discloses an adaptive control method for electric vehicle groups to participate in power grid peak regulation, which relates to the field of distribution network control. The method includes: obtaining a first electric vehicle; determining whether a first historical charging divergence is within a preset divergence threshold; if not, analyzing a first driving chain to obtain a first driving block set; obtaining a first electric vehicle cluster with a first departure block as a constraint; obtaining a second electric vehicle; predicting a first passing block set and a first destination block to obtain a first predicted charging characteristic, predicting a second passing block set and a second destination block to obtain a second predicted charging characteristic; and performing power grid peak regulation on a predetermined power grid area. The method solves the technical problems of inaccurate prediction of electric vehicle charging time and location and unsatisfactory peak regulation effect in the existing electric vehicle participation in power grid peak regulation control, and achieves the technical effect of improving power grid peak regulation efficiency through accurate prediction and adaptive scheduling of electric vehicle charging demand.

Description

Technical Field

[0001] The present application relates to the field of distribution network control, and in particular to an adaptive control method for electric vehicle groups participating in power grid peak regulation. Background Art

[0002] In recent years, with the continuous advancement of electric vehicle technology, the number of electric vehicles has shown a rapid growth trend. A large number of electric vehicles connected to the power grid for charging will have an important impact on the load and scheduling of the power grid. If not reasonably controlled and scheduled, the centralized charging behavior of electric vehicles may lead to a sharp increase in the load of the power grid and even cause power grid failure. Therefore, controlling the participation of electric vehicle groups in power grid peak regulation is an important measure to ensure the safe operation of the power grid. In the current power grid peak regulation strategy, electric vehicles are often dispatched as a whole, ignoring the differences in charging requirements between different electric vehicles, and the prediction of electric vehicle charging requirements is not accurate enough, resulting in unsatisfactory power grid peak regulation effect.

[0003] In the current relevant technologies, electric vehicles participating in grid peak load control have technical problems such as inaccurate prediction of electric vehicle charging time and location, and unsatisfactory peak load effect. Summary of the invention

[0004] The present application provides an adaptive control method for electric vehicle groups to participate in power grid peak regulation, which adopts the method of obtaining a first electric vehicle and judging whether its historical charging divergence is within a preset divergence threshold. If it is not within the preset divergence threshold, the driving chain of the first electric vehicle is analyzed in combination with the power grid block set to obtain a driving block set, and the electric vehicle cluster is obtained with the departure block as the clustering constraint, and the driving chain information of other electric vehicles in the cluster is obtained, and the charging prediction strategy is obtained. The predicted analysis is performed on the passing blocks and the destination blocks in the driving block set to obtain the predicted charging characteristics, and the predicted charging characteristics are combined to perform power grid peak regulation on the predetermined power grid area, and other technical means are used to achieve the technical effect of improving the power grid peak regulation efficiency through accurate prediction and adaptive scheduling of electric vehicle charging demand.

[0005] The present application provides an adaptive control method for electric vehicle groups participating in power grid peak regulation, including:

[0006] A first electric vehicle is obtained, where the first electric vehicle refers to any one of a predetermined electric vehicle group in a predetermined power grid area; whether a first historical charging divergence of the first electric vehicle is within a preset divergence threshold is determined; if not, a first driving chain of the first electric vehicle is analyzed in combination with a power grid block set obtained by dividing the predetermined power grid area to obtain a first driving block set, wherein the first driving block set includes a first departure block, a first passing block set, and a first destination block; a first electric vehicle cluster is obtained with the first departure block as a clustering constraint, wherein the first electric vehicle cluster refers to a clustering result of electric vehicles in a plurality of driving chains whose departure blocks are the first departure block; and the first A second electric vehicle in an electric vehicle cluster, wherein the second electric vehicle corresponds to a second driving chain, and the second driving chain corresponds to a second driving block set, wherein the second driving block set includes the first departure block, a second passing block set and a second destination block; a charging prediction strategy is obtained, and according to the charging prediction strategy, a prediction analysis is performed on the first passing block set and the first destination block to obtain a first predicted charging feature, and a prediction analysis is performed on the second passing block set and the second destination block to obtain a second predicted charging feature; the first predicted charging feature and the second predicted charging feature constitute a predicted charging feature, and the predicted charging feature is combined to perform grid peak regulation on the predetermined grid area.

[0007] In a possible implementation, the following processing is performed:

[0008] Analyze a first historical charge and discharge record of the first electric vehicle to obtain a first charging characteristic data time series; extract first charging characteristic data in the first charging characteristic data time series, the first charging characteristic data including a first charging time and a first charging block; obtain a first historical charging scatter plot of the first electric vehicle based on the first charging time and the first charging block; perform spline curve fitting on the first historical charging scatter plot to obtain a first fitting curve, and analyze the first fitting curve to obtain the first historical charging divergence.

[0009] In a possible implementation, the following processing is performed:

[0010] If so, the first charging characteristic data time series of the first electric vehicle is analyzed to obtain a first charging characteristic, and grid peak regulation is performed on the predetermined grid area according to the first charging characteristic.

[0011] In a possible implementation, the following processing is performed:

[0012] Obtaining predetermined dimensional indicators, the predetermined dimensional indicators including real-time power and the number of blocks passed; collecting dimensional information of the first electric vehicle and the second electric vehicle in turn according to the predetermined dimensional indicators, and obtaining first indicator information and second indicator information respectively; performing secondary clustering on the first electric vehicle and the second electric vehicle according to the first indicator information and the second indicator information, and obtaining secondary clustering results; obtaining a target clustering cluster in the secondary clustering results, and predicting a target predicted charging feature of the target clustering cluster; and using the target predicted charging feature as the predicted charging feature.

[0013] In a possible implementation, the following processing is performed:

[0014] Determine whether the first real-time power in the first indicator information is greater than 30%; if so, add the first electric vehicle to the first category; determine whether the first real-time power meets the driving requirement of a first number of passing blocks; if so, keep the first electric vehicle in the first category; if not, add the first electric vehicle to the second category; and use the second category as the target cluster.

[0015] In a possible implementation, the following processing is performed:

[0016] Determining whether the first real-time power level in the first indicator information is greater than 30%, also includes: if it is less than or equal to, adding the first electric vehicle to the second category.

[0017] In a possible implementation, the following processing is performed:

[0018] A third electric vehicle in the first category is randomly obtained, and a third driving chain of the third electric vehicle corresponds to a third driving block set; a third destination block in the third driving block set is used as a third predicted charging block of the third electric vehicle, and a time for the third electric vehicle to travel to the third destination block is used as a third predicted charging time of the third electric vehicle; a third predicted charging feature of the third electric vehicle is established according to the third predicted charging block and the third predicted charging time; and the third predicted charging feature is added to the predicted charging feature.

[0019] In a possible implementation, the following processing is performed:

[0020] The grid block set includes a plurality of grid blocks with identifications of whether or not they can be charged; a first route block in the first route block set is extracted according to the charging prediction strategy; when the first route block has an identification of being chargeable, multi-dimensional features of the first route block are collected to obtain a first comprehensive charging index; the first comprehensive charging index is used as a first predicted probability that the first electric vehicle will be charged in the first route block; the first predicted probability is analyzed to obtain a target route block that meets the charging prediction constraints; the target route block is used as a first predicted charging block for the first electric vehicle, and the time it takes the first electric vehicle to travel to the target route block is used as a first predicted charging time for the first electric vehicle; and the first predicted charging feature is formed according to the first predicted charging block and the first predicted charging time.

[0021] The adaptive control method for electric vehicle groups participating in power grid peak load regulation proposed in the present application first obtains a first electric vehicle, the first electric vehicle refers to any electric vehicle in a predetermined electric vehicle group in a predetermined power grid area, and then determines whether the first historical charging divergence of the first electric vehicle is within a preset divergence threshold. If not, the first driving chain of the first electric vehicle is analyzed in combination with the power grid block set obtained by dividing the predetermined power grid area to obtain a first driving block set, wherein the first driving block set includes a first departure block, a first passing block set and a first destination block, and then the first electric vehicle clustering cluster is obtained with the first departure block as a clustering constraint, the first electric vehicle clustering cluster refers to the clustering result of electric vehicles whose departure block is the first departure block in multiple driving chains, and then obtains Take the second electric vehicle in the first electric vehicle cluster, the second electric vehicle corresponds to the second driving chain, and the second driving chain corresponds to the second driving block set, wherein the second driving block set includes the first departure block, the second passing block set and the second destination block, then obtain the charging prediction strategy, and perform prediction analysis on the first passing block set and the first destination block according to the charging prediction strategy to obtain the first predicted charging feature, perform prediction analysis on the second passing block set and the second destination block to obtain the second predicted charging feature, the first predicted charging feature and the second predicted charging feature constitute the predicted charging feature, and combine the predicted charging feature to perform power grid peak regulation on the predetermined power grid area, thereby achieving the technical effect of improving the power grid peak regulation efficiency through accurate prediction and adaptive scheduling of electric vehicle charging demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the method according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0023] Figure 1 A schematic flow chart of an adaptive control method for a group of electric vehicles participating in grid peak load regulation provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of a process for obtaining a first historical charging divergence in an adaptive control method for an electric vehicle group participating in grid peak regulation provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0026] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0027] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0028] The embodiment of the present application provides an adaptive control method for electric vehicle groups to participate in power grid peak regulation, such as Figure 1 As shown, the method includes:

[0029] Step S100, obtaining a first electric vehicle, wherein the first electric vehicle refers to any electric vehicle in a predetermined electric vehicle group in a predetermined power grid area. Specifically, an electric vehicle in a predetermined electric vehicle group in a predetermined power grid area is identified and obtained as the first electric vehicle. The predetermined electric vehicle group is a group of electric vehicles located in a predetermined power grid area that is defined and monitored in advance. The predetermined power grid area is a pre-set geographical area in which a plurality of electric vehicles are included in the monitoring range to consider their power grid support capabilities.

[0030] Step S200, determining whether the first historical charging divergence of the first electric vehicle is within a preset divergence threshold. Specifically, the charging divergence of the first electric vehicle is calculated based on the historical charging records (including charging time, charging location, etc.). The first historical charging divergence is the degree of dispersion of charging behavior calculated based on the historical charging records of the first electric vehicle, and is an indicator reflecting the stability and distribution range of the charging behavior of the electric vehicle. If the calculated historical charging divergence exceeds the preset divergence threshold (used to determine whether the charging divergence reaches a critical value requiring special treatment), it is considered that the charging behavior is relatively dispersed; if it is less than or equal to the threshold, it is considered that the charging behavior is relatively concentrated.

[0031] like Figure 2As shown, in a possible implementation, step S200 further includes step S210, analyzing the first historical charging and discharging record of the first electric vehicle to obtain a first charging characteristic data sequence. Specifically, the charging and discharging record of the first electric vehicle in the past period of time is obtained, including the start time, end time, charging amount and charging location (or power grid block) of each charging. The charging record is sorted into time series data, that is, the first charging characteristic data sequence. Step S220, extracting the first charging characteristic data in the first charging characteristic data sequence, the first charging characteristic data including the first charging time and the first charging block. Specifically, extracting key information from the first charging characteristic data sequence, including the charging time and charging block of each charging. Among them, the charging time refers to the specific time point or time period when the charging behavior occurs. The charging block refers to the geographical location or power grid block where the charging behavior occurs. Step S230, based on the first charging time and the first charging block, a first historical charging scatter plot of the first electric vehicle is obtained. Specifically, using the extracted first charging characteristic data (charging time and charging block), a scatter plot is drawn on a two-dimensional plane. In this scatter plot, each point represents a charging behavior, the horizontal axis represents the charging time, and the vertical axis (or attributes such as color, size, etc.) represents the charging block. Step S240, perform spline curve fitting on the first historical charging scatter plot to obtain a first fitting curve, and analyze the first fitting curve to obtain the first historical charging divergence. Specifically, perform spline curve fitting on the first historical charging scatter plot to obtain a smooth curve (first fitting curve), which reflects the trend of the charging behavior of the first electric vehicle over time. Analyze the divergence of this curve (i.e., the degree of fluctuation or discreteness of the curve), and calculate the degree of deviation of the points on the curve from the average line (or center point) to obtain the first historical charging divergence. This implementation method analyzes the historical charging and discharging records of the first electric vehicle, uses the spline curve fitting method to convert discrete charging data points into a smooth curve, accurately evaluates the charging behavior pattern and trend of the first electric vehicle, and achieves the technical effect of accurately reflecting the changing trend of the charging behavior of the electric vehicle.

[0032] Step S300, if not, analyze the first driving chain of the first electric vehicle in combination with the grid block set obtained by dividing the predetermined grid area to obtain the first driving block set, wherein the first driving block set includes the first departure block, the first passing block set and the first destination block. Specifically, if the historical charging divergence of the first electric vehicle exceeds the preset divergence threshold, further analyze its driving chain, that is, the complete driving path information of the first electric vehicle from departure to destination. Among them, the predetermined grid area is divided into multiple grid blocks, and each block has an identification of whether charging is allowed. By analyzing the driving chain of the first electric vehicle, its departure block, passing block set and destination block are determined.

[0033] In a possible implementation, step S300 further includes step S310, if it is, analyzing the first charging characteristic data sequence of the first electric vehicle to obtain the first charging characteristic, and performing power grid peak shaving on the predetermined power grid area according to the first charging characteristic. Specifically, traverse the first charging characteristic data sequence, and count the number of charging times in each time period and each charging block. Find the charging time period and charging block with the highest frequency as the charging characteristic of the first electric vehicle. Use the first charging characteristic to predict the possible charging demand of the electric vehicle in the future, and formulate a power grid peak shaving strategy in combination with the load situation and peak shaving demand of the predetermined power grid area. For example, if it is predicted that the electric vehicle will be charged during peak hours, the power grid load is adjusted in advance to reduce the pressure during peak hours. This implementation method can accurately predict its future charging demand by analyzing the historical charging data of the first electric vehicle and determining its regular charging behavior (i.e., the first charging characteristic), which helps to more accurately adjust the load during the power grid peak shaving process to meet the power demand during peak hours.

[0034] Step S400, taking the first departure block as a clustering constraint, obtaining a first electric vehicle cluster, wherein the first electric vehicle cluster refers to the clustering result of electric vehicles whose departure block is the first departure block in multiple driving chains. Specifically, all electric vehicles whose departure block is the first departure block are clustered into a set, namely the first electric vehicle cluster.

[0035] Step S500, obtaining a second electric vehicle in the first electric vehicle cluster, wherein the second electric vehicle corresponds to a second driving chain, and the second driving chain corresponds to a second driving block set, wherein the second driving block set includes the first departure block, the second passing block set, and the second destination block. Specifically, an electric vehicle is selected from the first electric vehicle cluster as the second electric vehicle for analysis. The second electric vehicle also has its corresponding driving chain and driving block set.

[0036] Step S600, obtain a charging prediction strategy, and perform a prediction analysis on the first set of passing blocks and the first destination block according to the charging prediction strategy to obtain a first predicted charging feature, and perform a prediction analysis on the second set of passing blocks and the second destination block to obtain a second predicted charging feature. Specifically, obtain a predefined charging prediction strategy, which is a rule for predicting the possibility of an electric vehicle charging at a specific time and place. Perform a prediction analysis on the passing block sets and destination blocks of the first electric vehicle and the second electric vehicle to obtain a first predicted charging feature and a second predicted charging feature, respectively. Among them, the first predicted charging feature is the prediction information about the first electric vehicle charging at a specific time and place. The second predicted charging feature is the prediction information about the second electric vehicle charging at a specific time and place.

[0037] In a possible implementation, step S600 further includes step S610, wherein the grid block set includes a plurality of grid blocks with identifications indicating whether or not charging is possible. Specifically, when dividing a predetermined grid area, the area is divided into a plurality of grid blocks. Each grid block has an identification indicating whether the block has charging facilities (such as charging piles, charging points, etc.). Step S620, extracting the first passing block in the first passing block set according to the charging prediction strategy. Specifically, extracting the first passing block set from the first driving chain of the first electric vehicle. According to the charging prediction strategy, extracting any passing block from the first passing block set as the first passing block. Step S630, when the first passing block has an identification indicating that charging is possible, collecting multi-dimensional features of the first passing block to obtain a first charging comprehensive index. Specifically, checking whether the first passing block has an identification indicating that charging is possible. If so, collecting multi-dimensional features of the block, including the number of charging piles, charging power, other service configurations (such as tea, car wash, etc.), user evaluation, etc. Based on these characteristics, a comprehensive index, namely the first charging comprehensive index, is calculated to represent the overall satisfaction with charging in this block.

[0038] Step S640, using the first charging comprehensive index as the first predicted probability that the first electric vehicle is charged in the first passing block. Specifically, the first charging comprehensive index is used directly or after some conversion as the predicted probability that the first electric vehicle is charged in the first passing block, and this probability indicates the possibility that the electric vehicle is charged in the block. Step S650, analyzing the first predicted probability to obtain a target passing block that meets the charging prediction constraint. Specifically, for each block in the first passing block set (i.e., each passing block), a corresponding predicted probability is calculated according to its charging comprehensive index, and this probability indicates the possibility that the electric vehicle is charged in the block. According to the preset charging prediction constraints (for example, charging time window, charging amount demand, grid load balance, etc.), the blocks that meet these conditions are screened out, wherein the charging prediction constraints are used to ensure that the charging behavior of the electric vehicle will not cause excessive load impact on the grid, while meeting the charging demand of the electric vehicle. After screening out the blocks that meet the constraints, one or more optimal blocks are selected as the target passing blocks. Step S660, taking the target passing block as the first predicted charging block of the first electric vehicle, and taking the time when the first electric vehicle travels to the target passing block as the first predicted charging time of the first electric vehicle. Specifically, determining the target passing block as the predicted charging block of the first electric vehicle. According to the current position, speed and distance to the target passing block of the first electric vehicle, the time when the electric vehicle is expected to arrive at the block is calculated, and this time is the predicted charging time of the first electric vehicle. Step S670, forming the first predicted charging feature according to the first predicted charging block and the first predicted charging time. Specifically, the first predicted charging block and the first predicted charging time are combined to form the predicted charging feature of the first electric vehicle, which includes the expected charging location and time of the electric vehicle, and is an important input information for power grid peak regulation. This implementation method accurately predicts the charging demand and behavior of the electric vehicle by collecting and analyzing multi-dimensional features such as the driving chain of the electric vehicle, the charging facilities and service quality of the passing blocks, thereby accurately predicting the charging location and time of the electric vehicle, and providing reliable information support for power grid peak regulation.

[0039] Step S700, the first predicted charging feature and the second predicted charging feature form a predicted charging feature, and the predicted charging feature is combined to perform power grid peak regulation on the predetermined power grid area. Specifically, the first predicted charging feature and the second predicted charging feature (and other possible predicted charging features of electric vehicles) are combined into an overall predicted charging feature. These predicted charging features are used to adjust the power demand of the power grid to achieve the purpose of power grid peak regulation. The embodiment of the present application adopts the method of obtaining the first electric vehicle and judging whether its historical charging divergence is within the preset divergence threshold. If it is not within the preset divergence threshold, the driving chain of the first electric vehicle is analyzed in combination with the power grid block set to obtain the driving block set, and the electric vehicle cluster is obtained with the departure block as the clustering constraint, and the driving chain information of other electric vehicles in the cluster cluster is obtained, and the charging prediction strategy is obtained. The predicted charging feature is predicted and analyzed for the passing block and the destination block in the driving block set to obtain the predicted charging feature, and the predicted charging feature is combined to perform power grid peak regulation on the predetermined power grid area, and other technical means, so as to achieve the technical effect of improving the power grid peak regulation efficiency through accurate prediction and adaptive scheduling of the charging demand of electric vehicles.

[0040] In a possible implementation, step S700 further includes step S710, obtaining a predetermined dimension indicator, wherein the predetermined dimension indicator includes real-time power and the number of blocks passed. Specifically, the predetermined dimension indicator is an indicator used for cluster analysis, which reflects the key characteristics of the electric vehicle. The real-time power is the current battery power of the electric vehicle, and is an important indicator for evaluating the charging demand of the electric vehicle. The number of blocks passed is the number of power grid blocks passed by the electric vehicle during driving, reflecting the driving distance and path complexity of the electric vehicle. Step S720, according to the predetermined dimension indicator, the first electric vehicle and the second electric vehicle are sequentially collected for dimension information, and the first indicator information and the second indicator information are obtained respectively. Specifically, for the first electric vehicle and the second electric vehicle, their predetermined dimension indicator information is collected respectively. The collected information includes the real-time power and the number of blocks passed by each electric vehicle. That is, the first indicator information refers to the collection of the real-time power and the number of blocks passed by the first electric vehicle. The second indicator information is the collection of the real-time power and the number of blocks passed by the second electric vehicle (i.e., other electric vehicles in the cluster). Step S730, performing secondary clustering on the first electric vehicle and the second electric vehicle according to the first indicator information and the second indicator information to obtain a secondary clustering result. Specifically, after obtaining the indicator information of the first electric vehicle and the second electric vehicle, the electric vehicles are secondary clustered using the information, and the electric vehicles with similar charging characteristics and driving characteristics are divided into the same clustering cluster, so as to more accurately predict their charging needs. Step S740, obtaining the target clustering cluster in the secondary clustering result, and predicting the target predicted charging characteristics of the target clustering cluster. Specifically, after completing the secondary clustering, the target clustering cluster is selected for charging characteristic prediction. The target clustering cluster is a clustering cluster with a higher charging demand or a higher charging probability. For the selected target clustering cluster, the charging characteristics of the electric vehicles therein are predicted using the charging prediction strategy to obtain the target predicted charging characteristics, including the charging location, charging time, etc. Step S750, using the target predicted charging characteristics as the predicted charging characteristics. Specifically, the target predicted charging characteristics are used as the final predicted charging characteristics for power grid peak load control. This implementation method more accurately evaluates the charging needs and driving characteristics of electric vehicles by introducing more detailed dimensional indicators (real-time power and number of blocks passed). Through secondary clustering, electric vehicles with similar characteristics are divided into the same cluster, further reducing prediction errors and improving prediction accuracy.

[0041] In a possible implementation, step S730 further includes step S731, determining whether the first real-time power in the first indicator information is greater than 30%; step S732, if greater than, adding the first electric vehicle to the first category; step S733, if less than or equal to, adding the first electric vehicle to the second category; step S734, determining whether the first real-time power meets the first number of passing blocks; step S735, if satisfied, retaining the first electric vehicle in the first category, if not satisfied, adding the first electric vehicle to the second category; step S736, taking the second category as the target cluster.

[0042] Specifically, the first indicator information of the first electric vehicle is obtained, which includes the real-time power of the electric vehicle (i.e., the first real-time power). It is determined whether the real-time power is greater than the preset power threshold (30%). If the real-time power is greater than 30%, it is considered that the first electric vehicle has sufficient power during driving and does not need to be charged midway. The electric vehicle is added to the "first category", which represents electric vehicles that do not need to be charged on the way. If the first real-time power is less than or equal to 30%, it is considered that the power of the electric vehicle is low and needs to be charged during continued driving. Therefore, the electric vehicle is added to the second category set, which represents electric vehicles with low power and need to be charged. For the electric vehicles that have been added to the first category (i.e., those electric vehicles with real-time power greater than 30%), it is determined whether their power is sufficient to support them to complete the driving from the first departure block to the first destination block, which includes all the first passing blocks. If the power of the first electric vehicle is sufficient to support it to complete the driving from the first departure block to the first destination block (including all the first passing blocks), the electric vehicle is retained in the first category. If the power is insufficient, the electric vehicle is removed from the first category and added to the second category. After completing the classification of all electric vehicles, the second category (electric vehicles that need to be charged on the way) is determined as the target cluster, which is used for subsequent charging feature prediction and grid peak load regulation strategy formulation. This implementation method accurately predicts the charging needs of electric vehicles and improves the accuracy of grid peak load regulation by judging whether they need to be charged on the way based on the real-time power and driving path of the electric vehicles.

[0043] In a possible implementation, step S732 further includes step S7321, randomly obtaining a third electric vehicle in the first category, and the third driving chain of the third electric vehicle corresponds to the third driving block set. Specifically, the first category is a set of electric vehicles that do not need to be charged on the way, and an electric vehicle is randomly selected from the first category as the third electric vehicle. Step S7322, using the third destination block in the third driving block set as the third predicted charging block of the third electric vehicle, and using the time when the third electric vehicle travels to the third destination block as the third predicted charging time of the third electric vehicle. Specifically, the driving chain of the third electric vehicle is obtained, and a driving block set is extracted from it, including a departure block, a passing block set and a destination block. The destination block of the third electric vehicle is directly used as the third predicted charging block, because according to the previous judgment, the power of the third electric vehicle is sufficient to support its travel to the destination block without charging in the middle. At the same time, according to factors such as the driving speed, remaining power, and road conditions of the third electric vehicle, the time when the third electric vehicle travels to the destination block is estimated, and this time is used as the third predicted charging time. Step S7323, constructing the third predicted charging feature of the third electric vehicle according to the third predicted charging block and the third predicted charging time. Specifically, according to the third predicted charging block and the third predicted charging time, constructing the third predicted charging feature of the third electric vehicle, this feature is a two-dimensional data point including the charging block and the charging time, indicating the predicted charging demand of the third electric vehicle. Step S7324, adding the third predicted charging feature to the predicted charging feature. Specifically, adding the third predicted charging feature of the third electric vehicle to the entire predicted charging feature set, this set includes the predicted charging features of all electric vehicles (including electric vehicles that need to be charged on the way and those that do not need to be charged on the way), which is used for the formulation of the power grid peak-shaving strategy. This implementation method comprehensively considers the charging demand of all electric vehicles by randomly selecting the third electric vehicle in the first category (electric vehicles that do not need to be charged on the way) and constructing the predicted charging feature for it. By constructing the predicted charging feature for all electric vehicles, the charging demand of the entire predetermined power grid area is more accurately predicted, thereby improving the accuracy of the power grid peak-shaving strategy.

[0044] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An adaptive control method for electric vehicle groups participating in power grid peak load regulation, characterized in that: include: Acquire a first electric vehicle, where the first electric vehicle refers to any electric vehicle in a predetermined electric vehicle group in a predetermined power grid area; Determining whether a first historical charging divergence of the first electric vehicle is within a preset divergence threshold; If not, analyzing the first driving chain of the first electric vehicle in combination with the grid block set obtained by dividing the predetermined grid area to obtain a first driving block set, wherein the first driving block set includes a first departure block, a first passing block set and a first destination block; A first electric vehicle cluster is obtained by taking the first departure block as a clustering constraint, wherein the first electric vehicle cluster refers to a clustering result of electric vehicles whose departure block is the first departure block in a plurality of driving chains; Acquire a second electric vehicle in the first electric vehicle cluster, wherein the second electric vehicle corresponds to a second driving chain, and the second driving chain corresponds to a second driving block set, wherein the second driving block set includes the first departure block, a second passing block set, and a second destination block; Acquire a charging prediction strategy, and perform prediction analysis on the first set of passing blocks and the first destination block according to the charging prediction strategy to obtain a first predicted charging feature, and perform prediction analysis on the second set of passing blocks and the second destination block to obtain a second predicted charging feature; The first predicted charging feature and the second predicted charging feature constitute a predicted charging feature, and the predicted charging feature is combined to perform grid peak regulation on the predetermined grid area.

2. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 1 is characterized in that: include: Analyzing a first historical charge and discharge record of the first electric vehicle to obtain a first charging characteristic data time series; Extracting first charging characteristic data in the first charging characteristic data time series, wherein the first charging characteristic data includes a first charging time and a first charging block; Obtaining a first historical charging scatter plot of the first electric vehicle based on the first charging time and the first charging block; A spline curve is fitted to the first historical charging scatter plot to obtain a first fitting curve, and the first fitting curve is analyzed to obtain the first historical charging divergence.

3. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 2 is characterized in that: If so, the first charging characteristic data time series of the first electric vehicle is analyzed to obtain a first charging characteristic, and grid peak regulation is performed on the predetermined grid area according to the first charging characteristic.

4. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 1 is characterized in that: include: Obtaining predetermined dimension indicators, wherein the predetermined dimension indicators include real-time power and the number of blocks passed; According to the predetermined dimensional index, dimensional information of the first electric vehicle and the second electric vehicle are collected in sequence to obtain first index information and second index information respectively; Performing secondary clustering on the first electric vehicle and the second electric vehicle according to the first indicator information and the second indicator information to obtain a secondary clustering result; Obtaining a target cluster in the secondary clustering result, and predicting a target predicted charging feature of the target cluster; The target predicted charging characteristic is used as the predicted charging characteristic.

5. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 4 is characterized in that: include: Determine whether the first real-time power level in the first indicator information is greater than 30%; If it is greater, add the first electric vehicle to the first category; Determine whether the first real-time power level satisfies the driving of the first number of passing blocks; If the conditions are met, the first electric vehicle is retained in the first category; if the conditions are not met, the first electric vehicle is added to the second category; The second class is taken as the target cluster.

6. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 5 is characterized in that: Determining whether the first real-time power level in the first indicator information is greater than 30%, also includes: if it is less than or equal to, adding the first electric vehicle to the second category.

7. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 5 is characterized in that: Also includes: Randomly acquiring a third electric vehicle in the first category, wherein a third driving chain of the third electric vehicle corresponds to a third driving block set; Using a third destination block in the third travel block set as a third predicted charging block for the third electric vehicle, and using the time it takes for the third electric vehicle to travel to the third destination block as a third predicted charging time for the third electric vehicle; forming a third predicted charging feature of the third electric vehicle according to the third predicted charging block and the third predicted charging time; The third predicted charging characteristic is added to the predicted charging characteristic.

8. The adaptive control method for electric vehicle groups participating in power grid peak load regulation according to claim 1, characterized in that: include: The grid block set includes a plurality of grid blocks with identifications indicating whether charging is possible; Extracting a first route block from the first route block set according to the charging prediction strategy; When the first passing block has a chargeable identification, multi-dimensional features are collected for the first passing block to obtain a first charging comprehensive index; Using the first charging comprehensive index as a first predicted probability that the first electric vehicle is charged in the first passing block; Analyzing the first prediction probability to obtain a target passing block that meets the charging prediction constraint; Taking the target passing block as the first predicted charging block of the first electric vehicle, and taking the time for the first electric vehicle to travel to the target passing block as the first predicted charging time of the first electric vehicle; The first predicted charging feature is constructed according to the first predicted charging block and the first predicted charging time.

Images