Control method applied to train battery replacement

Abstract

The invention relates to a control method applied to train battery replacement, and belongs to the technical field of new energy train battery replacement. After the new energy train arrives at the station and confirmation is carried out, preparation and detection of new energy train battery replacement are carried out; the new energy train battery is replaced through the battery replacement robot; and meanwhile, data of each link is recorded, the data is processed to obtain the remaining working time of the battery, then the subsequent operation of the new energy train is optimized, and the battery replacement process of the new energy train is completed. According to the invention, the battery replacement efficiency and safety of the new energy train can be improved, and the emission and energy consumption cost is reduced, so that the wide application of the new energy train in the field of railway traffic is promoted. Through intelligent and automatic battery replacement operation, it can be ensured that the new energy train rapidly and accurately completes battery replacement in the battery replacement process, the energy supplementing time of the new energy train is shortened, and the operation efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of new energy train battery swapping technology, and in particular, a control method for train battery swapping. Background Technology

[0002] Electric trains are a type of rail transport traction equipment. Power comes from a traction motor that drives the wheels to rotate, and the friction between the wheels and the rails propels the train along the track. In this mode of operation, the traction force is limited not only by the power of the traction motor (or diesel engine) but also by the friction between the wheels and the rails. There are limitations on the gradient that trains can travel on; the typical gradient is 3‰, and local gradients cannot exceed 30‰. Based on their structure, they are classified into overhead wire electric trains and battery-powered electric trains.

[0003] The operation of an overhead line electric locomotive is as follows: High-voltage AC power is stepped down and rectified at the traction substation, with the positive terminal connected to the overhead line and the negative terminal connected to the rail. The pantograph on the train contacts the overhead line, introducing current into the train. The current then passes through an automatic air switch, controller, and resistor box to the traction motor, driving the motor to operate. The motor drives the wheels through a transmission device, thus propelling the train. The current flowing from the motor returns to the substation via the rails.

[0004] The working principle of battery-powered electric trains (i.e., new energy trains) is similar to that of overhead wire electric trains. The DC power provided by the battery drives the electric motor through explosion-proof pins, controllers, resistor boxes and other components, thereby traction of the train.

[0005] The battery capacity of most new energy trains is over 1000kWh, and may even reach over 3000kWh in the future. However, the output capacity of ordinary industrial-grade charging piles is currently limited, and the charging time is about 2-5 hours.

[0006] Pure electric new energy trains not only require megawatt-level charging systems but also power transmission and distribution facilities of equal capacity. This necessitates substantial initial investment for expansion, while low full-power utilization during actual operation leads to idle fixed asset investments. To balance the needs of power distribution, charging, and operation, and achieve efficient overall operation, battery swapping solutions for new energy trains have emerged.

[0007] However, although there are many design schemes for battery swapping of new energy trains, no standard has been formed and no implementation has been carried out. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a control method for train battery swapping, improving swapping efficiency and safety, reducing emissions and energy costs, thereby promoting the widespread application of new energy trains in the railway transportation sector. This invention ensures that new energy trains can quickly and accurately complete battery replacement through intelligent and automated battery swapping operations, shortening refueling time and improving operational efficiency. Simultaneously, this invention can also perform real-time battery monitoring and status assessment, ensuring battery safety and stability and avoiding safety hazards caused by battery failure.

[0009] The technical problem solved by this invention is achieved through the following technical solution: A control method for train battery swapping includes the following steps: Step S1, identifying and confirming the arrival of the new energy train; Step S2, after identifying and confirming the arrival, preparing for battery replacement and detecting its status; Step S3, performing the battery replacement operation of the new energy train through a battery swapping robot. Step S4: Record the battery swapping data for each stage in steps S1 to S3, analyze and process the data to obtain the remaining battery working time and battery swapping suggestions, and optimize the subsequent operation scheduling of new energy trains accordingly.

[0010] Furthermore, step S1 includes: setting an RFID identification card with a unique identification code stored on the front of the new energy train; pre-storing the information of each new energy train in the battery swapping station; when the new energy train arrives at the station, the battery swapping station obtains the identification information by reading the RFID identification card and matches it with the pre-stored information; if the match is successful, step S2 is executed; otherwise, an alarm is triggered.

[0011] Furthermore, step S2 includes: setting a battery swapping controller at the bottom of the power battery frame of the new energy train; the battery swapping controller communicates with the battery management system (BMS), vehicle control unit (VCU), and train monitoring system (TMS) via a CAN bus, and establishes a communication connection with the battery swapping station; the battery swapping controller performs status detection on the battery to be replaced to determine whether its working status is normal.

[0012] Furthermore, the status detection includes detecting the battery cell temperature, connector temperature, voltage, and fault status; the method for determining whether the battery is in normal working condition is as follows: compare the detected parameters with a preset safety threshold range. If all parameters are within the threshold range, the battery is determined to be in normal condition, and step S3 is executed; otherwise, the battery is determined to be in abnormal condition, a spare battery is replaced, and step S2 is executed again.

[0013] Furthermore, step S3 includes: the battery swapping station controls the battery swapping robot to perform the replacement of depleted and fully charged batteries based on the identified new energy train model and battery information; during the battery swapping process, the new energy train and the battery swapping station continuously interact and confirm their status; if the status of either party changes and the battery swapping conditions are not met, the battery swapping process is immediately terminated and an alarm is issued.

[0014] Furthermore, in step S4, the battery swapping data is analyzed and processed to obtain the remaining battery operating time and battery swapping suggestions. Specifically, this includes: calculating the remaining battery operating time based on a BP neural network model; and generating battery charging and pairing task orders and new energy train operation task orders based on a mixed integer linear programming (MILP) algorithm, which integrates line busyness data, electricity price data, and battery operation matching data, as battery swapping suggestions.

[0015] Furthermore, the calculation of the remaining battery operating time using the BP neural network model includes: extracting input features, including average discharge power, voltage drop slope, standard deviation of current sequence, cumulative battery cycle count, and indirect health status indicators. Extract the target variables, including the battery voltage, load current, and surface temperature at the current sampling time, as well as the voltage and current sequences over the past N sampling periods; Standardize the input features and target variables; Acquire historical data and divide it into training and testing sets; Construct a BP neural network model and train the model using the training set; Obtain the weights from the hidden layer to the output layer of the trained model; The model's prediction performance is verified using a test set. If the preset accuracy requirements are met, the model is used to calculate the remaining battery operating time; otherwise, an online adaptive module is introduced to continue training the model.

[0016] Furthermore, the generation of battery swapping recommendations includes: Obtain line congestion data A, including new energy train dispatching tasks, electricity consumption forecast list, and correction parameter list B; Obtain electricity prices and a list of available power loads (C); Obtain battery operation matching data table D, including the SOC, SOH, temperature and internal resistance parameters of multiple batteries; use data A, B, C and D as inputs and constraints, solve the problem using the mixed integer linear programming (MILP) algorithm, and output battery charging and grouping task sheet E and new energy train operation task sheet F.

[0017] The advantages and positive effects of this invention are: 1. This invention can improve battery swapping efficiency and safety, reduce emissions and energy consumption costs, thereby promoting the widespread application of electric vehicles in the railway transportation sector; at the same time, through intelligent and automated battery swapping operations, this invention can ensure that new energy trains can quickly and accurately complete battery replacement during the battery swapping process, shortening the refueling time of new energy trains and improving operational efficiency.

[0018] This invention enables real-time monitoring and status assessment of batteries, ensuring their safety and stability and avoiding safety hazards caused by battery failure. Therefore, this invention allows new energy trains to quickly replenish energy through battery swapping while maintaining safety. Furthermore, relying on train battery swapping stations, this invention achieves interconnectivity between megawatt-level high-power charging, battery swapping, and train dispatching, ensuring the smooth operation of new energy trains. Attached Figure Description

[0019] Figure 1 This is a flowchart of the present invention; Figure 2 This is a diagram of the BP neural network structure used in this invention; Figure 3 This is a BP model view of the present invention; Figure 4 This is a schematic diagram illustrating the prediction effect on the test set of the present invention; Figure 5 This is a schematic diagram illustrating the prediction effect of the training set in this invention. Detailed Implementation

[0020] The present invention will be further described in detail below with reference to the accompanying drawings.

[0021] A control method for train battery swapping, such as Figure 1 As shown, it includes the following steps: Step 1: Identification and confirmation of the arrival of new energy trains.

[0022] The specific implementation method of step 1 is as follows: Place an RFID identification card at the front of each new energy train, and write the unique identification code of the train into the RFID identification card. At the same time, input the information of each new energy train into the battery swapping station. After the new energy train arrives at the battery swapping station, the battery swapping station automatically identifies the identity information of the new energy train through radio frequency identification (RFID) technology and matches it with the information of each new energy train that has been entered into the battery swapping station. If the match is successful, proceed to step 2; otherwise, if the new energy train is not a new energy train that has been recorded, an alarm will be triggered.

[0023] Step 2: After recognizing and confirming the arrival of the new energy train, prepare and test the replacement batteries for the new energy train.

[0024] The specific implementation method of step 2 is as follows: Install a battery swapping controller at the bottom of the power battery frame of the new energy train. The battery swapping controller communicates with the battery management system (BMS), the vehicle control unit (VCU), and the battery swapping station, and performs status detection on the battery to be replaced to determine whether the battery is working normally.

[0025] Condition monitoring includes battery cell temperature, connector temperature, voltage, and fault status.

[0026] The specific method for detecting whether the battery is working properly is as follows: compare the detected battery cell temperature, connector temperature, voltage and fault status, battery capacity, system platform voltage, cumulative charge and discharge amount, SOC, SOH of multiple battery boxes, current battery internal resistance, etc. with a preset value range. If the detected data is within the preset value range, the battery is working properly and step 3 is performed; otherwise, the battery is not working properly, another battery is replaced and step 2 is repeated.

[0027] Step 3: Replace the batteries of the new energy train using a battery swapping robot.

[0028] The specific implementation method of step 3 is as follows: The battery swapping station uses a battery swapping robot to replace the depleted battery with a fully charged battery based on the identified new energy train model and battery information. During the battery swapping process, the new energy train will continuously interact and confirm its status with the battery swapping station. If the status of either party changes and the battery swapping conditions are not met, the battery swapping process will be terminated immediately and an alarm will be issued.

[0029] The specific implementation methods for status interaction and confirmation are as follows: The battery swapping station monitors the lock status, auxiliary power signal, and gear signal of the new energy train. If the new energy train is not in the battery swapping state, the battery swapping station immediately switches to the prohibited battery swapping state and stops operating immediately, issuing an alarm signal; otherwise, it proceeds with the normal battery swapping process.

[0030] If the battery swapping station issues a non-battery swapping status signal during the battery swapping process, the new energy train will immediately issue a signal to prohibit battery swapping, report to the battery swapping station, and issue an alarm signal to prompt station personnel to intervene in the battery swapping process to ensure the safety of the battery swapping process; otherwise, the normal battery swapping process will proceed.

[0031] Step 4: Record the data from each step from Step 1 to Step 3, calculate the battery swapping data for each step, obtain the remaining battery working time and battery swapping recommendations, use the remaining battery working time and battery swapping recommendations as the basic operational status of the battery swapping service, optimize the subsequent operation of the new energy train, and complete the battery swapping process of the new energy train.

[0032] Based on different electricity prices and the availability of battery swapping stations at different times, we recommend battery swapping during a specific time period without affecting the operation of customers' vehicles. We then send the recommended time period to the information display screen of the new energy train, which saves costs and ensures the operation of the battery swapping stations.

[0033] Integrate the following key data: (1) Line congestion data: Historical and planned transportation tasks: train timetable, freight volume (tons), transportation origin and destination, and planned departure / arrival times. Real-time location and status: GPS location of new energy trains, current speed, and traction / braking status. Line topology and operating conditions: line gradient, curve radius, station layout and distance. Energy consumption is calculated based on the above parameters.

[0034] Forecasted future demand: Based on business orders and production plans, forecast peak transportation demand for the next few hours or day.

[0035] For example: Planned transportation tasks for the next 3 days: train timetable, freight volume (tons), transportation origin and destination, and dispatching plan, as shown in Table 1. Output new energy train dispatching task A: Table 1

[0036] (2) Based on historical data from the same period, recent and similar business periods: weather temperature, actual operation data, vehicle power consumption, failure rate, etc., as shown in Table 2, output the power consumption forecast list and the correction parameter list B: Table 2

[0037] (3) Electricity price data: Time-of-use pricing refers to the electricity price at different times of the day (valley, average, peak, and high-peak). Based on the plant's electricity load, the electricity price and available power load list C are output.

[0038] (4) Battery status data: Based on the SOC (State of Charge) of multiple batteries in the operating system: the real-time remaining capacity of each new energy train battery. SOH (State of Health): battery health status, affecting actual usable capacity and charging efficiency. Temperature, internal resistance, cumulative charge / discharge, maintenance requirements, and maintenance history: used to assess battery safety, charge acceptance, and matching battery pack schemes. Output battery operation matching data table D.

[0039] (5) Operational planning Using data such as the new energy train scheduling task A, the electricity consumption forecast list and correction parameter list B, the electricity price and available power load list C, and the battery operation matching data table D as output data and constraints, the mixed integer linear programming (MILP) algorithm is used to output the battery charging and grouping task sheet E and the new energy train operation task sheet F.

[0040] The costs and constraints of new energy trains are transformed into linear expressions, which are then solved using CPLEX or Gurobi to obtain output battery charging and grouping task orders and new energy train operation task orders, thereby optimizing the subsequent operation of new energy trains. It integrates all information to determine "when," "which new energy train," and "which battery to replace." Through this intelligent scheduling, a shift from "passive battery swapping" (swapping only when power is depleted) to "proactive optimization" (swapping batteries at the right time and in the most economical way) can be achieved, ultimately significantly improving operational reliability and economy.

[0041] The specific calculation method for the remaining battery working time is as follows: After the battery swapping robot completes the battery swapping, the battery swapping controller automatically records the battery swapping time, the status of the original battery and the replacement battery, as well as the operating status of the battery swapping robot, and transmits the information to the data center for storage and analysis. Based on the analysis results, the system optimizes and guides the subsequent operation of the new energy train.

[0042] This invention calculates the remaining battery operating time using a BP neural network: like Figure 2 and Figure 3 As shown, a backpropagation (BP) neural network is a typical nonlinear algorithm. A BP neural network consists of an input layer, hidden layers (also called intermediate layers), and an output layer, where there is one or more hidden layers. Each layer can have a number of nodes. The connection states between nodes in different layers are represented by weights.

[0043] (1) Extracting input features and target variables: The input layer of the BP neural network of this invention is not a simple instantaneous parameter, but a carefully constructed multidimensional dynamic feature vector.

[0044] The target variables include: Real-time static parameters: battery voltage V(t), load current I(t), and surface temperature T(t) at the current sampling time.

[0045] Historical time series fragments: voltage sequence {V(t), V(t-1), ..., V(tN)} and current sequence {I(t), I(t-1), ..., I(tN)} over the past N sampling periods.

[0046] The input features include: feature values ​​calculated based on the aforementioned historical time series segments. Average discharge power P_avg.

[0047] Voltage drop slope ΔV / Δt.

[0048] The standard deviation I_std of the current series is used to reflect load fluctuations.

[0049] The cumulative cycle count of a battery serves as a macroscopic indicator of its historical aging status.

[0050] Indirect State of Health (SOH) indicator: Current battery internal resistance R_internal, obtained through online identification or table lookup. Target variable: Remaining battery operating time.

[0051] (2) Standardize input features and target variables: Z-score standardization standardizes data based on the mean and standard deviation of the original data. Its purpose is to unify data of different magnitudes into the same magnitude. Z=(X-μ) / σ Where X is the original data, μ is the mean of the data, and σ is the standard deviation of the data.

[0052] The mean is defined as the sum of all values ​​in a dataset divided by the number of values ​​in the dataset. The formula for calculating the mean is: $$ bar {x} = frac {1} {n} sum {i=1}^ {n} xi $$ Where $x_i$ represents each value in the dataset, and $n$ represents the number of values ​​in the dataset.

[0053] (3) Divide the training set and the test set: Data collection began when the SOC of the new energy train reached 100%. All collected data were divided into two sets: the first 80% was used as the training set, and the last 20% was used as the test set.

[0054] (4) Create and train the BP neural network model: Create 10 hidden layer neurons using MATLAB. 10 neurons are especially suitable for problems with medium complexity, balancing complexity and generalization: net=feedforwardnet(10); For many small to medium-sized datasets and network architectures, 1000 iterations are usually sufficient to allow the network weights to be adjusted to near the optimal solution. This ensures that the model has enough opportunity to learn: net.trainParam.epochs=1000; The error threshold 1e-6 is a very small number, meaning that a high level of accuracy is required for the model's predictions. Setting such a strict objective is to force the network to fit the subtle patterns in the training data as closely as possible: net.trainParam.goal=1e-6; A learning rate of 0.01 is a relatively conservative and stable initial value. It strikes a good balance between convergence speed and stability, making it a safe and efficient choice for many problems: net.trainParam.lr=0.01.

[0055] (5) Obtaining weights: The weights from the hidden layer to the output layer are: -0.3780, -0.0717, 0.4603, 0.0462, -0.4166, -0.3771, 0.4365, -0.3390, 0.5144, -0.5446.

[0056] (6) Detection results: like Figure 4 and Figure 5 As shown, the calculation results are as follows: mean square error = 0.0052744, coefficient of determination = 0.71715, mean absolute error = 0.055906, and mean relative error = 6.4498.

[0057] (7) Data prediction verification: At a certain moment, the SOC = 22.8%, the vehicle speed = 26 km / h, the total battery current = 183A, the battery internal voltage = 591.1V, and the current BMS voltage = 607.5V. After the above training, the BP neural network predicts that the remaining battery working time is 0.63 hours, and recommends that the driver go to the battery swapping station immediately.

[0058] (8) This invention adds an online adaptive function to the traditional BP neural network, so that the model keeps learning and becomes more and more accurate the more it is used.

[0059] Because battery performance slowly degrades with increasing cycle count, fixed models cannot capture this long-term change, leading to increasingly inaccurate predictions. Online adaptation enables models to track the battery's degradation trajectory. Furthermore, even batteries of the same model will experience different charging and discharging habits (such as rapid acceleration and fast charging) in different users. Online adaptation allows models to better adapt to the unique usage patterns of users.

[0060] 1. Real-time prediction: The deployed lightweight model receives real-time feature vectors X(t) from the sensors and outputs the predicted remaining working time RWL_pred(t).

[0061] 2. Data recording and caching: The system continuously records X(t) and its corresponding real timestamp.

[0062] 3. Tag generation: When a complete discharge cycle ends, the system can automatically backtrack and calculate the "true" remaining working time RWL_true(t) for each historical data point X(t) in the cache.

[0063] RWL_true(t) = T_end - t Where T_end is the time point at which this discharge ends, and t is the timestamp of the historical data point.

[0064] 4. New dataset formation: Pair (X(t), RWL_true(t)) to form new training samples with real labels, and store them in a fixed-size first-in-first-out buffer.

[0065] 5. Trigger Model Update: The system triggers a model update process when the number of new samples in the buffer reaches a preset threshold (e.g., data accumulated over 10 complete discharge cycles), or at a fixed time interval (e.g., every month). Updates are only performed when the distribution of the new dataset does not drastically change from that of the old dataset, preventing the model from being "skewed" by sensor malfunctions or extreme abnormal operating conditions.

[0066] Based on the above-mentioned control method for train battery swapping, the effectiveness of the invention was verified by performing a battery swapping operation on a new energy train.

[0067] Step 1: After the train arrives at the battery swapping station, the station system automatically identifies the train's model and serial number using RFID technology. Once the vehicle identification and battery type are successfully matched, a confirmation message is sent to the driver or dispatch center to prepare for the battery swapping operation.

[0068] Step 2: The battery swapping robot accurately identifies and locates the new energy train whose battery needs to be replaced using visual recognition and laser positioning technology.

[0069] Step 3: The battery swapping controller establishes a WIFI connection with the battery swapping station via a router.

[0070] Step 4: The battery swapping controller determines that the battery to be replaced is in good condition and without faults, and then sends the battery information to the battery swapping station. At the same time, it checks that the current backup battery meets the conditions for battery swapping with sufficient power and without faults, and then sends a request to the battery swapping station to allow battery swapping.

[0071] Step 5: The battery swapping robot moves to the designated location, picks up the depleted battery, and removes it from the vehicle. At this time, the battery swapping controller detects that the battery interface connection is tight and the battery condition is good, ensuring the safety and reliability of the replacement process.

[0072] Step 6: The battery swapping robot moves the fully charged battery to the designated location and precisely installs it onto the vehicle. During installation, the battery swapping controller monitors the battery interface connection status to prevent the battery from falling off or being damaged during use.

[0073] Step 7: After the new battery is installed, the battery swapping controller will perform a series of detection operations, including battery voltage detection, current detection, and temperature detection, to ensure that the new battery can work normally and meet the vehicle's operating requirements.

[0074] Step 8: The battery swapping controller will also record the battery swapping log, replacement results, and vehicle status information for subsequent management and maintenance, and will synchronously transmit the information to the battery swapping station for recording and storage.

[0075] It should be emphasized that the embodiments described in this invention are illustrative rather than limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementations derived by those skilled in the art based on the technical solutions of this invention are also within the scope of protection of this invention.

Claims

1. A control method for train battery swapping, characterized in that, Includes the following steps: Step S1: Identify and confirm the arrival of the new energy train; Step S2: After identifying and confirming the arrival, prepare for battery replacement and perform status detection; Step S3: Perform the battery replacement operation of the new energy train through the battery swapping robot. Step S4: Record the battery swapping data for each stage in steps S1 to S3, analyze and process the data to obtain the remaining battery working time and battery swapping suggestions, and optimize the subsequent operation scheduling of new energy trains accordingly.

2. The control method for train battery swapping according to claim 1, characterized in that, Step S1 includes: setting an RFID identification card with a unique identification code on the front of the new energy train; pre-storing the information of each new energy train in the battery swapping station; when the new energy train arrives at the station, the battery swapping station obtains the identification information by reading the RFID identification card and matches it with the pre-stored information; if the match is successful, step S2 is executed; otherwise, an alarm is triggered.

3. The control method for train battery swapping according to claim 1, characterized in that, Step S2 includes: setting a battery swapping controller at the bottom of the power battery frame of the new energy train; the battery swapping controller communicates with the battery management system (BMS), vehicle control unit (VCU), and train monitoring system (TMS) via a CAN bus, and establishes a communication connection with the battery swapping station; the battery swapping controller performs status detection on the battery to be replaced to determine whether its working status is normal.

4. The control method for train battery swapping according to claim 3, characterized in that, The status detection includes detecting the battery cell temperature, connector temperature, voltage, and fault status. The method for determining whether the battery is in normal working condition is as follows: compare the detected parameters with a preset safety threshold range. If all parameters are within the threshold range, the battery is determined to be in normal condition, and step S3 is executed; otherwise, the battery is determined to be in abnormal condition, a spare battery is replaced, and step S2 is executed again.

5. The control method for train battery swapping according to claim 1, characterized in that, Step S3 includes: the battery swapping station controls the battery swapping robot to perform the replacement of depleted and fully charged batteries based on the identified new energy train model and battery information; during the battery swapping process, the new energy train and the battery swapping station continuously interact and confirm the status; if the status of either party changes and the battery swapping conditions are not met, the battery swapping process is immediately terminated and an alarm is issued.

6. The control method for train battery swapping according to claim 1, characterized in that, In step S4, the battery swapping data is analyzed and processed to obtain the remaining battery working time and battery swapping suggestions. Specifically, this includes: calculating the remaining battery working time based on a BP neural network model; and generating battery charging and pairing task orders and new energy train operation task orders based on a mixed integer linear programming (MILP) algorithm, which integrates line busyness data, electricity price data, and battery operation matching data, as battery swapping suggestions.

7. A control method for train battery swapping according to claim 6, characterized in that, The calculation of the remaining battery operating time using a BP neural network model includes: extracting input features, including average discharge power, voltage drop slope, standard deviation of current sequence, cumulative battery cycle count, and indirect health status indicators; Extract the target variables, including the battery voltage, load current, and surface temperature at the current sampling time, as well as the voltage and current sequences over the past N sampling periods; Standardize the input features and target variables; Acquire historical data and divide it into training and testing sets; Construct a BP neural network model and train the model using the training set; Obtain the weights from the hidden layer to the output layer of the trained model; The model's prediction performance is verified using a test set. If the preset accuracy requirements are met, the model is used to calculate the remaining battery operating time; otherwise, an online adaptive module is introduced to continue training the model.

8. A control method for train battery swapping according to claim 6, characterized in that, The generated battery swapping recommendations include: Obtain line congestion data A, including new energy train dispatching tasks, electricity consumption forecast list, and correction parameter list B; Obtain electricity prices and a list of available power loads (C); Obtain battery operation matching data table D, including the SOC, SOH, temperature and internal resistance parameters of multiple batteries; use data A, B, C and D as inputs and constraints, solve the problem using the mixed integer linear programming (MILP) algorithm, and output battery charging and grouping task sheet E and new energy train operation task sheet F.

Images