Unmanned vehicle charging cooperation method based on peak load shifting of Internet of Vehicles

Abstract

The invention belongs to the technical field of unmanned vehicle charging, and relates to an unmanned vehicle charging cooperation method based on Internet of Vehicles peak load shifting, which comprises the steps of predicting a power grid load in a period of time in the future, and determining a peak period, a low peak period and a relative low peak period in the peak period; predicting order demands in a period of time in the future, and calculating the total electricity consumption of each unmanned vehicle in a peak period; counting the total power storage quantity of each unmanned vehicle, comparing the total power storage quantity with the obtained total power consumption of each unmanned vehicle in the peak period, and determining whether the vehicle with sufficient power can meet the power consumption demand of the vehicle with insufficient power; and the charging strategy is optimized. The charging demand of the unmanned vehicle is monitored in real time through the Internet of Vehicles technology, the charging strategy is optimized, part of the charging demand is transferred to the time period when the load of the power grid is low, the effect of peak load shifting is achieved, and the load pressure of the power grid in the peak period is effectively relieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned vehicle charging, and relates to an unmanned vehicle charging coordination method, in particular to an unmanned vehicle charging coordination method based on peak shaving and valley filling of vehicle networking. Background Art

[0002] With the rapid development and popularization of driverless technology, more and more driverless vehicles are beginning to play an important role in urban transportation. These vehicles need to be charged frequently to maintain their operation, and traditional charging methods often rely on power supply from the power grid, which will bring additional load pressure to the power grid during peak hours. At the same time, since the charging demand of driverless vehicles is uncertain and random, how to reasonably dispatch the charging behavior of these vehicles to reduce the load on the power grid and improve energy efficiency has become an urgent problem to be solved.

[0003] The emergence of networking technology has provided a new approach to solving this problem. The Internet of Vehicles enables real-time sharing and coordination of vehicle information through communication between vehicles and between vehicles and infrastructure. Based on the Internet of Vehicles technology, the power status, driving routes, charging requirements and other information of driverless vehicles can be monitored in real time, thereby achieving accurate scheduling and optimization of the charging behavior of these vehicles.

[0004] The following problems exist: Unbalanced load on the power grid: Under the traditional charging method, driverless vehicles are charged in a concentrated manner during peak hours, resulting in a sharp increase in the load on the power grid, affecting the stability and safety of the power grid. Low energy efficiency: Due to the uncertainty and randomness of the charging needs of driverless vehicles, traditional charging methods often cannot fully utilize the power supply capacity of the power grid, resulting in energy waste. Low utilization rate of charging facilities: Some charging facilities are in short supply during peak hours, and idle during off-peak hours, resulting in a waste of resources. Summary of the invention

[0005] The purpose of the present invention is to provide a coordinated charging method for unmanned vehicles based on peak shaving and valley filling in vehicle networking to solve the problems raised in the above background technology.

[0006] The present invention solves the technical problem by the following technical solutions:

[0007] A coordinated charging method for unmanned vehicles based on peak load shaving and valley filling in vehicle networking, characterized in that it comprises the following steps:

[0008] S1, collects and processes recent grid load data, and uses time series analysis methods to predict grid load in the future;

[0009] S2, according to the grid load forecast results, determine the peak period, off-peak period and the relatively low-peak period in the peak period;

[0010] S3, collects and analyzes recent autonomous vehicle order data and uses machine learning algorithms to predict order demand in the future;

[0011] S4, calculate the total power consumption of each driverless vehicle during the peak period according to the order demand forecast results;

[0012] S5, counting the total power storage of each driverless vehicle, and comparing it with the total power consumption of each driverless vehicle during the peak period, to determine whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power;

[0013] S5, based on whether vehicles with sufficient power can meet the power demand of vehicles with insufficient power, optimize the charging strategy to achieve the purpose of peak shaving and valley filling.

[0014] Furthermore, the method for determining the peak period, the low peak period and the relatively low valley section in the peak period according to the grid load prediction result comprises the following steps:

[0015] S2.1, obtain the power grid load data of the recent period from the power grid dispatching center or relevant data sources, the power grid load data includes daily load values, and is refined to hourly or half-hourly load values;

[0016] S2.2, cleaning the original power grid load data, removing outliers, and arranging the power grid load data in chronological order to form time series data;

[0017] S2.3, calculate the daily load characteristic values, including the daily maximum load, daily minimum load and daily average load;

[0018] S2.4, determine the peak period and the off-peak period: according to the daily load characteristic value, set a threshold value, determine the period when the load value is higher than the threshold value as the peak period, and determine the period when the load value is lower than the threshold value as the off-peak period;

[0019] S2.5, determine the relative valley section during the peak period and calculate the peak period load fluctuation: during the peak period, calculate the deviation of the load value at each time point from the average load value during the period

[0020] Di=Pi-P avg_peak

[0021] Where Pi represents the load value at the i-th time point in a day, and P avg_peak represents the average load value during the peak period;

[0022] Determine the relative low valley section: Sort the load deviation values ​​during the peak period, find the time period with smaller deviation values, set a deviation threshold according to actual needs, and determine the time period with load deviation less than the deviation threshold as the relative low valley section during the peak period.

[0023] Moreover, the unmanned vehicle order data includes order quantity, destination, and mileage.

[0024] Furthermore, the method for calculating the total power consumption of each driverless vehicle during the peak period according to the order demand prediction result comprises the following steps:

[0025] S4.1, collect historical order data and use machine learning algorithms to predict vehicle order data for the next day;

[0026] S4.2, Calculate the power consumption of a single vehicle: Calculate the power consumption of a single driverless vehicle in a single order based on the vehicle type and mileage:

[0027] E single = D × C / 100

[0028] Among them, E single is the power consumption of a single driverless vehicle in a single order, kWh; D is the mileage, km; C is the power consumption per unit mileage, kWh / km, which is determined by the vehicle type;

[0029] S4.3, count the number of orders during peak period: count the number of orders for each type of driverless vehicle during peak period; assume that the number of orders for the i-th type of driverless vehicle during peak period is Ni;

[0030] S4.4, calculate the total peak electricity consumption: Based on the electricity consumption of a single vehicle and the number of orders during the peak period, calculate the total peak electricity consumption of each driverless vehicle:

[0031] E total_i=N i×E single_i

[0032] Where, E total_i is the total power consumption of the i-th driverless vehicle during the peak period, kWh; Ni is the number of orders for the i-th driverless vehicle during the peak period; E single_i is the power consumption of the i-th driverless vehicle in a single order, kWh;

[0033] S4.5, determine the total peak load of each type of unmanned vehicle, assuming that all unmanned vehicles of the same type have the same charging power or average charging power when charging,

[0034] Ptotal_i=E total_i / T charge_i

[0035] Among them, P total_i is the total load of the i-th unmanned vehicle during peak hours, kW; E total_i is the total power consumption of the i-th unmanned vehicle during peak hours, kWh; T charge_i is the average charging time of the i-th unmanned vehicle, which can be estimated based on parameters such as vehicle type and battery capacity.

[0036] Moreover, the vehicle order data includes vehicle type, order time, mileage, etc.

[0037] Furthermore, the method for determining whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power comprises the following steps:

[0038] S5.1, define the vehicle set: let the vehicle set with sufficient power be A, and the vehicle set with insufficient power be B. Each vehicle has a current power attribute, denoted as Q. Define power demand: For each vehicle in set B, there is a power demand attribute, denoted as D, which indicates the amount of power that the vehicle needs to replenish. Define matching conditions: The condition for a supplier vehicle to match another demander vehicle is that the remaining power of the supplier is greater than or equal to the power demand of the demander.

[0039] S5.2, perform the greedy algorithm steps: sort the vehicles in set A from high to low according to the current power Q, and sort the vehicles in set B from low to high according to the power demand D. Matching process: initialize two pointers to point to the beginning of sets A and B respectively, traverse each vehicle in set B, and for each vehicle, check whether the vehicle pointed to by the current pointer in set A can meet its power demand; if it can be met, match it, that is, reduce the power of the supplier by D, meet the power demand of the demander, record the successful match, and move the pointer of set A backward; if it cannot be met, try the next vehicle in set A; if all vehicles in set A cannot meet the current power demand of the demander, record the match failure, and continue to check the next vehicle in set B; Termination condition: When all vehicles in set B have been checked, or all vehicles in set A have been tried to match, the algorithm terminates;

[0040] S5.3, Calculation of matching success rate: The matching success rate can be defined as the number of successfully matched vehicles divided by the total number of vehicles in set B. The matching success rate is used to determine whether a vehicle with sufficient power can meet the power demand of a vehicle with insufficient power.

[0041] Moreover, the method of optimizing the charging strategy to achieve the purpose of peak load shifting based on whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power is:

[0042] When vehicles with sufficient power can meet the power needs of vehicles with insufficient power, vehicles can be charged interactively during peak hours, eliminating the need to use the power grid for charging during peak hours.

[0043] When the grid load is higher than the threshold, vehicles with sufficient power will be used to discharge power to the grid;

[0044] When vehicles with sufficient power cannot meet the power needs of vehicles with insufficient power, some vehicles will be charged alternately during peak hours. At the same time, when vehicles are short of power, the power grid will be used to charge vehicles with insufficient power during the nearest relatively low-peak period to minimize the impact on the power grid during peak hours.

[0045] The advantages and beneficial effects of the present invention are:

[0046] This unmanned vehicle charging coordination method based on peak shaving and valley filling of the Internet of Vehicles monitors the charging needs of unmanned vehicles in real time through the Internet of Vehicles technology, optimizes the charging strategy, and transfers part of the charging demand to the period when the grid load is lower, achieving the effect of peak shaving and valley filling, and effectively reducing the load pressure on the grid during peak periods.

[0047] The charging coordination method based on the Internet of Vehicles can dynamically adjust the charging power and charging time according to the grid load and the charging needs of unmanned vehicles, thereby achieving efficient use of energy.

[0048] By accurately scheduling the charging behavior of unmanned vehicles and enabling charging interactions between vehicles, the power supply capacity of charging facilities can be fully utilized and waste of resources can be avoided.

[0049] The present invention can adapt to changes in charging demands in different scenarios, has strong adaptability and robustness, and provides strong support for the widespread application of driverless vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flow chart of the interaction between vehicles of the present invention;

[0051] Figure 2 A flow chart of the interaction between the vehicle and the power grid of the present invention;

[0052] Figure 3 A flow chart for determining peak and off-peak periods for the present invention;

[0053] Figure 4 A flow chart for determining the total load and total power consumption of an unmanned vehicle for the present invention;

[0054] Figure 5 Flowchart for determining power demand for the present invention. DETAILED DESCRIPTION

[0055] The present invention is further described in detail below through specific examples. The following examples are only illustrative and not restrictive, and the protection scope of the present invention cannot be limited thereto.

[0056] The peak-shaving and valley-filling of the Internet of Vehicles aims to reduce the load pressure on the power grid during peak periods by rationally scheduling the charging behavior of driverless vehicles, while ensuring that the power needs of driverless vehicles are met.

[0057] A charging coordination method for unmanned vehicles based on peak load shaving and valley filling based on vehicle networking, the innovation of which is that it includes the following steps:

[0058] S1, collects and processes recent power grid load data, and uses time series analysis methods, such as ARIMA and LSTM methods, to predict the power grid load in the future;

[0059] S2, according to the grid load forecast results, determine the peak period, off-peak period and the relatively low-peak period in the peak period;

[0060] S3, collects and analyzes recent autonomous vehicle order data, and uses machine learning algorithms, such as decision trees and random forests, to predict order demand in the future;

[0061] S4, calculate the total power consumption of each driverless vehicle during the peak period according to the order demand forecast results;

[0062] S5, counting the total power storage of each driverless vehicle, and comparing it with the total power consumption of each driverless vehicle during the peak period, to determine whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power;

[0063] S5, based on whether vehicles with sufficient power can meet the power demand of vehicles with insufficient power, optimize the charging strategy to achieve the purpose of peak shaving and valley filling.

[0064] In this embodiment, the “recent period” is one month, half a month or three months, which is a fixed value set in advance; the “future period” is 24 hours, 12 hours or 6 hours, which is also a fixed value set in advance.

[0065] According to the recent grid load data, the specific steps to determine the peak period, off-peak period and the relatively low-peak period during the peak period are as follows:

[0066] S2.1, obtain the power grid load data of the recent period from the power grid dispatching center or relevant data sources. The data should include daily load values ​​and be broken down into hourly or half-hourly load values;

[0067] S2.2, clean the raw data, remove outliers (such as abnormally high or low load values ​​caused by equipment failure or data transmission errors), and arrange the data in chronological order to form time series data;

[0068] S2.3, calculate the daily load characteristic values: Daily maximum load (Pmax): the maximum value of the grid load in a day. Daily minimum load (Pmin): the minimum value of the grid load in a day. Daily average load (Pavg): the average value of the grid load in a day;

[0069] S2.4, determine the peak period and the off-peak period: according to the daily load characteristic value, set a load threshold (such as Pthreshold), which can be determined according to a certain proportion of the daily average load (such as Pthreshold = k × Pavg, where k is the proportional coefficient); determine the period when the load value is higher than the threshold as the peak period Tpeak, and determine the period when the load value is lower than the threshold as the off-peak period;

[0070] S2.5, determine the relative valley section during the peak period and calculate the peak period load fluctuation: during the peak period, calculate the deviation of the load value at each time point from the average load value during the period

[0071] Di=Pi-P avg_peak

[0072] Where Pi represents the load value at the i-th time point in a day, and P avg_peak represents the average load value during the peak period. Determine the relatively low-peak period: Sort the load deviation values ​​during the peak period and find the period with a smaller deviation value (i.e., the period with a relatively low load). According to actual needs, set a deviation threshold (such as Dthreshold) and determine the period with a load deviation less than the threshold as the relatively low-peak period during the peak period.

[0073] Determine the total load of each type of unmanned vehicle during peak hours based on recent orders, and determine the specific steps for the total power consumption of each type of unmanned vehicle during peak hours:

[0074] S4.1, collect historical order data, and use machine learning algorithms (such as decision trees, random forests, etc.) to predict the vehicle order data for the next day. The vehicle order data includes vehicle type, order time, and mileage. The specific steps of S201 are: feature construction, construct new features according to business needs and data characteristics. For example, time-related features can be constructed, such as the day of the week, whether it is a holiday, whether it is a working day, etc. Statistical features can also be constructed, such as the average mileage and maximum mileage of historical orders. Feature conversion: Encode categorical features, such as using One-Hot Encoding or Label Encoding. Discretize continuous features, such as using binning or quantization. Perform model training and data partitioning: Divide historical data into training sets and test sets. The training set is used to train the model, and the test set is used to evaluate the performance of the model. Model selection: Select a suitable machine learning algorithm, such as random forest, gradient boosting tree (GBDT), support vector machine (SVM), etc. In this example, the random forest algorithm can be selected for prediction. Model training: Use the training set data to train the model. Adjust the model parameters, such as the number of trees in the random forest, the maximum depth, etc., to optimize the model's performance. Model evaluation: Use the test set data to evaluate the model's performance. Common evaluation indicators include accuracy, recall, F1 score, mean square error (MSE), etc. Based on the evaluation results, adjust the model's parameters or select other algorithms for training. Prediction and result analysis Use the trained model to predict future vehicle order data. The prediction results can include key information such as vehicle type, order time, and mileage.

[0075] S4.2, Calculate the power consumption of a single vehicle: Calculate the power consumption of a single driverless vehicle in a single order based on the vehicle type (type of work) and mileage:

[0076] E single = D × C / 100

[0077] Where, E single is the electricity consumption of a single driverless vehicle in a single order (kWh); D is the mileage (km); C is the power consumption per unit mileage (kWh / km), which is determined according to the vehicle type (type of work);

[0078] S4.3, count the number of orders during the peak period: count the number of orders for each type of unmanned vehicle (type of work) during the peak period Tpeak. Assume that the number of orders for the i-th type of unmanned vehicle during the peak period is Ni;

[0079] S4.4, calculate the total peak electricity consumption: Based on the electricity consumption of a single vehicle and the number of orders during the peak period, calculate the total peak electricity consumption of each driverless vehicle:

[0080] E total_i=N i×E single_i

[0081] Where, E total_i is the total power consumption of the i-th driverless vehicle during the peak period (kWh), N i is the number of orders for the i-th driverless vehicle during the peak period, and E single_i is the power consumption of the i-th driverless vehicle in a single order (kWh);

[0082] S4.5, determine the total load of each type of unmanned vehicle during peak hours. Since the charging power of each unmanned vehicle may be different and the charging process is not instantaneous, it is more complicated to directly calculate the total load (power). A simplified method is to assume that all unmanned vehicles of the same type have the same charging power (or average charging power) when charging:

[0083] Ptotal_i=E total_i / T charge_i

[0084] Among them, Ptotal_i is the total load (kW) of the i-th unmanned vehicle during peak hours, Etotal_i is the total power consumption (kWh) of the i-th unmanned vehicle during peak hours, and Tcharge_i is the average charging time (hours) of the i-th unmanned vehicle, which can be estimated based on parameters such as vehicle type and battery capacity.

[0085] Power demand calculation:

[0086] Calculate the power demand of each driverless vehicle during peak hours based on the model, battery capacity, mileage and other information of the driverless vehicle

[0087] E demand i = Ci × Di × 1 / E capacity i

[0088] Among them, Edemandi represents the power demand of the i-th unmanned vehicle, Ci represents the order quantity of the i-th unmanned vehicle, Di represents the average mileage of the i-th unmanned vehicle, and Ecapacityi represents the battery capacity of the i-th unmanned vehicle.

[0089] The specific steps of using the greedy algorithm to match supply and demand and determine whether vehicles with sufficient power can meet the power demand of vehicles with insufficient power are as follows:

[0090] S5.1, define the vehicle set: let A be the set of vehicles with sufficient power, and B be the set of vehicles with insufficient power. Each vehicle has a current power attribute, denoted as Q. Define power demand: For each vehicle in set B, there is a power demand attribute, denoted as D, which indicates the amount of power that the vehicle needs to replenish. Define matching conditions: The condition for a vehicle (supplier) to match another vehicle (demander) is that the remaining power of the supplier is greater than or equal to the power demand of the demander.

[0091] S5.2, perform the greedy algorithm steps: Sort the vehicles in set A from high to low according to the current power Q. Sort the vehicles in set B from low to high according to the power demand D. Matching process: Initialize two pointers to point to the beginning of sets A and B respectively. Traverse each vehicle in set B, and for each vehicle: check whether the vehicle pointed to by the current pointer in set A can meet its power demand. If it can be met, match it, that is, reduce the power of the supplier by D, meet the power demand of the demander, record the successful match, and move the pointer of set A back. If it cannot be met, try the next vehicle in set A (if the pointer has not reached the end of set A). If all vehicles in set A cannot meet the current power demand of the demander, record the match failure and continue to check the next vehicle in set B. Termination condition: When all vehicles in set B have been checked, or all vehicles in set A have been tried to match (but may not all be successful), the algorithm terminates.

[0092] S5.3, Calculation of matching success rate: The matching success rate can be defined as the number of successfully matched vehicles divided by the total number of vehicles in set B. Formula: Matching success rate = total number of vehicles in set B / number of successfully matched vehicles. The matching success rate is used to determine whether a vehicle with sufficient power can meet the power demand of a vehicle with insufficient power.

[0093] When vehicles with sufficient power can meet the power needs of vehicles with insufficient power, vehicles can be charged alternately during peak hours, eliminating the need to use the grid for charging. Secondly, when the grid load is higher than the threshold, vehicles with sufficient power will be used to discharge to the grid.

[0094] When vehicles with sufficient power cannot meet the power needs of vehicles with insufficient power, some vehicles will be charged alternately during peak hours. At the same time, when vehicles are short of power, the power grid will be used to charge vehicles with insufficient power during the nearest relatively low-peak period to minimize the impact on the power grid during peak hours.

[0095] According to the grid load forecast results, the charging time and charging power in the relatively low-peak period are determined.

[0096] Use Python and other programming languages ​​to implement the above algorithms, including modules such as data preprocessing, forecasting, supply and demand matching, and charging strategy optimization. Design a user-friendly interface to facilitate real-time monitoring and scheduling by managers.

[0097] The performance of the algorithm is evaluated through simulation experiments, including indicators such as peak shaving and valley filling effect, grid load fluctuation, and unmanned vehicle power satisfaction rate. The algorithm is tuned and improved based on the evaluation results.

[0098] The present invention proposes a coordinated charging method for unmanned vehicles based on peak shaving and valley filling based on vehicle networking, which achieves the effect of peak shaving and valley filling by accurately predicting grid load and order demand and optimizing charging strategy.

[0099] Although the embodiments and drawings of the present invention are disclosed for illustrative purposes, those skilled in the art will appreciate that various substitutions, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.

Claims

1. A coordinated charging method for unmanned vehicles based on peak load shaving and valley filling based on vehicle networking, characterized in that: The steps include: S1, collects and processes recent grid load data, and uses time series analysis methods to predict grid load in the future; S2, according to the grid load forecast results, determine the peak period, off-peak period and the relatively low-peak period in the peak period; S3, collects and analyzes recent autonomous vehicle order data and uses machine learning algorithms to predict order demand in the future; S4, calculate the total power consumption of each driverless vehicle during the peak period according to the order demand forecast results; S5, counting the total power storage of each driverless vehicle, and comparing it with the total power consumption of each driverless vehicle during the peak period, to determine whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power; S5, based on whether vehicles with sufficient power can meet the power demand of vehicles with insufficient power, optimize the charging strategy to achieve the purpose of peak shaving and valley filling.

2. According to claim 1, a coordinated charging method for unmanned vehicles based on peak load shaving and valley filling based on vehicle networking, characterized in that: The method for determining the peak period, the low peak period and the relatively low valley section in the peak period according to the power grid load prediction result comprises the following steps: S2.1, obtain the power grid load data of the recent period from the power grid dispatching center or relevant data sources, the power grid load data includes daily load values, and is refined to hourly or half-hourly load values; S2.2, cleaning the original power grid load data, removing outliers, and arranging the power grid load data in chronological order to form time series data; S2.3, calculate the daily load characteristic values, including the daily maximum load, daily minimum load and daily average load; S2.4, determine the peak period and the off-peak period: according to the daily load characteristic value, set a threshold value, determine the period when the load value is higher than the threshold value as the peak period, and determine the period when the load value is lower than the threshold value as the off-peak period; S2.5, determine the relative valley section during the peak period and calculate the peak period load fluctuation: during the peak period, calculate the deviation of the load value at each time point from the average load value during the period Di=Pi-P avg_peak Where Pi represents the load value at the i-th time point in a day, and P avg_peak represents the average load value during the peak period; Determine the relative low valley section: Sort the load deviation values ​​within the peak period, find the time period with smaller deviation values, set a deviation threshold according to actual needs, and determine the time period with load deviation less than the deviation threshold as the relative low valley section during the peak period.

3. The unmanned vehicle charging coordination method based on vehicle networking for peak load shaving and valley filling according to claim 1 is characterized by: The driverless vehicle order data includes order quantity, destination, and mileage.

4. The unmanned vehicle charging coordination method based on vehicle networking for peak load shaving and valley filling according to claim 1 is characterized by: The method for calculating the total power consumption of each driverless vehicle during the peak period according to the order demand prediction result comprises the following steps: S4.1, collect historical order data and use machine learning algorithms to predict vehicle order data for the next day; S4.2, Calculate the power consumption of a single vehicle: Calculate the power consumption of a single driverless vehicle in a single order based on the vehicle type and mileage: E single = D × C / 100 Among them, E single is the power consumption of a single driverless vehicle in a single order, kWh; D is the mileage, km; C is the power consumption per unit mileage, kWh / km, which is determined by the vehicle type; S4.3, count the number of orders during peak period: count the number of orders for each type of driverless vehicle during peak period; assume that the number of orders for the i-th type of driverless vehicle during peak period is Ni; S4.4, calculate the total peak electricity consumption: Based on the electricity consumption of a single vehicle and the number of orders during the peak period, calculate the total peak electricity consumption of each driverless vehicle: E total_i = Ni × E single_i Where, E total_i is the total power consumption of the i-th driverless vehicle during the peak period, kWh; Ni is the number of orders for the i-th driverless vehicle during the peak period; E single_i is the power consumption of the i-th driverless vehicle in a single order, kWh; S4.5, determine the total peak load of each type of unmanned vehicle, assuming that all unmanned vehicles of the same type have the same charging power or average charging power when charging, Ptotal_i=E total_i / T charge_i Among them, P total_i is the total load of the i-th unmanned vehicle during peak hours, kW; E total_i is the total power consumption of the i-th unmanned vehicle during peak hours, kWh; T charge_i is the average charging time of the i-th unmanned vehicle, which can be estimated based on parameters such as vehicle type and battery capacity.

5. The unmanned vehicle charging coordination method based on vehicle networking for peak load shaving and valley filling according to claim 1 is characterized by: The vehicle order data includes vehicle type, order time, mileage, etc.

6. The unmanned vehicle charging coordination method based on vehicle networking for peak load shaving and valley filling according to claim 1, characterized in that: The method for determining whether a vehicle with sufficient power can meet the power demand of a vehicle with insufficient power comprises the following steps: S5.1, define the vehicle set: let the vehicle set with sufficient power be A, and the vehicle set with insufficient power be B. Each vehicle has a current power attribute, denoted as Q. Define power demand: For each vehicle in set B, there is a power demand attribute, denoted as D, which indicates the amount of power that the vehicle needs to replenish. Define matching conditions: The condition for a supplier vehicle to match another demander vehicle is that the remaining power of the supplier is greater than or equal to the power demand of the demander. S5.2, perform the greedy algorithm steps: sort the vehicles in set A from high to low according to the current power Q, and sort the vehicles in set B from low to high according to the power demand D. Matching process: initialize two pointers to point to the beginning of sets A and B respectively, traverse each vehicle in set B, and for each vehicle, check whether the vehicle pointed to by the current pointer in set A can meet its power demand; if it can be met, match it, that is, reduce the power of the supplier by D, meet the power demand of the demander, record the successful match, and move the pointer of set A backward; if it cannot be met, try the next vehicle in set A; if all vehicles in set A cannot meet the current power demand of the demander, record the match failure, and continue to check the next vehicle in set B; Termination condition: When all vehicles in set B have been checked, or all vehicles in set A have been tried to match, the algorithm terminates; S5.3, Calculation of matching success rate: The matching success rate can be defined as the number of successfully matched vehicles divided by the total number of vehicles in set B. The matching success rate is used to determine whether a vehicle with sufficient power can meet the power demand of a vehicle with insufficient power.

7. The unmanned vehicle charging coordination method based on vehicle networking for peak load shaving and valley filling according to claim 1 is characterized by: The method for optimizing the charging strategy and achieving the purpose of peak load shifting based on whether the vehicle with sufficient power can meet the power demand of the vehicle with insufficient power is as follows: When vehicles with sufficient power can meet the power needs of vehicles with insufficient power, vehicles can be charged interactively during peak hours, eliminating the need to use the power grid for charging during peak hours. When the grid load is higher than the threshold, vehicles with sufficient power will be used to discharge power to the grid; When vehicles with sufficient power cannot meet the power needs of vehicles with insufficient power, some vehicles will be charged alternately during peak hours. At the same time, when vehicles are short of power, the power grid will be used to charge vehicles with insufficient power during the nearest relatively low-peak period to minimize the impact on the power grid during peak hours.

Images