Charging pile scheduling method and system

Abstract

The invention provides a charging pile scheduling method and system, and belongs to the technical field of charging pile energy management, and the method comprises the steps: predicting the second state data of a charging pile group based on the historical state data of the charging pile group and the first state data of the charging pile group, taking the historical state data, the first state data and the second state data as state data of the charging pile group; wherein the first state data is the state data of the charging pile group in the current preset time period; the second state data is predicted state data of the charging pile group in a future target time period; processing the state data of the charging pile group, the demand data of the charging vehicle and the power grid load data based on an ant colony algorithm to obtain an initial charging scheduling strategy; and adjusting the initial charging scheduling strategy based on a Q learning algorithm to obtain a target charging scheduling strategy. According to the charging pile scheduling method and system provided by the invention, the charging efficiency can be improved.

Description

Technical Field

[0001] The present application belongs to the technical field of charging pile energy management, and more specifically, relates to a charging pile scheduling method and system. Background Art

[0002] With the explosive growth of electric vehicles, charging pile scheduling has become a key bottleneck restricting the development of the new energy industry. Traditional charging pile scheduling methods rely on static allocation strategies or simple greedy algorithms, which are unable to cope with complex and changing application scenarios. Furthermore, the coexistence of multiple types of charging piles (fast charging and slow charging) further exacerbates scheduling complexity, making traditional solutions difficult to achieve multi-objective coordinated optimization. Therefore, an intelligent and dynamic charging pile scheduling method is urgently needed. Summary of the Invention

[0003] The purpose of this application is to provide a charging pile scheduling method and system to achieve multi-objective collaborative optimization, thereby improving charging efficiency.

[0004] A first aspect of an embodiment of the present application provides a charging pile scheduling method, comprising: Predicting second status data of the charging pile group based on historical status data of the charging pile group and first status data of the charging pile group, and using the historical status data, the first status data and the second status data as status data of the charging pile group; The first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period; An initial charging scheduling strategy is obtained by processing the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on an ant colony algorithm; The initial charging scheduling strategy is adjusted based on the Q learning algorithm to obtain a target charging scheduling strategy.

[0005] A second aspect of an embodiment of the present application provides a charging pile scheduling device, comprising: a data acquisition module, configured to predict second status data of the charging pile group based on historical status data of the charging pile group and first status data of the charging pile group, and use the historical status data, the first status data, and the second status data as status data of the charging pile group; The first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period; An initial strategy determination module is used to process the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on the ant colony algorithm to obtain an initial charging scheduling strategy; The target strategy determination module is used to adjust the initial charging scheduling strategy based on the Q learning algorithm to obtain a target charging scheduling strategy.

[0006] According to a third aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned charging pile scheduling method when executing the computer program.

[0007] In a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned charging pile scheduling method are implemented.

[0008] The beneficial effects of the charging pile scheduling method and system provided by the embodiments of the present application are: This application comprehensively acquires multiple types of data and takes into account various key factors in the charging scenario, providing a basis for the formulation of a reasonable scheduling strategy and effectively avoiding scheduling errors caused by missing information. Afterwards, the ant colony algorithm is used to process the data to obtain the initial charging scheduling strategy, which can quickly generate a more reasonable initial charging scheduling strategy and improve scheduling efficiency. Finally, the initial charging scheduling strategy is adjusted based on the Q learning algorithm. The scheduling strategy can be continuously optimized according to actual conditions, so that the target charging scheduling strategy is more in line with the dynamically changing charging demand and grid load, so as to achieve multi-objective collaborative optimization and thus improve charging efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0010] Figure 1 A flow chart of a charging pile scheduling method provided in one embodiment of the present application; Figure 2 This is a structural block diagram of a charging pile scheduling device provided in one embodiment of the present application; Figure 3 A schematic block diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0011] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0012] In order to make the purpose, technical solutions and advantages of this application clearer, specific embodiments will be described below with reference to the accompanying drawings.

[0013] Please refer to Figure 1 , Figure 1 A flow chart of a charging pile scheduling method provided in one embodiment of the present application, the method comprising: S101: Predict the second status data of the charging pile group based on the historical status data of the charging pile group and the first status data of the charging pile group, and use the historical status data, the first status data and the second status data as the status data of the charging pile group; wherein, the first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period.

[0014] In this embodiment, the status data of the charging pile group is real-time information on the physical status and available energy of the charging piles, including but not limited to location coordinates, type, power capacity, current status, number of vehicles in the queue, and estimated available time. The location coordinates are latitude and longitude or area codes, the type is fast charging, slow charging, or battery swap station, the power capacity is the maximum output power, and the current status is idle, charging, faulty, or maintenance. This embodiment can obtain charging pile status data through charging pile hardware sensors, real-time status of the management system, and data uploaded through communication protocols.

[0015] Historical status data is a record of the charging pile fleet's operations over a period of time, including busy / idle status, charging duration, and failure rate, reflecting long-term operational patterns. First-level status data is real-time status data for the current preset time period (e.g., the last hour). This data is collected in real time through the IoT and reflects the current operational status. Second-level status data is predicted status data for a future target time period (e.g., the next hour). This data is used to predict energy demand based on historical patterns and real-time trends.

[0016] The prediction model can be a time series analysis model, such as the Long Short-Term Memory (LSTM) model, which is used to capture periodic patterns and sudden changes.

[0017] In this embodiment, historical state data is extracted in batches from a database, and first-state data is aggregated in real time by edge computing nodes. An LSTM model is then trained with this historical state data to learn periodic patterns. This model then dynamically adjusts prediction parameters based on the first-state data to generate the future state, or second-state data. This embodiment facilitates the simultaneous consideration of both first-state and second-state data by subsequent algorithms, enabling proactive energy allocation planning.

[0018] This embodiment takes historical status data into consideration, fully exploring the operating patterns and characteristics of the charging pile group in the past and providing rich experience references for subsequent analysis and decision-making. By obtaining the first status data within the current preset time period, the real-time dynamics of the charging pile group can be grasped in a timely manner, ensuring the freshness and accuracy of the information. The second status data for the future target time period is predicted based on historical and current data, which can predict the status change trend of the charging pile group in advance. The combination of these three as the status data of the charging pile group can provide comprehensive, accurate, and forward-looking information support for decisions such as charging pile scheduling.

[0019] S102: Based on the ant colony algorithm, the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data are processed to obtain an initial charging scheduling strategy.

[0020] In this embodiment, the demand data of the charging vehicle is the user demand and physical characteristics of the vehicle to be charged, including but not limited to the current location, remaining power, target power, charging priority, time window and vehicle model, among which the target power is the expected fullness, the charging priority is normal and urgent, and the time window is the latest start and end time; the grid load data is the real-time operating status and constraints of the power system, including but not limited to regional load rate, grid zoning, real-time electricity price, predicted load and safety threshold.

[0021] This embodiment obtains charging vehicle demand data through reports from vehicle terminals, user APP reservation information, and historical behavior data; and obtains grid load data through real-time data from the grid dispatching center, distributed energy management systems, and meteorological data.

[0022] After obtaining the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data, the above data can also be pre-processed: First, the data is aligned in time windows, with the three types of data aligned according to the same timestamp (e.g., every minute) to ensure consistency in status. Then, spatial mapping is performed to map the charging pile and vehicle locations to the grid partitions. Finally, anomaly monitoring and standardization are performed, that is, after filtering out unreasonable data, the data from different units are normalized.

[0023] The ant colony algorithm (ACO) simulates the swarm intelligence of ants, which use pheromone communication to find the shortest path. It gradually converges to the optimal solution through the accumulation and evaporation of pheromones. The ACO inputs charging pile status data, charging vehicle demand data, and grid load data. The ACO outputs a preliminary allocation plan for vehicles and charging piles, which serves as the initial charging scheduling strategy.

[0024] Specifically, the steps of this embodiment are: First, data preprocessing and model building.

[0025] Abstract nodes and tasks: Charging piles are abstracted as nodes in the graph, and vehicle demands are regarded as tasks to be assigned.

[0026] Calculate heuristic information: Calculate a heuristic score for each path (vehicle-to-charger assignment) based on the distance between the vehicle and the charger, the current load at the charger, the regional load rate of the power grid, etc. For example, paths with shorter distances and lower loads have higher scores.

[0027] Initialize the pheromone matrix: Create a two-dimensional matrix to represent the initial pheromone concentration of each path. It can be set to a uniform distribution or assigned an initial value based on historical data.

[0028] Second, ants build solutions.

[0029] Task allocation cycle: Each ant selects a charging station for each car in turn until all cars are allocated.

[0030] State transition probability calculation: The probability of an ant choosing a charging station is determined by the pheromone concentration and heuristic information.

[0031] Third, evaluation of the solution.

[0032] Calculate the objective function value: Score the solution generated by each ant based on the preset goal (such as minimizing the total waiting time and balancing the grid load).

[0033] Fourth, pheromone update.

[0034] Global update: Add pheromones to the current optimal solution path to increase the probability of subsequent ants choosing this path.

[0035] Local renewal: Each ant slightly reduces the pheromone after passing through the path to encourage exploration of new paths.

[0036] Fifth, iterative optimization and termination.

[0037] Repeat the loop: Continue to execute steps 2-4 until the maximum number of iterations is reached or the objective function converges.

[0038] Output result: The optimal solution of the final iteration is used as the initial charging scheduling strategy.

[0039] S103: Adjust the initial charging scheduling strategy based on the Q learning algorithm to obtain a target charging scheduling strategy.

[0040] In this embodiment, the Q-learning algorithm is a model-free reinforcement learning algorithm that optimizes policies by learning value functions for state-action pairs, independent of a dynamic model of the environment. The initial charging scheduling strategy is a static allocation generated by an ant colony algorithm that does not account for real-time dynamic changes (such as new vehicles joining or charging station failures). The target charging scheduling strategy is the final solution dynamically adjusted by the Q-learning algorithm, balancing long-term benefits with real-time constraints. This embodiment uses the Q-learning algorithm to optimize the initial charging scheduling strategy in real time within a dynamic environment. The key is to continuously update the value function through interaction with the environment, achieving policy iteration and ultimately obtaining the optimal strategy, which is the target charging scheduling strategy.

[0041] From the above, it can be concluded that this application comprehensively considers various key factors in the charging scenario by comprehensively acquiring multiple types of data, providing a basis for the formulation of a reasonable scheduling strategy and effectively avoiding scheduling errors caused by missing information. Afterwards, the ant colony algorithm is used to process the data to obtain the initial charging scheduling strategy, which can quickly generate a more reasonable initial charging scheduling strategy and improve scheduling efficiency. Finally, the initial charging scheduling strategy is adjusted based on the Q learning algorithm, and the scheduling strategy can be continuously optimized according to actual conditions, so that the target charging scheduling strategy is more in line with the dynamically changing charging demand and grid load, so as to achieve multi-objective collaborative optimization and thus improve charging efficiency.

[0042] In one embodiment of the present application, an initial charging scheduling strategy is obtained by processing the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on the ant colony algorithm, including: Determine the number of ants in the ant colony algorithm based on the status data of the charging pile group; Determine the pheromone importance factor of the ant colony algorithm based on the demand data of charging vehicles; Determine the heuristic information importance factor of the ant colony algorithm based on power grid load data; The reference charging scheduling value is iteratively optimized based on the number of ants, the pheromone importance factor, and the heuristic information importance factor until the first condition is met to obtain the initial charging scheduling strategy; the first condition is that the difference between the objective functions of two consecutive iterations of the target number is less than the target threshold.

[0043] In this embodiment, the number of ants determines the parallel search capability of the algorithm and is positively correlated with the number of charging piles. That is, the more charging piles there are, the larger the search space is, and the corresponding number of ants increases.

[0044]

[0045] The number of available charging piles can be obtained from the status data of the charging pile group. is the first adjustment coefficient, The number of ants is rounded down.

[0046] The pheromone importance factor controls the weight of historical experience (pheromone concentration) in route selection and is negatively correlated with the urgency of vehicle demand, which can be calculated from the proportion of emergency vehicles.

[0047]

[0048] in, is the pheromone importance factor, is the basic value of the pheromone importance factor, is the second adjustment coefficient, is the total number of charging vehicles. Specifically, the higher the proportion of emergency vehicles, the lower the pheromone importance factor. The base value and second adjustment coefficient of the pheromone importance factor can be set manually. The number of emergency vehicles and the total number of charging vehicles can be obtained from charging vehicle demand data.

[0049] The heuristic information importance factor can control the weight of real-time information (such as distance and load) in path selection and is positively correlated with grid load fluctuation.

[0050]

[0051] in, is the heuristic information importance factor, is the basic value of the heuristic information importance factor, is the third adjustment coefficient. Specifically, the closer the load is to the threshold, the more emphasis is placed on load balancing, i.e., the heuristic information importance factor is increased. The current load and safety threshold can be obtained from the grid load data.

[0052] The reference charging scheduling strategy is initially randomly generated or based on a greedy algorithm. The objective function comprehensively evaluates the performance of the scheduling scheme. The first condition is the convergence criterion, ensuring that the algorithm terminates after the solution space stabilizes.

[0053] In this embodiment, the objective function is: ; in, is the output value of the objective function, is the average waiting time of users, is the load balance degree of the power grid, is the average utilization rate of charging piles, is the weight of the average waiting time of users, is the weight of the grid load balance, is the weight of the average utilization rate of charging piles. , , , The value range of is [0, 1].

[0054] The formula for calculating the average waiting time of users is:

[0055] in, For vehicles The waiting time is the time from when the battery reaches the charging station to when charging starts.

[0056] The calculation formula for grid load balance is:

[0057] in, is the number of power grid areas, For the region The real-time load rate, is the average load factor of all areas. Minimizing the square difference between the load of each area and the average value can avoid local overload.

[0058] The calculation formula for the average utilization rate of charging piles is:

[0059] in, is the total number of charging piles, For charging piles Maximize the efficiency of charging piles and reduce idle time.

[0060] From the above, it can be concluded that this embodiment determines the number of ants based on the charging pile group status data, so that the size of the ant colony simulated by the ant algorithm is adapted to the actual charging pile situation, thereby improving search efficiency. Determining the pheromone importance factor based on charging vehicle demand data can highlight the impact of demand factors on path selection, making scheduling more closely aligned with the actual vehicle needs. Determining the heuristic information importance factor based on grid load data can balance the role of grid load factors in scheduling. Through iterative optimization and using the first condition as the termination criterion, it can be ensured that the resulting initial charging scheduling strategy meets accuracy requirements while being both reasonable and efficient.

[0061] In one embodiment of the present application, the initial charging scheduling strategy includes a plurality of charging scheduling strategies; The initial charging scheduling strategy is adjusted based on the Q-learning algorithm to obtain the target charging scheduling strategy, including: Determine the action space vector of the Q-learning algorithm corresponding to each charging scheduling strategy; determining a state space vector of a Q-learning algorithm based on the second state data; Determine the value function of the Q-learning algorithm based on the action space vector, state space vector, reward function and discount factor; The target charging scheduling strategy is determined based on the feedback value of the value function.

[0062] In this embodiment, the action space vector is a set of optional actions corresponding to each scheduling strategy, such as reallocating vehicles, adjusting charging priorities, etc. The state space vector is represented by the system state constructed by the second state data (future prediction), including the probability of charging pile availability, load forecast, etc.

[0063] The reward function quantifies the effectiveness of an action, balancing multiple objectives such as user waiting time and grid load. The discount factor (γ∈[0,1]) is a parameter that balances short-term and long-term rewards. The value function estimates the long-term cumulative reward for a state-action pair, with Q(s,a) representing the expected value of performing action a in state s. The feedback value is the Q-value output of the value function, which is used to evaluate the effectiveness of the strategy.

[0064] The reward function is obtained by weighted calculation of the reduction in vehicle waiting time and the saving in charging cost.

[0065] In this embodiment, the process of adjusting the discount factor includes: In response to a fluctuation rate of the grid load data being greater than a first fluctuation rate threshold, increasing the discount factor by a first step; In response to a fluctuation rate of the grid load data being less than or equal to a first fluctuation rate threshold, reducing the discount factor by a second step size; The value function of the Q-learning algorithm is determined based on the action space vector, state space vector, reward function, and discount factor, including: The value function of the Q-learning algorithm is determined based on the action space vector, the state space vector, the reward function, and the adjusted discount factor.

[0066] In this embodiment, the discount factor ranges from [0 to 1] and is used to measure the importance of future rewards. The closer the discount factor is to 1, the more the algorithm prioritizes long-term gains; the closer it is to 0, the more immediate returns are prioritized. The volatility of grid load data reflects the severity of grid load fluctuations over a period of time and is calculated by calculating the standard deviation, coefficient of variation, or percentage difference between adjacent moments in the load data. The first volatility threshold is a manually set reference value used to determine whether grid load fluctuations exceed the normal range. The first and second step sizes are parameters for adjusting the magnitude of the discount factor (e.g., 0.05 for the first step size and 0.02 for the second step size), controlling the speed of parameter updates.

[0067] When the grid load data volatility exceeds the first volatility threshold, indicating significant grid load fluctuations, increasing the discount factor in the first step allows the Q-learning algorithm to prioritize long-term benefits, guiding the charging scheduling strategy toward stabilizing grid loads and avoiding exacerbating grid fluctuations due to short-term scheduling. When the grid load data volatility is less than or equal to the first threshold, the grid load is relatively stable. Lowering the discount factor in the second step allows the algorithm to prioritize current benefits, respond promptly to charging demands, and improve charging efficiency. This dynamic adjustment allows the algorithm to flexibly adapt to varying grid load conditions, enhancing the rationality and adaptability of charging scheduling.

[0068] From the above, it can be concluded that this embodiment clearly defines the action space vector corresponding to each charging scheduling strategy, provides a clear definition for the adjustment actions of different scheduling strategies in the Q learning algorithm, and facilitates algorithm operation. The state space vector is determined based on the second state data, so that the algorithm can combine the future predicted state of the charging pile group for learning optimization, which is in line with the actual scenario. The value function is determined by the action, state space vector, reward function, and discount factor, and the rewards and long-term impact of the scheduling strategy are comprehensively considered. Finally, the target strategy is determined based on the feedback of the value function, which can adaptively select the optimal charging scheduling plan, improve the charging pile scheduling effect, and enhance the intelligence and adaptability of the system.

[0069] In one embodiment of the present application, a charging pile scheduling method further includes: Monitoring first status data of a charging pile group to obtain a target monitoring result, where the target monitoring result includes normal and abnormal data; the charging pile group includes a plurality of charging piles; In response to the target monitoring result being abnormal, a power-off protection instruction is output; the power-off protection instruction is used to control the power-off device of the corresponding charging pile.

[0070] In this embodiment, the target monitoring result is the operational status conclusion obtained after analyzing the first status data, which is categorized as "normal" or "abnormal." Abnormality occurs when charging pile data exceeds safety thresholds or exhibits fault characteristics (such as overvoltage, overcurrent, overtemperature, or charging interruption). The power-off protection command is a control signal issued by the system that triggers the charging pile's built-in power-off device to cut off power. The power-off device is a safety module within the charging pile (such as a circuit breaker or relay), which immediately stops power upon receiving the command.

[0071] The target monitoring results can be obtained by analyzing the threshold values of voltage, current, temperature, etc. or by using machine learning models for fault monitoring.

[0072] Specifically, if a metric exceeds a threshold or a fault mode is detected, it is marked as "abnormal." This embodiment sends a power-off command to the corresponding charging pile via an IoT protocol (such as MQTT). Upon receiving the command, the charging pile's power-off device (such as a relay) cuts off power within 100 milliseconds.

[0073] As can be seen from the above, this embodiment monitors the first status data of multiple charging piles, can grasp the operating status of the charging pile group in real time, and promptly identify potential problems. When the target monitoring result is abnormal, it quickly outputs a power-off protection instruction to control the power-off device of the corresponding charging pile, which can effectively avoid safety accidents caused by abnormal conditions, such as short circuits and overloads, and ensure the safety of charging equipment and users. At the same time, this rapid response mechanism helps to reduce equipment damage and maintenance costs, and can also prevent abnormal conditions from impacting the power grid, maintain the stable operation of the power grid, and improve the reliability and safety of the entire charging pile system.

[0074] Corresponding to a charging pile scheduling method in the above embodiment, Figure 2 This is a structural block diagram of a charging pile scheduling device provided in one embodiment of the present application. For ease of explanation, only the parts related to the embodiment of the present application are shown. Figure 2 The charging pile scheduling device 20 includes: a data acquisition module 21, an initial strategy determination module 22 and a target strategy determination module 23.

[0075] The data acquisition module 21 is configured to predict the second state data of the charging pile group based on the historical state data of the charging pile group and the first state data of the charging pile group, and use the historical state data, the first state data, and the second state data as the state data of the charging pile group; The first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period; An initial strategy determination module 22 is configured to process the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on an ant colony algorithm to obtain an initial charging scheduling strategy; The target strategy determination module 23 is configured to adjust the initial charging scheduling strategy based on a Q learning algorithm to obtain a target charging scheduling strategy.

[0076] In one embodiment of the present application, the initial strategy determination module 22 is specifically configured to determine the number of ants in the ant colony algorithm based on the status data of the charging pile group; Determine the pheromone importance factor of the ant colony algorithm based on the demand data of charging vehicles; Determine the heuristic information importance factor of the ant colony algorithm based on power grid load data; The reference charging scheduling value is iteratively optimized based on the number of ants, the pheromone importance factor, and the heuristic information importance factor until the first condition is met to obtain the initial charging scheduling strategy; the first condition is that the difference between the objective functions of two consecutive iterations of the target number is less than the target threshold.

[0077] In one embodiment of the present application, the objective function is: ; in, is the output value of the objective function, is the average waiting time of users, is the load balance degree of the power grid, is the average utilization rate of charging piles, is the weight of the average waiting time of users, is the weight of the grid load balance, is the weight of the average utilization rate of charging piles.

[0078] In one embodiment of the present application, the initial charging scheduling strategy includes a plurality of charging scheduling strategies; A target strategy determination module 23 is specifically used to determine the action space vector of the Q learning algorithm corresponding to each charging scheduling strategy; determining a state space vector of a Q-learning algorithm based on the second state data; Determine the value function of the Q-learning algorithm based on the action space vector, state space vector, reward function and discount factor; The target charging scheduling strategy is determined based on the feedback value of the value function.

[0079] In one embodiment of the present application, a charging pile scheduling device 20 further includes: a discount factor adjustment module, configured to increase the discount factor by a first step in response to a fluctuation rate of grid load data being greater than a first fluctuation rate threshold; In response to a fluctuation rate of the grid load data being less than or equal to a first fluctuation rate threshold, reducing the discount factor by a second step size; The target strategy determination module 23 is further configured to determine the value function of the Q learning algorithm based on the action space vector, the state space vector, the reward function, and the adjusted discount factor.

[0080] In one embodiment of the present application, the reward function is obtained by weighted calculation of the reduction in vehicle waiting time and the saving in charging cost.

[0081] In one embodiment of the present application, a charging pile scheduling device 20 further includes: a power-off protection module for monitoring first status data of a charging pile group to obtain a target monitoring result, the target monitoring result including normal and abnormal; the charging pile group includes a plurality of charging piles; In response to the target monitoring result being abnormal, a power-off protection instruction is output; the power-off protection instruction is used to control the power-off device of the corresponding charging pile.

[0082] See also Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided in one embodiment of the present application. Figure 3 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 2 The functions of the data acquisition module 21, the initial strategy determination module 22 and the target strategy determination module 23 are shown.

[0083] It should be understood that in the embodiments of the present application, the processor 301 may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0084] The input device 302 may include a touchpad, a fingerprint collection sensor (for collecting user fingerprint information and fingerprint direction information), a microphone, etc. The output device 303 may include a display (LCD, etc.), a speaker, etc.

[0085] The memory 304 may include a read-only memory and a random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include a nonvolatile random access memory.

[0086] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present application can execute the implementation methods described in the first and second embodiments of a charging pile scheduling method provided in the embodiments of the present application, and can also execute the implementation methods of the electronic device described in the embodiments of the present application, which will not be repeated here.

[0087] In another embodiment of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, all or part of the process of the method in the above embodiment is implemented. The computer program can also be used to instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of each of the above method embodiments are implemented. The computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium.

[0088] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. Furthermore, the computer-readable storage medium can include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or is about to be output.

[0089] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0090] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0091] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or can be an electrical, mechanical or other form of connection.

[0092] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0093] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0094] The above are only specific embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A charging pile scheduling method, characterized in that: include: Predicting second status data of the charging pile group based on historical status data of the charging pile group and first status data of the charging pile group, and using the historical status data, the first status data and the second status data as status data of the charging pile group; The first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period; An initial charging scheduling strategy is obtained by processing the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on an ant colony algorithm; The initial charging scheduling strategy is adjusted based on the Q learning algorithm to obtain a target charging scheduling strategy.

2. A charging pile scheduling method according to claim 1, characterized in that: The ant colony algorithm is used to process the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data to obtain an initial charging scheduling strategy, including: Determining the number of ants in the ant colony algorithm based on the status data of the charging pile group; Determining the pheromone importance factor of the ant colony algorithm based on the demand data of the charging vehicle; Determining a heuristic information importance factor of the ant colony algorithm based on the power grid load data; The reference charging scheduling value is iteratively optimized based on the number of ants, the pheromone importance factor, and the heuristic information importance factor until a first condition is met to obtain the initial charging scheduling strategy; the first condition is that the difference between the objective functions of two consecutive iterations of the target number is less than a target threshold.

3. A charging pile scheduling method according to claim 2, characterized in that: The objective function is: ； in, is the output value of the objective function, is the average waiting time of users, is the load balance degree of the power grid, is the average utilization rate of charging piles, is the weight of the average waiting time of users, is the weight of the grid load balance, is the weight of the average utilization rate of charging piles.

4. A charging pile scheduling method according to claim 1, characterized in that: The initial charging scheduling strategy includes multiple charging scheduling strategies; The adjusting the initial charging scheduling strategy based on the Q learning algorithm to obtain a target charging scheduling strategy includes: Determine the action space vector of the Q-learning algorithm corresponding to each charging scheduling strategy; determining a state space vector of a Q-learning algorithm based on the second state data; Determining a value function of the Q-learning algorithm based on the action space vector, the state space vector, the reward function, and the discount factor; A target charging scheduling strategy is determined based on the feedback value of the value function.

5. A charging pile scheduling method according to claim 4, characterized in that: The discount factor adjustment process includes: In response to a fluctuation rate of the grid load data being greater than a first fluctuation rate threshold, increasing the discount factor by a first step; In response to a fluctuation rate of the grid load data being less than or equal to a first fluctuation rate threshold, reducing the discount factor by a second step size; Determining a value function of the Q-learning algorithm based on the action space vector, the state space vector, the reward function, and the discount factor includes: A value function of the Q-learning algorithm is determined based on the action space vector, the state space vector, the reward function, and the adjusted discount factor.

6. A charging pile scheduling method according to claim 4, characterized in that: The reward function is obtained by weighted calculation of the reduction in vehicle waiting time and the saving in charging cost.

7. A charging pile scheduling method according to claim 1, characterized in that: Also includes: Monitoring the first status data of the charging pile group to obtain a target monitoring result, wherein the target monitoring result includes normal and abnormal; The charging pile group includes a plurality of charging piles; In response to the target monitoring result being abnormal, a power-off protection instruction is output; the power-off protection instruction is used to control the power-off device of the corresponding charging pile.

8. A charging pile scheduling device, characterized in that: include: a data acquisition module, configured to predict second status data of the charging pile group based on historical status data of the charging pile group and first status data of the charging pile group, and use the historical status data, the first status data, and the second status data as status data of the charging pile group; The first status data is the status data of the charging pile group in the current preset time period; the second status data is the predicted status data of the charging pile group in the future target time period; An initial strategy determination module is used to process the status data of the charging pile group, the demand data of the charging vehicles, and the grid load data based on the ant colony algorithm to obtain an initial charging scheduling strategy; The target strategy determination module is used to adjust the initial charging scheduling strategy based on the Q learning algorithm to obtain a target charging scheduling strategy.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images