Power distribution method and system for charging piles in charging station and medium

Abstract

The invention relates to a power distribution method and system for charging piles in a charging station and a medium, and the method comprises the steps: collecting the real-time demand information of the charging piles and the real-time state information of the charging piles, and carrying out the data preprocessing of the real-time demand information of the power and the real-time state information of the charging piles; inputting the historical data of the charging piles and the real-time demand data into a pre-trained power demand prediction model, outputting the power demand of each charging pile in a certain time in the future through the power demand prediction model, and generating a power curve of each charging pile based on power demand prediction results at different times; and inputting the power curve of the charging pile into a pre-established power distribution model, and solving the power distribution model based on an optimization algorithm to obtain an optimal power distribution scheme of each charging pile, thereby optimizing the resource utilization rate of the charging station and improving the charging efficiency by dynamically adjusting the power distribution of the charging piles.

Description

Technical Field

[0001] The present application relates to the technical field of charging systems for charging stations, and in particular to a power distribution method, system, and medium for charging piles in a charging station. Background Art

[0002] As a clean and efficient means of transportation, electric vehicles have received significant attention and widespread recognition from governments and consumers worldwide. In recent years, the electric vehicle market has experienced explosive growth. According to the International Energy Agency (IEA), the global electric vehicle fleet has grown from less than 100,000 in 2010 to over 14 million in 2022, and is projected to reach 245 million by 2030.

[0003] The rapid adoption of electric vehicles has created a huge demand for charging, driving the rapid development of charging infrastructure. As crucial energy refueling points for electric vehicles, the scale and operational efficiency of charging stations directly impact the user experience and adoption of electric vehicles. According to statistics, by the end of 2022, the number of public charging stations worldwide exceeded 2 million. However, compared to the rapidly growing number of electric vehicles, the pace of charging infrastructure development lags significantly, and problems such as difficulty and slow charging continue to plague many electric vehicle users.

[0004] Charging stations are typically equipped with multiple charging piles to meet the charging needs of different electric vehicles. The power requirements of each charging pile vary, depending primarily on factors such as the electric vehicle's battery capacity, remaining charge, and charging mode. For example, the charging power of a household electric vehicle is generally around 7kW, while the charging power of a commercial electric vehicle can reach 350kW or even higher. Furthermore, the overall power capacity of a charging station is limited, typically determined by the transformer capacity, and cannot simultaneously meet the maximum power requirements of all charging piles. How to rationally allocate the power of each charging pile within this limited power capacity to meet the charging needs of different electric vehicles and maximize the utilization and operational efficiency of the charging station has become a key issue that needs to be addressed.

[0005] Existing charging station power allocation methods mainly use a simple equal distribution strategy, that is, evenly distributing power capacity according to the number of charging piles. Although this strategy is simple to implement, it has many drawbacks:

[0006] Unable to dynamically adjust power allocation based on actual demand: The equal-sharing strategy ignores the differences in charging requirements of different electric vehicles, resulting in some charging piles having insufficient power to meet fast charging needs, while other charging piles may have excess power, resulting in a waste of resources.

[0007] Leading to waste of resources and low charging efficiency: Due to unreasonable power distribution, some electric vehicles need to wait longer to complete charging, which reduces the operating efficiency and service quality of charging stations and increases users' waiting time and usage costs. Summary of the Invention

[0008] The purpose of the embodiments of the present application is to provide a method, system and medium for power distribution of charging piles in a charging station, which can optimize the resource utilization of the charging station and improve the overall charging efficiency of the charging station by dynamically adjusting the power distribution of each charging pile.

[0009] To achieve the above objectives, this application provides the following technical solutions:

[0010] In a first aspect, an embodiment of the present application provides a method for distributing power to charging piles in a charging station, comprising the following steps:

[0011] Collecting real-time demand information and real-time charging pile status information of the charging pile, and performing data preprocessing on the real-time power demand information and real-time charging pile status information;

[0012] Inputting the historical data and the real-time demand data of the charging pile into a pre-trained power demand prediction model, outputting the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generating a power curve for each charging pile based on the power demand prediction results at different times;

[0013] The power curve of the charging pile is input into a pre-established power distribution model, and the power distribution model is solved based on an optimization algorithm to obtain the optimal power distribution solution for each charging pile.

[0014] The method further comprises the following steps:

[0015] Real-time demand information and real-time charging pile status information are collected and monitored in real time, and the optimal power allocation scheme of each charging pile is iteratively adjusted according to the real-time demand information and real-time charging pile status information.

[0016] Inputting the historical data and the real-time demand data of the charging pile into a pre-trained power demand prediction model, outputting the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generating a power curve for each charging pile based on the power demand prediction results at different times, specifically including:

[0017] The historical data and real-time demand data of the charging piles are input into a pre-trained LSTM neural network model, and the power demand of each charging pile in a certain period of time in the future is output through the LSTM neural network model. The power curve of each charging pile is generated based on the power demand prediction results at different times.

[0018] Inputting the power curve of the charging pile into a pre-established power allocation model, and solving the power allocation model based on an optimization algorithm to obtain the optimal power allocation solution for each charging pile, specifically including:

[0019] The objective function of the power allocation model is to minimize the total power fluctuation of the charging station. The specific formula is:

[0020]

[0021] in, is the total power of the charging station at time t, is the average power of the charging station,

[0022] The constraint functions of the power allocation model include:

[0023]

[0024] Among them, P i (t) represents the power of the i-th device, P tatal,max The maximum operating power allowed by the system.

[0025] The power allocation model is pre-established and trained specifically as follows:

[0026] Initialization parameters:

[0027] 1) Set the maximum power limit Pmax of the charging station;

[0028] 2) Set the time step Δt;

[0029] 3) Obtain the current charging demand Ei of all charging piles;

[0030] 4) Set the vehicle charging power range:

[0031] 0≤Pi(t)≤Pi,max where Pi,max is the maximum power of the i-th charging pile;

[0032] Calculate the objective function:

[0033] The optimization goal is to minimize the power fluctuation of the charging station:

[0034]

[0035] in: is the total power at the current time t; is the average power: ;

[0036] Calculate the current power demand:

[0037] For each charging station

[0038] 1) Calculate the remaining charging time:

[0039] If Ti is too short, i.e. the vehicle is about to be fully charged, its power allocation priority is lowered;

[0040] 2) Calculate the total power demand in the current time window: ;

[0041] 3) Calculate the power fluctuation degree:

[0042] , adjust the power allocation based on σ(t);

[0043] Perform dynamic optimization:

[0044] Optimization is performed using gradient descent:

[0045] 1) Calculate the adjustment direction

[0046] If the total power P(t) exceeds the limit, reduce the power of some charging piles with higher power;

[0047] If the power fluctuation is large, the power curve is smoothed; if some vehicles are low on power, power is allocated first;

[0048] 2) Structural optimization problem

[0049] Constraints: , ;

[0050] 3) Solve using optimization algorithm: Use gradient descent method to solve;

[0051] Adjust charging power in real time:

[0052] At each time step t:

[0053] 1) Update Pi(t) according to the new power allocation scheme calculated by optimization;

[0054] 2) Calculate the new total power P(t) to ensure that it does not exceed the maximum power;

[0055] 3) Record the power fluctuations. If the fluctuations are still large, continue optimization.

[0056] The method also includes pre-establishing and training an LSTM neural network model,

[0057] The input data of the LSTM neural network model is the historical 4-hour charging pile power data P(t);

[0058] The model output of the LSTM neural network model is the power prediction value for the next 10 minutes, that is, the power prediction curve P(t+1), P(t+2),...P(t+n).

[0059] In a second aspect, an embodiment of the present application provides a power distribution system for charging piles in a charging station, comprising:

[0060] The data acquisition and preprocessing module is used to collect real-time demand information and real-time charging pile status information of the charging pile, and perform data preprocessing on the real-time power demand information and real-time charging pile status information;

[0061] a power demand prediction module, configured to input the historical data and the real-time demand data of the charging piles into a pre-trained power demand prediction model, output the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generate a power curve for each charging pile based on the power demand prediction results at different times;

[0062] The power distribution module is used to input the power curve of the charging pile into a pre-established power distribution model, and solve the power distribution model based on an optimization algorithm to obtain an optimal power distribution solution for each charging pile.

[0063] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores program code, and when the program code is executed by a processor, the steps of the power distribution method for charging piles in a charging station as described above are implemented.

[0064] In a fourth aspect, an embodiment of the present application provides an electronic device, including:

[0065] Memory for storing computer programs;

[0066] The processor is configured to execute the steps of the method for distributing power to charging piles in a charging station as described above when executing the computer program stored in the memory.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] a. Reduce power fluctuations: Dynamically adjust power distribution through optimization algorithms to reduce total power fluctuations at charging stations and improve grid stability;

[0069] b. Improve charging efficiency: Allocate power based on the actual needs of each charging station to avoid resource waste and improve charging efficiency;

[0070] c. Enhanced adaptability: Through real-time monitoring and dynamic adjustments, the charging station's ability to adapt to emergencies is enhanced to ensure the smooth progress of the charging process. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0072] Figure 1 is a flow chart of the method of this application;

[0073] Figure 2 This is the LSTM neural network model diagram of this application;

[0074] Figure 3 This is the power allocation flow chart of this application;

[0075] Figure 4 This is a system block diagram of this application;

[0076] Figure 5 This is a diagram of the electronic device used in this application. DETAILED DESCRIPTION

[0077] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings. It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0078] The terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0079] The terms "first," "second," etc. are only used to distinguish one entity or operation from another entity or operation, and are not to be understood as indicating or implying relative importance, nor are they to be understood as requiring or implying any actual relationship or order between these entities or operations.

[0080] like Figure 1 As shown, the present application provides a method for distributing power to charging piles in a charging station, comprising the following steps:

[0081] S1: Collecting real-time power demand information and real-time charging pile status information of charging piles, and performing data preprocessing on the real-time power demand information and real-time charging pile status information, wherein the real-time demand information and charging pile status information include power demand information, battery status information, charging time information, etc. The above information can be collected by equipping each charging pile with a sensor or by other means that can obtain the above information;

[0082] S2: Input the historical data of the charging pile and the real-time demand data into a pre-trained power demand prediction model, output the power demand of each charging pile in a certain period of time in the future through the power demand prediction model, and generate a power curve for each charging pile based on the power demand prediction results at different times; It is worth noting that before using the power allocation method to allocate power, it is necessary to train the power demand prediction model, which can be an LSTM neural network model. The power demand prediction flow chart is as follows: Figure 2 shown.

[0083] S3: Input the power curve of the charging pile into the pre-established power distribution model, and solve the power distribution model based on the optimization algorithm to obtain the optimal power distribution scheme for each charging pile. It is worth noting that before using the power distribution method to distribute power, the power distribution model needs to be established and trained. For example, the power distribution model solution flow chart is as follows: Figure 3 shown.

[0084] In one embodiment of the present invention, the charging station further includes a charging station control center. The collected power demand information, battery status information, charging time information, and other information are transmitted to the charging station control center via a wireless communication module (exemplary: HPLC / Lora). The charging station control center performs operations such as data cleaning and format conversion on the received data to ensure data accuracy and consistency.

[0085] In one embodiment of the present invention, the method further includes the following steps: real-time collection and monitoring of real-time demand information and real-time charging pile status information, and iteratively adjusting the optimal power allocation scheme of each charging pile according to the real-time demand information and real-time charging pile status information.

[0086] It is worth noting that the power distribution method for charging piles in a charging station provided above is based on the real-time status of the charging piles and the needs of users. It predicts the power demand at different times and inputs the power demand into a pre-trained power distribution model to dynamically adjust the power distribution. This process is a dynamic process. After outputting the optimal power distribution plan at the current moment, the iteration will continue to input the current real-time collection and monitoring real-time demand information and real-time charging pile status information, and iteratively adjust the optimal power distribution plan for each charging pile based on the real-time demand information and real-time charging pile status information. This method reduces the total power fluctuation of the charging station, improves the stability of the power grid, distributes power according to the actual needs of each charging pile, avoids resource waste, improves charging efficiency, and enhances the adaptability of the charging station to emergencies through real-time monitoring and dynamic adjustment, ensuring the smooth progress of the charging process.

[0087] In one embodiment of the present invention, the historical data of the charging pile and the real-time demand data are input into a pre-trained LSTM neural network model, and the power demand of each charging pile in a certain period of time in the future is output through the LSTM neural network model. The power curve of each charging pile is generated based on the power demand prediction results at different times. For example, the output power demand of each charging pile in a certain period of time in the future can be the power at multiple time points in the next 15 minutes, such as 1 minute, 2 minutes...15 minutes, etc., and then the power curve of each charging pile is output with time and power as the horizontal and vertical coordinates, and the power demand curve of each charging pile is used as the input of the power allocation model.

[0088] As an improved recurrent neural network, LSTM adds a cell state structure to the original network structure to control the transmission of global information, and controls the update of cell state information values ​​through three gating units: forget gate, input gate, and output gate. LSTM has greatly alleviated the long-term dependency problem of traditional RNN models, reduced the loss of long-distance historical information, and output more accurate prediction results. The specific model of LSTM is as follows: Figure 2 ,

[0089] The historical data is used to train the model and construct an LSTM model suitable for ultra-short-term power prediction of charging piles.

[0090] The following is the python build code

[0091] # Build LSTM network

[0092] model = Sequential([

[0093] LSTM(units=50, return_sequences=True, input_shape=(lookback, 1)),

[0094] LSTM(units=50),

[0095] Dense(units=25, activation="relu"),

[0096] Dense(units=1) # Output predicted power)

[0097] # Compile the model

[0098] model.compile(optimizer="adam", loss="mse")

[0099] # Train the model

[0100] model.fit(X_train, y_train, epochs=50, batch_size=32, validation_data=(X_test, y_test)).

[0101] Predicted power data curve

[0102] Input data: 4 hours of historical charging pile power data P(t).

[0103] Model output: Power forecast value for the next 10 minutes, i.e. power forecast curve.

[0104] P(t+1),P(t+2),...P(t+n).

[0105] In one embodiment of the present invention, inputting the power curve of the charging pile into a pre-established power distribution model, and solving the power distribution model based on an optimization algorithm to obtain the optimal power distribution solution for each charging pile specifically includes:

[0106] The objective function of the power allocation model does not minimize the total power fluctuation of the charging station. The specific formula is:

[0107]

[0108] in, is the total power of the charging station at time t, is the average power of the charging station,

[0109] Exemplarily, an ant colony algorithm is used to solve the power distribution model to obtain the optimal power distribution plan for each charging pile. The solution is transmitted to the control unit of the charging pile to adjust the power output of each charging pile in real time.

[0110] In one embodiment of the present invention, the constraint function of the power allocation model includes:

[0111]

[0112] Among them, P i (t) represents the power of the i-th device, P tatal,max The maximum operating power allowed by the system.

[0113] First, the power allocation model is established and trained. Specifically, this method comprehensively considers:

[0114] 1) Charging requirements (ensuring each vehicle is fully charged)

[0115] 2) Total power smoothing (reducing fluctuations)

[0116] 3) Battery charging characteristics (avoiding sudden changes)

[0117] 4) Peak shaving and valley filling (rationally distribute power to avoid overload)

[0118] Actual implementation steps

[0119] 1. Initialization parameters

[0120] 1) Set the maximum power limit Pmax of the charging station.

[0121] 2) Set the time step Δt (e.g. 1 min or 5 min).

[0122] 3) Get the current charging demand Ei (unit: kWh) of all charging piles.

[0123] 4) Set the vehicle charging power range:

[0124] 0≤Pi(t)≤Pi,max where Pi,max is the maximum power of the i-th charging pile.

[0125] 2. Calculate the objective function

[0126] The optimization goal is to minimize the power fluctuation of the charging station:

[0127]

[0128] in: is the total power at the current time t; is the average power:

[0129] 3. Calculate current power requirements

[0130] For each charging station

[0131] 1) Calculate the remaining charging time:

[0132] If Ti is too short (i.e. the vehicle is about to be fully charged), reduce its power allocation priority.

[0133] 2) Calculate the total power demand in the current time window:

[0134] 3) Calculate the power fluctuation degree:

[0135] , if σ(t) is too large, adjust the power distribution.

[0136] Perform dynamic optimization

[0137] Optimization is performed using gradient descent:

[0138] 1) Calculate the adjustment direction

[0139] If the total power P(t) exceeds the limit, reduce the power of some charging piles with higher power;

[0140] If the power fluctuation is large, smooth the power curve; if some vehicles are low on power, prioritize power allocation.

[0141] 2) Structural optimization problem

[0142] Constraints: ( )

[0143] 3) Solve using optimization algorithm: Use gradient descent method to solve.

[0144] 5. Real-time adjustment of charging power

[0145] At each time step t:

[0146] 1) Update Pi(t) according to the new power allocation scheme calculated by optimization;

[0147] 2) Calculate the new total power P(t) to ensure that it does not exceed the maximum power;

[0148] 3) Record the power fluctuations. If the fluctuations are still large, continue optimization.

[0149] In one embodiment of the present invention, Figure 4 As shown, a power distribution system for charging piles in a charging station is provided, including the following modules:

[0150] A data acquisition and preprocessing module is used to collect real-time demand information and real-time charging pile status information of charging piles, and perform data preprocessing on the real-time power demand information and real-time charging pile status information;

[0151] a power demand prediction module, configured to input the historical data and the real-time demand data of the charging piles into a pre-trained power demand prediction model, output the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generate a power curve for each charging pile based on the power demand prediction results at different times;

[0152] The power distribution module is used to input the power curve of the charging pile into a pre-established power distribution model, and solve the power distribution model based on an optimization algorithm to obtain an optimal power distribution solution for each charging pile.

[0153] In one embodiment of the present application, an electronic device is provided, such as Figure 5 As shown, the electronic device includes:

[0154] At least one processor 601, and a memory 602 connected to the at least one processor 601. The specific connection medium between the processor 601 and the memory 602 is not limited in the embodiment of the present application. Figure 5 In the example, the processor 601 and the memory 602 are connected via a bus 600. Figure 5 The bus 600 can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 5 The diagram is represented by only one thick line, but this does not mean that there is only one bus or one type of bus. Alternatively, the processor 601 may also be referred to as a controller, without limitation to the name. In one embodiment of the present application, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the above-mentioned method for distributing power to charging piles in a charging station is implemented.

[0155] In the embodiment of the present application, the memory 602 stores instructions that can be executed by at least one processor 601. The at least one processor 601 can execute the power distribution method for charging piles in the charging station described above by executing the instructions stored in the memory 602. The processor 601 can implement Figure 5 The functions of each module in the device shown.

[0156] The processor 601 is the control center of the device, and can use various interfaces and lines to connect the various parts of the entire control device. By running or executing instructions stored in the memory 602 and calling data stored in the memory 602, the various functions of the device and processing data, the device can be monitored as a whole. In one possible design, the processor 601 may include one or more processing units. The processor 601 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 601. In some embodiments, the processor 601 and the memory 602 may be implemented on the same chip. In some embodiments, they may also be implemented separately on separate chips.

[0157] The processor 601 can be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the power distribution method for charging piles in a charging station disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.

[0158] The memory 602 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The memory 602 may include at least one type of storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory, a random access memory (Random Access Memory, RAM), a static random access memory (Static Random Access Memory, SRAM), a programmable read-only memory (Programmable Read Only Memory, PROM), a read-only memory (Read Only Memory, ROM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a magnetic memory, a disk, an optical disk, etc. The memory 602 is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 602 in the embodiment of the present application can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0159] By designing and programming the processor 601, the code corresponding to the power distribution method of the charging pile in the charging station in the above embodiment can be fixed into the chip, so that the chip can execute the power distribution method when running. Figure 1 The steps of the power distribution method for charging piles in a charging station in the embodiment shown are as follows: How to design and program the processor 601 is a technology well known to those skilled in the art and will not be described in detail here.

[0160] In one embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of the power distribution method for charging piles in a charging station as described above are performed.

[0161] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for distributing power to charging piles in a charging station, characterized in that: The following steps are involved: Collecting real-time demand information and real-time charging pile status information of the charging pile, and performing data preprocessing on the real-time power demand information and real-time charging pile status information; Inputting the historical data and the real-time demand data of the charging pile into a pre-trained power demand prediction model, outputting the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generating a power curve for each charging pile based on the power demand prediction results at different times; The power curve of the charging pile is input into a pre-established power distribution model, and the power distribution model is solved based on an optimization algorithm to obtain the optimal power distribution solution for each charging pile.

2. The power distribution method for charging piles in a charging station according to claim 1, characterized in that: The method further comprises the following steps: Real-time demand information and real-time charging pile status information are collected and monitored in real time, and the optimal power allocation scheme of each charging pile is iteratively adjusted according to the real-time demand information and real-time charging pile status information.

3. The power distribution method for charging piles in a charging station according to claim 1, characterized in that: Inputting the historical data and the real-time demand data of the charging pile into a pre-trained power demand prediction model, outputting the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generating a power curve for each charging pile based on the power demand prediction results at different times, specifically including: The historical data and real-time demand data of the charging piles are input into a pre-trained LSTM neural network model, and the power demand of each charging pile in a certain period of time in the future is output through the LSTM neural network model. The power curve of each charging pile is generated based on the power demand prediction results at different times.

4. The power distribution method for charging piles in a charging station according to claim 3, characterized in that: Inputting the power curve of the charging pile into a pre-established power allocation model, and solving the power allocation model based on an optimization algorithm to obtain the optimal power allocation solution for each charging pile, specifically including: The objective function of the power allocation model is to minimize the total power fluctuation of the charging station. The specific formula is: ， in, is the total power of the charging station at time t, is the average power of the charging station, T represents the total number of time periods.

5. The power distribution method for charging piles in a charging station according to claim 4, characterized in that: The constraint functions of the power allocation model include: ， Among them, P i (t) represents the power of the i-th device, P tatal,max The maximum operating power allowed by the system.

6. The power distribution method for charging piles in a charging station according to claim 5, characterized in that: The power allocation model is pre-established and trained specifically as follows: Initialization parameters: 1) Set the maximum power limit Pmax of the charging station; 2) Set the time step Δt; 3) Obtain the current charging demand Ei of all charging piles; 4) Set the vehicle charging power range: 0≤Pi(t)≤Pi,max where Pi,max is the maximum power of the i-th charging pile; Calculate the objective function: The optimization goal is to minimize the power fluctuation of the charging station: ， in: is the total power at the current time t; is the average power: ; Calculate the current power demand: For each charging station 1) Calculate the remaining charging time: , If Ti is too short, i.e. the vehicle is about to be fully charged, its power allocation priority is lowered; 2) Calculate the total power demand in the current time window: ; 3) Calculate the power fluctuation degree: , adjust the power allocation based on σ(t); Perform dynamic optimization: Optimization is performed using gradient descent: 1) Calculate the adjustment direction If the total power P(t) exceeds the limit, reduce the power of some charging piles with higher power; If the power fluctuation is large, the power curve is smoothed; if some vehicles are low on power, power is allocated first; 2) Structural optimization problem Constraints: , ; 3) Solve using optimization algorithm: Use gradient descent method to solve; Adjust charging power in real time: At each time step t: 1) Update Pi(t) according to the new power allocation scheme calculated by optimization; 2) Calculate the new total power P(t) to ensure that it does not exceed the maximum power; 3) Record the power fluctuations. If the fluctuations are still large, continue optimization.

7. The power distribution method for charging piles in a charging station according to claim 6, characterized in that: The method also includes pre-establishing and training an LSTM neural network model, The input data of the LSTM neural network model is the historical 4-hour charging pile power data P(t); The model output of the LSTM neural network model is the power prediction value for the next 10 minutes, that is, the power prediction curve P(t+1), P(t+2),...P(t+n).

8. A power distribution system for charging piles in a charging station, characterized in that: include, The data acquisition and preprocessing module is used to collect real-time demand information and real-time charging pile status information of the charging pile, and perform data preprocessing on the real-time power demand information and real-time charging pile status information; a power demand prediction module, configured to input the historical data and the real-time demand data of the charging piles into a pre-trained power demand prediction model, output the power demand of each charging pile within a certain period of time in the future through the power demand prediction model, and generate a power curve for each charging pile based on the power demand prediction results at different times; The power distribution module is used to input the power curve of the charging pile into a pre-established power distribution model, and solve the power distribution model based on an optimization algorithm to obtain an optimal power distribution solution for each charging pile.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores program code, and when the program code is executed by a processor, the steps of the power distribution method for charging piles in a charging station are implemented as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: Memory for storing computer programs; The processor is configured to execute the steps of the method for power distribution of charging piles in a charging station as claimed in any one of claims 1 to 7 when executing the computer program stored in the memory.

Images