Electric vehicle charging method, device and equipment based on time-of-use electricity price and medium

Abstract

The invention discloses an electric vehicle charging method, device and equipment based on time-of-use electricity price, and a medium, and relates to the field of smart power grids, and the method comprises the steps: generating a target power data set according to the load data of a power grid side, the charging data of a target charging pile, and a preset time-of-use electricity price parameter; performing power grid load curve prediction based on the target power data set through a preset prediction model, and performing peak-valley period division on a predicted target prediction power grid load curve to obtain divided periods; generating a time-of-use electricity price signal according to the divided time periods, and constructing a charging strategy function according to the target power data set and the time-of-use electricity price signal; and solving the charging strategy function to obtain a corresponding target solution set, scoring a plurality of solutions in the target solution set through a preset weight, and generating a charging strategy based on the target solution with the highest score, so that the electric vehicle can be charged based on the charging strategy. Therefore, the electric vehicle charging cost of the user can be reduced while the intra-day load peak-valley difference is reduced.

Description

Technical Field

[0001] This invention relates to the field of smart grids, and in particular to a method, apparatus, equipment and medium for charging electric vehicles based on time-of-use pricing. Background Technology

[0002] With the rapid growth in the number of electric vehicles, the demand for charging is increasing daily. User charging behavior is becoming increasingly random and unpredictable, primarily manifested in the unpredictability of charging time, location, and power demand. This randomness directly impacts power grid operation, leading to problems such as increased peak-to-valley load differences, overload of regional distribution transformers, and localized grid voltage instability, posing significant challenges to the safe and economical operation of the power grid.

[0003] Current mainstream charging technologies include AC slow charging and DC fast charging. The concentrated use of high-power fast charging facilities further exacerbates the instantaneous load pressure on the power grid. At the same time, traditional charging systems lack a coordinated scheduling mechanism with the power grid, making it difficult to optimize and control based on the real-time status of the power grid. This results in uneven distribution of charging load in time and space, peaks in the power grid load curve, reduced equipment utilization, and risks to power supply reliability. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method, apparatus, equipment, and medium for charging electric vehicles based on time-of-use pricing. This can significantly reduce the daily peak-valley load difference while ensuring the safety boundaries of power distribution equipment, and reduce overall charging costs within an acceptable range for users, while improving the on-time departure rate and user satisfaction. The specific solution is as follows: In a first aspect, this application discloses a method for charging electric vehicles based on time-of-use pricing, comprising: Collect load data from the power grid and charging data from the target charging piles to generate a target power dataset based on the load data, the charging data, and preset time-of-use electricity price parameters; The power grid load curve within a preset time threshold is predicted based on the target power dataset using a preset prediction model. The predicted target power grid load curve is then divided into peak and valley periods according to preset fuzzy rules to obtain the divided time periods. Time-of-use (TOU) electricity price signals are generated based on the divided time periods, and a charging strategy function is constructed based on the target power dataset and the TOU electricity price signals. The charging strategy function is solved based on preset constraints and the target power dataset to obtain the corresponding target solution set. Several solutions in the target solution set are scored by preset weights, and a charging strategy is generated based on the target solution with the highest score. The charging strategy is sent to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy.

[0005] Optionally, the step of collecting load data from the power grid and charging data from the target charging pile to generate a target power dataset based on the load data, the charging data, and preset time-of-use pricing parameters includes: Historical load data and real-time load data from the power grid are collected using preset data acquisition equipment based on preset time granularity. The charging data of the target charging pile is collected by a preset monitoring unit in the target charging pile; the charging data includes the charging power, cumulative charging time and fault status data of the target charging pile. Preset time-of-use electricity price parameters are obtained from a preset electricity data source, and the historical load data, real-time load data, charging data, and preset time-of-use electricity price parameters are standardized to generate a target electricity dataset based on the obtained standardized data.

[0006] Optionally, the step of predicting the grid load curve within a preset time threshold based on the target power dataset using a preset prediction model, and dividing the predicted target grid load curve into peak and valley periods according to preset fuzzy rules to obtain the divided time periods, includes: The load data in the target power dataset is used as an input variable and input into a preset prediction model so that the preset prediction model can predict the power grid load curve within a preset time threshold in the future, thereby obtaining the target predicted power grid load curve. The segmented loads are determined based on the amplitude in the load data, and the target membership degree between the target predicted power grid load curve and the segmented loads is determined according to a preset fuzzy membership function; the segmented loads include peak loads, average loads, and off-peak loads; Based on the target membership degree, the target predicted power grid load curve is divided into peak and valley periods to obtain the divided time periods.

[0007] Optionally, the step of generating time-of-use electricity price signals based on the divided time periods, and constructing a charging strategy function based on the target electricity dataset and the time-of-use electricity price signals, includes: Based on the divided time periods, corresponding time-of-use electricity price signals are generated; the time-of-use electricity price signals include peak period electricity price signals, flat period electricity price signals, and off-peak period electricity price signals; wherein, the peak period electricity price signal is a preset first multiplier of the flat period electricity price signal, and the off-peak period electricity price signal is a preset second multiplier of the flat period electricity price signal; A grid-side load function is constructed based on the load data in the target power dataset; the grid-side load function is the ratio between the difference between the daily maximum load and the daily minimum load in the bound data and the rated capacity of the power distribution equipment; A user-side charging function is constructed based on the product of preset charging duration parameters, charging power parameters, and the time-of-use electricity price signal. The grid-side load function and the user-side charging function are used as the charging strategy function.

[0008] Optionally, the step of solving the charging strategy function based on preset constraints and the target power dataset to obtain the corresponding target solution set includes: An initial population is generated based on preset constraints and the target power dataset, and the output population is encoded to obtain the initial parent population. The initial parent population is sorted non-dominated to obtain a sorted parent population. The sorted parent population is then processed through competitive selection, normal crossover, and mutation operations to generate corresponding child populations. The sorted parent population and the child population are merged to obtain a merged population, and the merged population is deduplicated to obtain the target merged population. If the number of individuals in the target merged population is less than a preset individual threshold, the target merged population will be reprocessed through competitive selection, normal distribution crossover, and mutation operations until the number of individuals in the target merged population is not less than the preset individual threshold. If the number of individuals in the target merged population is not less than the preset individual threshold, then determine whether the current iteration number is greater than the preset iteration number threshold. If the current iteration number is not greater than the preset iteration number threshold, then the current iteration number is incremented according to the preset increment threshold to obtain a new current iteration number; The target population after merging is used as the initial parent population for the next iteration, and the process jumps to the step of performing non-dominated sorting on the initial parent population to obtain the sorted parent population. The sorted parent population is then processed through competitive selection, normal crossover and mutation operations to generate the corresponding child populations, until the current iteration number is greater than the preset iteration number threshold to obtain the target population. The target population is used as the target solution set.

[0009] Optionally, the step of scoring several solutions in the target solution set using preset weights to generate a charging strategy based on the target solution with the highest score includes: The congestion distance corresponding to each solution in the target solution set is calculated by the superior-inferior solution distance method, and the product between the preset weight and the congestion distance is calculated to obtain the comprehensive score corresponding to each solution. The target comprehensive score with the highest comprehensive score is selected from the comprehensive scores, and a charging strategy is generated based on the target solution corresponding to the target comprehensive score.

[0010] Optionally, before the charging strategy is sent to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy, the method further includes: The charging strategy is sent to the target charging pile so that the target charging pile can apply the corresponding parameters in the charging strategy to its local application. Accordingly, the step of distributing the charging strategy to a preset vehicle terminal platform, so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy, includes: The charging strategy is sent to the target charging pile and the preset vehicle terminal platform, so that the preset vehicle terminal platform applies the corresponding parameters in the charging strategy to the corresponding electric vehicle locally, and charges the target charging pile based on the charging strategy.

[0011] Secondly, this application discloses an electric vehicle charging device based on time-of-use pricing, comprising: The dataset generation module is used to collect load data from the power grid and charging data from the target charging pile, so as to generate a target power dataset based on the load data, the charging data and preset time-of-use electricity price parameters. The time period segmentation module is used to predict the power grid load curve within a preset time threshold based on the target power dataset using a preset prediction model, and to segment the predicted target power grid load curve into peak and valley time periods according to preset fuzzy rules to obtain the segmented time periods. The function construction module is used to generate time-of-use electricity price signals based on the divided time periods, and to construct charging strategy functions based on the target power dataset and the time-of-use electricity price signals. The strategy generation module is used to solve the charging strategy function based on preset constraints and the target power dataset to obtain the corresponding target solution set, and to score several solutions in the target solution set by preset weights, so as to generate a charging strategy based on the target solution with the highest score. The strategy application module is used to send the charging strategy to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy.

[0012] Thirdly, this application discloses an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the electric vehicle charging method based on time-of-use pricing as described above.

[0013] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned electric vehicle charging method based on time-of-use pricing.

[0014] In this application, load data from the power grid and charging data from the target charging pile can be collected to generate a target power dataset based on the load data, the charging data, and preset time-of-use pricing parameters. A preset prediction model is used to predict the power grid load curve within a preset time threshold based on the target power dataset. The predicted target power grid load curve is then divided into peak and valley periods according to preset fuzzy rules to obtain the divided time periods. A time-of-use pricing signal is generated based on the divided time periods, and a charging strategy function is constructed based on the target power dataset and the time-of-use pricing signal. The charging strategy function is solved based on preset constraints and the target power dataset to obtain a corresponding target solution set. Several solutions in the target solution set are scored using preset weights, and a charging strategy is generated based on the highest-scoring target solution. The charging strategy is then distributed to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy.

[0015] Therefore, the method of this application can, after collecting load data from the power grid and charging data from the target charging pile, generate a target power dataset based on preset time-of-use pricing parameters and the aforementioned two types of data. Then, a preset prediction model predicts the power grid load curve within a preset time threshold based on the target power dataset, and uses preset fuzzy rules to divide the predicted target power grid load curve into peak and valley periods. Time-of-use pricing signals are then generated based on the divided periods, and a charging strategy function is constructed based on the time-of-use pricing signals and the target power dataset. Finally, the constructed charging strategy function is solved using preset constraints, and the solutions in the preset full solution set are evaluated to select the optimal solution with the highest score, generating the charging strategy. The charging strategy is then distributed to a preset vehicle terminal platform so that the electric vehicles corresponding to the preset vehicle terminal platform can charge based on the charging strategy. In this way, the daily peak-valley load difference can be significantly reduced while ensuring the safety boundary of the power distribution equipment, and the overall charging cost can be reduced within an acceptable range for users, while improving the on-time departure rate and satisfaction. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 This application discloses a flowchart of a method for charging electric vehicles based on time-of-use pricing. Figure 2 This is a schematic diagram of a data model consisting of a charging strategy function and a constraint function disclosed in this application; Figure 3 This is a schematic diagram of a charging strategy generation process disclosed in this application; Figure 4 This is a demand response framework diagram for an electric vehicle charging method based on time-of-use pricing disclosed in this application; Figure 5 This is a schematic diagram of a charging device structure for an electric vehicle based on time-of-use pricing disclosed in this application; Figure 6 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Current mainstream charging technologies include AC slow charging and DC fast charging. The concentrated use of high-power fast charging facilities further exacerbates the instantaneous load pressure on the power grid. At the same time, traditional charging systems lack a coordinated scheduling mechanism with the power grid, making it difficult to optimize and control based on the real-time status of the power grid. This results in uneven distribution of charging load in time and space, peaks in the power grid load curve, reduced equipment utilization, and risks to power supply reliability.

[0020] To overcome the aforementioned technical problems, this application discloses a method, apparatus, equipment, and medium for charging electric vehicles based on time-of-use pricing. This can significantly reduce the daily peak-valley load difference while ensuring the safety boundaries of power distribution equipment, and reduce overall charging costs within an acceptable range for users, while improving the on-time departure rate and user satisfaction.

[0021] See Figure 1 As shown, this embodiment of the invention discloses a method for charging electric vehicles based on time-of-use pricing, including: Step S11: Collect load data from the power grid and charging data from the target charging pile to generate a target power dataset based on the load data, the charging data, and preset time-of-use electricity price parameters.

[0022] In this embodiment, it is first necessary to collect load data from the power grid and charging data from the target charging pile. Specifically, historical load data and real-time load data from the power grid need to be collected using a preset data acquisition device based on a preset time granularity. This requires the SCADA (Supervisory Control And Data Acquisition) system of the power grid dispatch center to collect both historical and real-time load data. The time granularity of the historical load data should not exceed 15 minutes and the duration should not be less than 90 days, while the update frequency of the real-time load data should not be less than 1 minute. Simultaneously, it is necessary to acquire charging data from the target charging pile collected by a preset monitoring unit within the target charging pile. This charging data includes the charging power, cumulative charging time, and fault status data of the target charging pile.

[0023] Furthermore, it is necessary to obtain preset time-of-use (TOU) electricity price parameters from a preset electricity data source, specifically an official electricity marketing system. These TOU parameters include the price range for each time period and an initial time period division scheme. Then, the historical load data, the real-time load data, the charging data, and the preset TOU parameters need to be standardized to generate the target electricity dataset based on the standardized data. This ensures the quality of the collected data, accelerating model convergence during subsequent processing and improving the efficiency of charging strategy generation.

[0024] Step S12: Based on the target power dataset, the target power load curve within a preset time threshold is predicted using a preset prediction model. The predicted target power load curve is then divided into peak and valley periods according to preset fuzzy rules to obtain the divided time periods.

[0025] In this embodiment, a preset prediction model is used to predict the power grid load curve for a future period of time in order to obtain the corresponding target predicted power grid load curve. Specifically, the load data in the target power dataset needs to be input as an input variable into the preset prediction model so that the preset prediction model can predict the power grid load curve within a preset time threshold in the future, thereby obtaining the target predicted power grid load curve. The preset prediction model needs to be a short-term prediction model, which is used to predict the power grid load curve for the next 1 to 7 days.

[0026] Furthermore, it is necessary to determine segmented loads based on the amplitude in the load data, and to determine the target membership degree between the target predicted grid load curve and the segmented loads according to a preset fuzzy membership function; the segmented loads include peak load, flat load, and valley load. It should be noted that fuzzy rules are used to divide peak and valley periods. This fuzzy rule uses load amplitude as the input variable and defines three fuzzy subsets: "peak load," "flat load," and "valley load." The fuzzy membership function uses a triangular membership function. The target predicted grid load curve needs to be divided into peak and valley periods based on the obtained target membership degrees to obtain the divided periods. It should be noted that when the peak load membership degree is not less than 0.8, the corresponding period is classified as a peak period; when the valley load membership degree is not less than 0.8, the corresponding period is classified as a valley period; the remaining periods are classified as flat periods.

[0027] Furthermore, in identifying peak, off-peak, and valley periods, a fuzzy rule base and inference mechanism are needed. This involves integrating elements such as "load change trends," "historical congestion probability," "user price sensitivity," and "weather and holiday tags" into similarity scores for the three time periods, and determining the classification based on the principle of maximum similarity. To avoid frequent boundary switching, protective zones need to be set at category transition points to limit repeated crossings within a short period. A smooth transition strategy should be adopted for instructions before and after the switch to ensure continuous perception between the pile-side and vehicle-mounted terminals. For special days such as holidays, large events, and extreme weather, pre-trained time period templates can be loaded and fine-tuned according to real-time load, ensuring consistency with historical patterns while quickly adapting to the current situation.

[0028] Step S13: Generate time-of-use electricity price signals according to the divided time periods, and construct a charging strategy function based on the target power dataset and the time-of-use electricity price signals.

[0029] In this embodiment, it is necessary to generate time-of-use (TOU) electricity price signals based on the divided time periods, and construct a charging strategy function based on the target power dataset and the TOU electricity price signals. Specifically, the charging strategy function needs to be constructed with "minimizing the peak-valley difference rate on the grid side" and "minimizing the charging cost on the user side" as dual objective functions. This involves generating corresponding TOU electricity price signals based on the divided time periods. The TOU electricity price signals include peak-hour electricity price signals, flat-hour electricity price signals, and off-peak electricity price signals. The peak-hour electricity price signal is a preset first multiplier of the flat-hour electricity price signal, and the off-peak electricity price signal is a preset second multiplier of the flat-hour electricity price signal. The preset first multiplier is not less than 1.5 times, and the preset second multiplier is not more than 0.5 times.

[0030] The function constructed with the objective of "minimizing the peak-valley difference rate on the grid side" is the grid-side load function. This function is based on load data from the target power dataset and is defined as the ratio of the difference between the daily maximum load and the daily minimum load in the bound data to the rated capacity of the distribution equipment. The function constructed with the objective of "minimizing the charging cost on the user side" is the user-side charging function. This function is based on the product of preset charging duration parameters, charging power parameters, and the time-of-use electricity price signal. Both the grid-side load function and the user-side charging function are used as the charging strategy function.

[0031] It should be noted that the constructed charging strategy function needs to satisfy preset constraints, i.e., constraint functions, and such as... Figure 2 As shown, the charging strategy function and constraint function can form the corresponding data model. The constraint function includes transformer capacity constraint, charging power constraint, battery power constraint, charging time period constraint, and peak-valley electricity price ratio constraint.

[0032] The transformer capacity constraint can be described as the total active power of the transformer not exceeding its rated capacity at any given time, as expressed below: ; in, For the basic load of residents, For electric vehicle load, Rated capacity of the transformer The system's rated power factor is used to convert apparent power into active power. Transformer efficiency is the ratio of actual output power to input power.

[0033] The charging power constraint can be described as the charging power of a single charging pile being between the minimum and maximum charging power at any given time, as shown below: ; in, as well as These represent the minimum and maximum charging power, respectively.

[0034] The battery capacity constraint can be expressed as follows: ; in, as well as These represent the minimum and maximum battery capacity of 0.1 and 0.9, respectively.

[0035] The charging period constraint can be described as follows: the charging start time of the electric vehicle is no earlier than the vehicle's arrival time, and the charging end time is no later than the vehicle's departure time. This can be expressed as follows: ; in, The starting time for daily charging for car owners. This marks the end of the charging period for car owners.

[0036] The peak-valley electricity price ratio constraint can be described as the peak-valley electricity price ratio falling between a preset minimum limit and a preset maximum limit, as shown below: ; in, This is the minimum limit for the peak-valley electricity price ratio. Maximum limit on peak-valley electricity pricing ratio Peak hour electricity price, The electricity price is for off-peak hours.

[0037] Step S14: Solve the charging strategy function based on preset constraints and the target power dataset to obtain the corresponding target solution set, and score several solutions in the target solution set using preset weights, so as to generate a charging strategy based on the target solution with the highest score.

[0038] In this embodiment, the charging strategy function needs to be solved based on preset constraints and the target power dataset to obtain the corresponding target solution set. Specifically, the improved NSGA-II (Non-dominated Sorting Genetic Algorithm II) algorithm needs to be used to solve the charging strategy function based on the preset constraints and the target power dataset. For example... Figure 3 As shown, an initial population needs to be generated based on preset constraints and the target power dataset, and the output population needs to be encoded to obtain the initial parent population P(t). Specifically, the initial population needs to be generated based on data samples in the target power dataset and preset constraints, and then the current iteration number is set to the first generation. It should be noted that when generating the initial population, an initial charging strategy needs to be randomly selected from the initial population as the baseline solution. The peak-valley difference rate of the power grid and the total charging cost for users under this strategy are calculated. Then, according to the multi-objective optimization criterion (minimizing the peak-valley difference rate + minimizing the cost), the comprehensive fitness value F of each candidate strategy is calculated using a weighted method. When generating the initial population, high-quality individuals are selected using the F value, and during the evolutionary process, the F value is used as an auxiliary selection pressure to help the algorithm converge to the high-weight reorganization region more quickly. Furthermore, the weight coefficients... The algorithm can be dynamically adjusted according to the optimization process, allowing it to focus on different objectives at different stages. The overall fitness value F can be expressed by the following formula: ; in, For grid load, For user-side charging costs, These are the preset weighting coefficients.

[0039] It can be represented as: ; in, The maximum load for power grid operation It is the minimum load for power grid operation.

[0040] It can be represented as: ; Where n is the number of electric vehicles; The time-of-use electricity price for that period; Electric vehicle charging power (kW); Let represent the charging state of vehicle i at time j. A value of 1 indicates that the vehicle is charging, while a value of 0 indicates that the vehicle is not charging. t represents the optimization time interval.

[0041] Next, the initial parent population needs to be non-dominated sorted to obtain a sorted parent population. Then, the sorted parent population is processed through competitive selection, normal distribution crossover, and mutation operations to generate corresponding child populations. It should be noted that the crossover and mutation operations are based on a normal distribution, and adaptive step sizes and boundary adsorption rules are set for time-power type decision variables, allowing the offspring to expand the search range without disrupting the available charging window and the device power limit. After obtaining the child populations, the sorted parent population and the child populations need to be merged to obtain a merged population. The merged population is then deduplicated to obtain the target merged population. Finally, the number of individuals in the target merged population needs to be judged. If the number of individuals in the target merged population is less than a preset individual threshold pt, then... Figure 3 As shown, the target merged population needs to be reprocessed through a bidding process, normal distribution crossover, and mutation operations to generate a new subpopulation. This subpopulation is then merged with the current target merged population until the number of individuals in the target merged population is not less than a preset individual threshold pt. It should be noted that introducing a duplicate individual removal mechanism to remove duplicates from the solution set that are highly similar in key dimensions can maintain population diversity and avoid premature convergence. Furthermore, the crossover probability and mutation intensity can be dynamically adjusted during the iteration process. The exploration ratio is automatically increased when population diversity decreases, the target improvement rate slows down, or external perturbations increase; the development ratio is increased when the solution set boundary is stable, the proportion of feasible solutions is high, or rapid finalization is required for on-site execution.

[0042] Once the number of individuals in the merged population is not less than a preset individual threshold pt, the iteration count needs to be checked. If the current iteration count is not greater than a preset iteration count threshold maxGEN, the current iteration count needs to be incremented according to a preset increment threshold to obtain a new current iteration count. In this embodiment, the preset increment threshold is 1. Then, the merged population is used as the initial parent population for the next iteration, and then... Figure 3 As shown, the process jumps to the step of performing a non-dominated sort on the initial parent population for the next round of iterations, until the number of iterations exceeds a preset threshold value maxGEN. The final output target population is then used as the optimal target solution set.

[0043] Furthermore, such as Figure 3 As shown, the optimal scheduling scheme needs to be selected using TOPSIS (Technique for Order Preference by Similarity to Ideal Solution, dual-baseline method). This involves calculating the congestion distance between each solution in the target solution set using the superior-inferior solution distance method, and then calculating the product of the preset weight and the congestion distance to obtain the comprehensive score for each solution. It should be noted that a larger distance indicates better solution diversity, and the preset weight is a price weight set according to the grid operation priority. The sum of the grid peak-valley difference rate weight and the user charging cost weight is 1. The peak-valley difference rate weight is 0.6 to 0.8 when the grid is heavily loaded, and 0.2 to 0.4 when the grid is lightly loaded. Then, the target comprehensive score with the highest comprehensive score needs to be selected from the comprehensive scores, and a charging strategy is generated based on the target solution corresponding to the target comprehensive score. In this way, a fast non-dominated sorting strategy, crowding degree and crowding comparison operators, and elite selection strategy are proposed through non-dominated sorting genetic algorithm, which reduces the computational complexity while ensuring the diversity of the population. It has better convergence and robustness when solving multi-objective optimization problems, thereby improving the effectiveness of the generated charging strategy.

[0044] Further explanation is needed, such as Figure 3 As shown in the figure, the number of iterations is greater than the preset iteration threshold maxGEN as the termination condition. However, in practice, the termination condition can be met when one of the following three conditions is met: the maximum number of iterations, the target improvement stagnation threshold, and the coverage of feasible solutions. Before termination, a global screening and local fine-tuning are performed to output a balanced set of candidate solutions that can be directly distributed.

[0045] Step S15: The charging strategy is sent to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy.

[0046] In this embodiment, since the electric vehicle is charged at a charging station, it is necessary to maintain consistency between the charging station and the electric vehicle's charging strategy. Therefore, before sending the charging strategy to the preset vehicle terminal platform, the charging strategy needs to be applied to the target charging station so that the target charging station can apply the corresponding parameters in the charging strategy locally. Then, the charging strategy needs to be sent to the target charging station and the preset vehicle terminal platform so that the preset vehicle terminal platform can apply the corresponding parameters in the charging strategy to the corresponding electric vehicle locally and charge at the target charging station based on the charging strategy.

[0047] In this embodiment, after collecting load data from the power grid and charging data from the target charging pile, a target power dataset is generated based on preset time-of-use pricing parameters and the aforementioned two types of data. Then, a preset prediction model predicts the power grid load curve within a preset time threshold based on the target power dataset, and uses preset fuzzy rules to divide the predicted target power grid load curve into peak and valley periods. A time-of-use pricing signal is then generated based on the divided periods, and a charging strategy function is constructed based on the time-of-use pricing signal and the target power dataset. Finally, the constructed charging strategy function is solved using preset constraints, and the solutions in the preset full solution set are evaluated to select the optimal solution with the highest score, generating the charging strategy. The charging strategy is then distributed to a preset vehicle terminal platform so that the electric vehicles corresponding to the preset vehicle terminal platform can charge based on the charging strategy. This significantly reduces the intraday peak-valley load difference while ensuring the safety boundary of the power distribution equipment, and reduces overall charging costs within an acceptable range for users while improving the on-time departure rate and satisfaction.

[0048] As a preferred embodiment, to ensure the practical application of this application, a hierarchical constraint system needs to be established. Specifically, equipment-side constraints convert the thermal limits, protection settings, and voltage stability ranges of transformers and feeders into dynamic available capacity, automatically tightening the margin in situations such as sudden increases in external load, rising ambient temperature, or temporary equipment maintenance; charging pile-side constraints set stable operating ranges and derating curves for charging terminals of different models and power levels, automatically converging the planned power towards a safer side when signals such as high temperature, low bus voltage, or fan failure are detected; vehicle-side constraints require that the battery status throughout the charging process fall within the manufacturer's recommended range, the start and end times of the plan must be within the vehicle's available window, and the user-set target must be reached before leaving the station, while limiting frequent starts and stops and drastic power fluctuations in a short period to protect battery life; price and policy constraints ensure that the ratio between peak, flat, and valley periods is maintained according to the operator's settings, compatible with overlapping policies such as local subsidies and membership discounts, ensuring effective guidance and user understanding; fairness and experience constraints dynamically prioritize long-term restricted users by statistically analyzing each user's historical number of times they have been affected and their waiting time, avoiding long-term experience deviations.

[0049] Furthermore, to ensure the robustness of the charging strategy during execution, a combination of random sampling and rolling evaluation can be used to model and validate user behavior and external disturbances during strategy execution. For uncertainties such as vehicle arrival time, departure time, initial battery level, temporary plugging / unplugging, user responsiveness to price signals, and minor terminal communication delays, value ranges and probability weights are set based on historical statistics and operational experience, generating multiple sets of random samples according to scenario scale. For each set of samples, the execution process of the collaborative charging scheme is reproduced within a fixed rolling time window, recording instantaneous and cumulative key performance indicators, including peak load, average load, peak-to-valley difference, and user weighted satisfaction within the station and feeder. The satisfaction level is measured by the proportion of timely departures and compliance with standards, as well as waiting time, electricity price guidance achievement rate, and instruction execution deviation. When the mean or quantile of any indicator in the sample set reaches a preset threshold, or when extreme samples that pose a significant risk to safe operation appear, the method automatically triggers scheme correction. Priority is given to adjusting the power allocation during low-impact periods without changing the user's planned departure time. Secondly, with the user's permission, the start time of charging or the amount of electricity allocated to adjacent periods is fine-tuned. The above evaluation is repeated at fixed intervals, and the latest collected load, price, and arrival information are included in the sampling and calculation of the next window, thereby continuously verifying the robustness and executability of the strategy throughout the entire cycle.

[0050] As a preferred embodiment, in order to deal with some emergency situations that require replanning of charging strategies, a full-process governance mechanism of "tiered triggering - step-by-step control - smooth execution - traceable recovery" is needed to ensure that charging stations can quickly reduce risks and maintain service continuity when there is a sharp increase in load, abnormal fluctuations in price signals, or sudden changes in the status of critical equipment, thereby improving the user experience. The system uses a dual-threshold, hysteresis logic to classify anomalies into three levels: warning, tightening, and exceeding limits. The warning level is usually triggered by trend indicators, such as an abnormal increase in load slope within a short period of time or an expected increase in available margin within the next few minutes. In this case, only low-priority tasks that have not yet started are lightly rearranged, and the user is prompted with "may be delayed." The tightening level is triggered by the approach of hard boundaries, such as the predicted value entering the inner layer of the safety zone, the bus voltage remaining low, or the transformer temperature rising approaching the upper limit. The system will temporarily increase the peak weight and decrease the trough weight, and set convergent power limits and duty cycle limits for the current and adjacent time periods, prioritizing the reduction of available power during high-impact periods. The exceeding-limit level is activated when the actual limit is exceeded or the protection action is about to be triggered. The system enters a rapid voltage drop channel, adopts stricter power capping and short-term charging stop, and freezes non-critical optimizations while focusing on safe recovery. To ensure fairness and social benefits, the system implements a "stop first, compensate later" rule and a multi-factor priority queue. Priority is given to vehicles nearing departure and not yet meeting standards, public service vehicles, and special support vehicles. Remaining vehicles are allocated remaining power based on a comprehensive ranking of waiting time, historical impact frequency, registration priority, and user membership level. At the execution level, the magnitude and frequency of single power changes are limited, and all issued commands use a ramp transition to avoid frequent activation of the charging pile or battery-side protection systems. Redundant confirmation and local fallback strategies are implemented for communication links to ensure safety is maintained with conservative power even if cloud control is temporarily unavailable. During the recovery phase, the system gradually releases compressed time slots according to the principle of "prioritizing urgent needs and ensuring essential needs are met before supplementing reserves." Priority is given to compensating users with longer affected durations and greater target deviations, and fee reductions or points compensation are provided based on the degree of impact. Simultaneously, an emergency event report is automatically generated, fully recording the triggering conditions, handling process, power reduction curve, list of affected vehicles, recovery sequence, and effect evaluation, providing a basis for subsequent threshold calibration, rule optimization, and personnel drills. In this way, through this hierarchical and step-by-step mechanism, the site can complete risk reduction within minutes and restore order within hours, taking into account safety, efficiency and fairness, thereby improving the reliability of the charging method of this application.

[0051] As a preferred embodiment, such as Figure 4The diagram illustrates the overall framework of this application, which requires incorporating the charging load within a site or community into a dual-channel management system of "price-based response" and "incentive-based response." The price-based response side comprises price signals such as time-of-use pricing, peak pricing, and real-time pricing. Before the operating day, the system determines peak, flat, and valley boundaries based on historical load, weather, and holiday labels, generating a basic price curve. During the operating day, it gently fine-tunes the price signals based on real-time load slope and congestion risk, thereby guiding users to migrate more charging power to lower-price or lower-load periods. The incentive-based response side consists of direct load control, interruptible load, demand-side bidding, and emergency demand response. The system provides tiered compensation schemes for users who agree to participate in dispatch or accept temporary load reductions. During normal periods, direct load control and interruptible load are the primary methods. Emergency demand response is activated during anticipated over-limit periods or when the upper-level feeder issues an alarm. Demand-side bidding is opened during specific activity periods to form a market-based dispatch sequence. The aforementioned dual-channel strategy is uniformly accessed through the strategy engine of the control server. Price guidance and incentive compensation are presented in a way that is interpretable by the user terminal. When reserving the arrival and departure time and the target amount of electricity, the user chooses whether to accept the scheduling and incentive intensity. Based on this, the system generates an individual profile and response level and dynamically calls it in the subsequent optimization and distribution process, thereby achieving orderly charging behavior shaped by price as the main factor and incentive as the auxiliary factor, with price and incentive working together.

[0052] See Figure 5 As shown, this embodiment of the invention discloses a method for charging electric vehicles based on time-of-use pricing, including: The dataset generation module 11 is used to collect load data from the power grid side and charging data from the target charging pile, so as to generate a target power dataset based on the load data, the charging data and the preset time-of-use electricity price parameters. The time period segmentation module 12 is used to predict the power grid load curve within a preset time threshold based on the target power dataset using a preset prediction model, and to segment the predicted target power grid load curve into peak and valley time periods according to preset fuzzy rules to obtain the segmented time periods. The function construction module 13 is used to generate time-of-use electricity price signals according to the divided time periods, and to construct a charging strategy function according to the target power dataset and the time-of-use electricity price signals. The strategy generation module 14 is used to solve the charging strategy function based on preset constraints and the target power dataset to obtain the corresponding target solution set, and to score several solutions in the target solution set by preset weights, so as to generate a charging strategy based on the target solution with the highest score. The strategy application module 15 is used to send the charging strategy to a preset vehicle terminal platform so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy.

[0053] In this embodiment, after collecting load data from the power grid and charging data from the target charging pile, a target power dataset is generated based on preset time-of-use pricing parameters and the aforementioned two types of data. Then, a preset prediction model predicts the power grid load curve within a preset time threshold based on the target power dataset, and uses preset fuzzy rules to divide the predicted target power grid load curve into peak and valley periods. A time-of-use pricing signal is then generated based on the divided periods, and a charging strategy function is constructed based on the time-of-use pricing signal and the target power dataset. Finally, the constructed charging strategy function is solved using preset constraints, and the solutions in the preset full solution set are evaluated to select the optimal solution with the highest score, generating the charging strategy. The charging strategy is then distributed to a preset vehicle terminal platform so that the electric vehicles corresponding to the preset vehicle terminal platform can charge based on the charging strategy. This significantly reduces the intraday peak-valley load difference while ensuring the safety boundary of the power distribution equipment, and reduces overall charging costs within an acceptable range for users while improving the on-time departure rate and satisfaction.

[0054] In some embodiments, the dataset generation module 11 may specifically include: The first data acquisition unit is used to acquire historical load data and real-time load data of the power grid side through a preset data acquisition device based on a preset time granularity. The second data acquisition unit is used to acquire the charging data of the target charging pile collected by the preset monitoring unit in the target charging pile; the charging data includes the charging power, cumulative charging time and fault status data of the target charging pile. The data structure unit is used to obtain preset time-of-use electricity price parameters from preset power data sources, and to standardize the historical load data, the real-time load data, the charging data, and the preset time-of-use electricity price parameters to generate a target power dataset based on the obtained standardized data.

[0055] In some embodiments, the time period division module 12 may specifically include: The load curve prediction unit is used to input the load data in the target power dataset as an input variable into a preset prediction model, so that the preset prediction model can predict the power grid load curve within a preset time threshold in the future, and obtain the target predicted power grid load curve. The membership determination unit is used to determine the segmented load based on the amplitude in the load data, and to determine the target membership degree between the target predicted power grid load curve and the segmented load according to a preset fuzzy membership function; the segmented load includes peak load, flat load and valley load; The time period division unit is used to divide the target predicted power grid load curve into peak and valley time periods based on the target membership degree, so as to obtain the divided time periods.

[0056] In some embodiments, the function construction module 13 may specifically include: The electricity price signal generation unit is used to generate corresponding time-of-use electricity price signals according to the divided time periods; the time-of-use electricity price signals include peak period electricity price signals, flat period electricity price signals and off-peak period electricity price signals; wherein, the peak period electricity price signal is a preset first multiplier of the flat period electricity price signal, and the off-peak period electricity price signal is a preset second multiplier of the flat period electricity price signal; The first function construction unit is used to construct a grid-side load function based on the load data in the target power dataset; the grid-side load function is the ratio between the difference between the daily maximum load and the daily minimum load in the bound data and the rated capacity of the power distribution equipment. The second function construction unit is used to construct a user-side charging function based on the product of preset charging duration parameters, charging power parameters and the time-of-use electricity price signal. The charging strategy function determination unit is used to take the grid-side load function and the user-side charging function as the charging strategy function.

[0057] In some embodiments, the policy generation module 14 may specifically include: An initial population generation unit is used to generate an initial population based on preset constraints and the target power dataset, and to encode the output population to obtain an initial parent population. The subpopulation generation unit is used to perform non-dominated sorting on the initial parent population to obtain a sorted parent population, and to process the sorted parent population through competitive selection, normal crossover and mutation operations to generate corresponding subpopulations. The population merging unit is used to merge the sorted parent population with the child population to obtain a merged population, and to remove duplicates from the merged population to obtain the target merged population. The population processing unit is used to reprocess the target merged population through competitive selection, normal distribution crossover and mutation operations if the number of individuals in the target merged population is less than a preset individual threshold, until the number of individuals in the target merged population is not less than the preset individual threshold. The iteration count determination unit is used to determine whether the current iteration count is greater than the preset iteration count threshold if the number of individuals in the target merged population is not less than the preset individual threshold. An iteration count increment unit is used to increment the current iteration count according to the preset increment threshold if the current iteration count is not greater than the preset iteration count threshold, so as to obtain a new current iteration count. The step jump unit is used to take the target merged population as the initial parent population for the next iteration, and jump to the step of performing non-dominated sorting on the initial parent population to obtain the sorted parent population, and processing the sorted parent population through competitive selection, normal distribution crossover and mutation operations to generate the corresponding child population, until the current iteration number is greater than the preset iteration number threshold to obtain the target population; The target solution set determination unit is used to determine the target population as the target solution set.

[0058] In some embodiments, the policy generation module 14 may specifically include: The scoring calculation unit is used to calculate the congestion distance corresponding to each solution in the target solution set by using the superior-inferior solution distance method, and to calculate the product between the preset weight and the congestion distance to obtain the comprehensive score corresponding to each solution. The strategy generation unit is used to select the target comprehensive score with the highest comprehensive score from the comprehensive scores, and generate a charging strategy based on the target solution corresponding to the target comprehensive score.

[0059] In some embodiments, the policy application module 15 may specifically include: The charging strategy is sent to the target charging pile so that the target charging pile can apply the corresponding parameters in the charging strategy to its local application. Accordingly, the step of distributing the charging strategy to a preset vehicle terminal platform, so that the electric vehicle corresponding to the preset vehicle terminal platform can charge based on the charging strategy, includes: The charging strategy is sent to the target charging pile and the preset vehicle terminal platform, so that the preset vehicle terminal platform applies the corresponding parameters in the charging strategy to the corresponding electric vehicle locally, and charges the target charging pile based on the charging strategy.

[0060] Furthermore, embodiments of this application also disclose an electronic device, Figure 6 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0061] Figure 6This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the time-of-use pricing-based electric vehicle charging method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0062] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0063] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0064] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the time-of-use pricing-based electric vehicle charging method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0065] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned electric vehicle charging method based on time-of-use pricing. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0066] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0067] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0068] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0069] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0070] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A time-of-use rate-based electric vehicle charging method, characterized by, The method comprises the following steps: Collecting load data of a power grid side and charging data of a target charging pile to generate a target power data set according to the load data, the charging data and preset time-of-use electricity price parameters; Predicting, by a preset prediction model, a power grid load curve within a preset time threshold in the future based on the target power data set, and dividing a peak-valley period of the predicted target predicted power grid load curve according to a preset fuzzy rule to obtain a divided period; Generating a time-of-use electricity price signal according to the divided period, and constructing a charging strategy function according to the target power data set and the time-of-use electricity price signal; Solving the charging strategy function based on preset constraint conditions and the target power data set to obtain a corresponding target solution set, and scoring a plurality of solutions in the target solution set by a preset weight to generate a charging strategy based on a target solution with the highest score; Downlinking the charging strategy to a preset vehicle terminal platform so that an electric vehicle corresponding to the preset vehicle terminal platform charges based on the charging strategy.

2. The time-of-use electricity price based electric vehicle charging method according to claim 1, wherein, The method comprises the following steps: Collecting historical load data and real-time load data of a power grid side based on a preset time granularity by a preset data collection device; Obtaining charging data of a target charging pile collected by a preset monitoring unit in the target charging pile; the charging data comprises charging power, cumulative charging duration and fault state data of the target charging pile; Obtaining preset time-of-use electricity price parameters from a preset power data source, and performing standardization processing on the historical load data, the real-time load data, the charging data and the preset time-of-use electricity price parameters to generate a target power data set based on the obtained standardized data.

3. The time-of-use electricity pricing based electric vehicle charging method of claim 1, wherein, The method comprises the following steps: Inputting the load data in the target power data set as an input variable into a preset prediction model so that the preset prediction model predicts a power grid load curve within a preset time threshold in the future to obtain a target predicted power grid load curve; Determining a segmented load based on an amplitude in the load data, and determining a target membership between the target predicted power grid load curve and the segmented load according to a preset fuzzy membership function; the segmented load comprises a peak load, a flat load and a valley load; Dividing a peak-valley period of the target predicted power grid load curve based on the target membership to obtain a divided period.

4. The time-of-use electricity pricing based electric vehicle charging method of claim 1, wherein, The method comprises the following steps: Generate a corresponding time-of-use electricity price signal according to the divided time period; the time-of-use electricity price signal includes a peak time period electricity price signal, a flat time period electricity price signal, and a valley time period electricity price signal; wherein the peak time period electricity price signal is a preset first multiple of the flat time period electricity price signal, and the valley time period electricity price signal is a preset second multiple of the flat time period electricity price signal; Construct a power grid side load function based on the load data in the target power data set; the power grid side load function is a ratio between a difference between a daily maximum load and a daily minimum load in the bound data and a rated capacity of a power distribution device; Construct a user side charging function based on a product of a preset charging time length parameter, a charging power parameter, and the time-of-use electricity price signal; Take the power grid side load function and the user side charging function as the charging strategy function.

5. The time-of-use electricity pricing based electric vehicle charging method of claim 1, wherein, Solve the charging strategy function based on a preset constraint condition and the target power data set to obtain a corresponding target solution set, including: Generate an initial population based on a preset constraint condition and the target power data set, and encode the output population to obtain an initial parent population; Perform non-dominated sorting on the initial parent population to obtain a sorted parent population, and process the sorted parent population through auction selection, normal distribution crossover, and mutation operation to generate a corresponding child population; Merge the sorted parent population and the child population to obtain a merged population, and remove duplicates from the merged population to obtain a target merged population; If the number of individuals in the target merged population is less than a preset individual threshold, re-process the target merged population through auction selection, normal distribution crossover, and mutation operation until the number of individuals in the target merged population is not less than the preset individual threshold; If the number of individuals in the target merged population is not less than the preset individual threshold, determine whether a current iteration number is greater than a preset iteration number threshold; If the current iteration number is not greater than the preset iteration number threshold, perform a self-increment operation on the current iteration number according to a preset self-increment threshold to obtain a new current iteration number; Take the target merged population as the initial parent population of the next round of iteration, and jump to the step of performing non-dominated sorting on the initial parent population to obtain a sorted parent population, and processing the sorted parent population through auction selection, normal distribution crossover, and mutation operation to generate a corresponding child population, until the current iteration number is greater than the preset iteration number threshold, to obtain a target population; Take the target population as the target solution set.

6. The time-of-use electricity pricing based electric vehicle charging method of claim 5, wherein, Score a plurality of solutions in the target solution set through a preset weight to generate a charging strategy based on a target solution with the highest score, including: Calculate a crowding distance corresponding to each solution in the target solution set through a Pareto optimal solution distance method, and calculate a product between the preset weight and the crowding distance to obtain a comprehensive score corresponding to each solution; screening a target comprehensive score with the highest comprehensive score from the comprehensive scores, and generating a charging strategy based on a target solution corresponding to the target comprehensive score.

7. The time-of-use electricity pricing based electric vehicle charging method according to any one of claims 1 to 6, wherein, The charging strategy is further delivered to a preset vehicle terminal platform before the preset vehicle terminal platform corresponding to an electric vehicle charges based on the charging strategy. The charging strategy is further delivered to the target charging pile so that the target charging pile applies corresponding parameters in the charging strategy locally. Correspondingly, the charging strategy is delivered to the preset vehicle terminal platform so that the preset vehicle terminal platform corresponding to the electric vehicle charges based on the charging strategy. The charging strategy is delivered to the target charging pile and the preset vehicle terminal platform so that the preset vehicle terminal platform applies corresponding parameters in the charging strategy to the local corresponding electric vehicle and charges at the target charging pile based on the charging strategy.

8. A time-of-use electricity price-based electric vehicle charging device, characterized by, It includes: a data set generation module for collecting load data on the power grid side and charging data of the target charging pile to generate a target power data set according to the load data, the charging data and preset time-of-use electricity price parameters; a time period division module for predicting the power grid load curve within a preset time threshold in the future based on the target power data set through a preset prediction model, and dividing the target predicted power grid load curve obtained by prediction into peak and valley periods according to a preset fuzzy rule to obtain divided periods; a function construction module for generating a time-of-use electricity price signal according to the divided periods, and constructing a charging strategy function according to the target power data set and the time-of-use electricity price signal; a strategy generation module for solving the charging strategy function based on a preset constraint condition and the target power data set to obtain a corresponding target solution set, and scoring a number of solutions in the target solution set through a preset weight to generate a charging strategy based on a target solution with the highest score; a strategy application module for delivering the charging strategy to a preset vehicle terminal platform so that the preset vehicle terminal platform corresponding to an electric vehicle charges based on the charging strategy.

9. An electronic device, comprising: It includes: a memory for saving a computer program; a processor for executing the computer program to implement the time-of-use electricity price based electric vehicle charging method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, a memory for saving a computer program, wherein the computer program is executed by a processor to implement the time-of-use electricity price based electric vehicle charging method according to any one of claims 1 to 7. a memory for saving a computer program, wherein the computer program is executed by a processor to implement the time-of-use electricity price based electric vehicle charging method according to any one of claims 1 to 7.

Images