Battery examination and vehicle network interaction control method based on SSA and Autoformer

Abstract

The invention discloses a battery examination and vehicle network interaction control method based on SSA and Autoformer, and relates to the technical field of power system regulation and control. According to the method, charging data of the electric vehicle are collected in real time through a high-precision battery charging information collection device and transmitted to an upper computer for analysis. And utilizing SSA to optimize the BHI to accurately evaluate the health condition of the battery, and generating health early warning information. And constructing a correlation model of the battery health state and the vehicle network interaction behavior based on the BHI, and optimizing an interaction strategy between the electric vehicle and the power grid. And processing historical charging data by adopting an Autoformer model, and predicting a charging and discharging depth adjustment coefficient and an interaction time shortening coefficient in a vehicle network interaction process. And generating a battery maintenance strategy according to the BHI, and dynamically adjusting the charging and discharging power and the interaction time interval, so that the accuracy of battery health assessment and the intelligent level of vehicle network interaction are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system regulation, and in particular to a battery health check and vehicle-grid interaction control method based on SSA and Autoformer. Background Art

[0002] With the increasing popularity of electric vehicles, battery health monitoring and vehicle-to-grid (V2G) control technologies are playing an increasingly important role in the coordinated development of electric vehicles and smart grids. Effective battery state-of-health (SOH) monitoring not only extends battery life and improves safety, but also reduces grid risks and promotes secondary utilization by optimizing V2G scheduling, ultimately maximizing the economic, reliable, and environmental benefits of battery technology and energy systems.

[0003] However, existing technologies have significant deficiencies in charging information collection, battery health assessment, and vehicle-grid interactive control. First, the frequency of charging information collection is low and lacks sufficient granularity, making it impossible to capture instantaneous changes in current, voltage, etc. during battery charging. In addition, there are problems with incomplete and standardized data, resulting in incomplete monitoring of charging behavior and battery status. Secondly, existing battery health assessment methods mainly rely on battery capacity decay curves, ignoring the influence of environmental factors and usage habits, resulting in insufficient accuracy in battery health assessment and lack of support from intelligent optimization algorithms. Finally, existing vehicle-grid interactive control strategies fail to fully consider the health status of batteries, which may lead to excessive discharge of unhealthy batteries, shorten battery service life, and also increase safety risks of the power grid. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this paper proposes a battery health check and vehicle-to-grid interaction control method based on SSA and Autoformer. This method improves the accuracy and frequency of electric vehicle charging information collection by enhancing the precision and completeness of data. It also utilizes SSA to optimize the Battery Health Index (BHI) evaluation metric for precise battery health assessment. A BHI-based vehicle-to-grid safety interaction model is constructed to optimize V2G power scheduling, ensuring a balance between battery health and grid scheduling. Finally, a BHI-based battery maintenance and V2G control strategy is proposed to extend battery life and improve V2G efficiency.

[0005] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0006] A battery health check and vehicle-grid interaction control method based on SSA and Autoformer includes the following steps:

[0007] Step 1: Collect battery charging information through the electric vehicle charging information collection device, and transmit the battery charging information to the host computer through the WIFI module;

[0008] Step 2: Construct the battery health index (BHI) and use the salp swarm algorithm to optimize the indicator weights to evaluate the battery health status. The battery health index is calculated based on the predetermined health assessment indicators and health warning information is generated.

[0009] Step 3: Based on the battery health index (BHI), a correlation model between battery health status and vehicle-grid interaction behavior is established to analyze the relationship between battery charge and discharge depth, interaction time interval, and interaction depth, and determine the interaction strategy between the electric vehicle and the grid;

[0010] Step 4: Use the Autoformer model to process historical charging data and predict the charge and discharge depth adjustment coefficient and interaction time reduction coefficient during the vehicle-grid interaction process;

[0011] Step 5: Generate a battery maintenance strategy based on the battery health index BHI and control the behavior of the electric vehicle in participating in the grid interaction, including adjusting the charging and discharging power and setting the interaction time interval.

[0012] In one embodiment, the electric vehicle charging information collection device is integrated into a charging pile and includes:

[0013] CAN transceiver terminal, used to obtain electric vehicle charging information in the form of CAN messages through the CAN communication interfaces S+ and S- of the charging gun;

[0014] The main control chip receives and parses the message through the CAN transceiver, generates a formatted data packet, and communicates with the WIFI module through the SCI serial port;

[0015] Power supply motherboard, used to power the CAN transceiver, main control chip and WIFI module;

[0016] The expression of the battery charging information is:

[0017] X C =[X1,X2,X3,X4,X5,X6,X7,X8,X9,X 10 ]

[0018] Where: X c The collected battery charging information, X1 is the battery type, X2 is the battery capacity, X3 is the battery rated voltage, X4 is the charging voltage measurement value, X5 is the charging current measurement value, X6 is the highest single cell voltage, X7 is the current state of charge SOC, X8 is the estimated remaining charging time, X9 is the cumulative charging time, X 10 The maximum battery temperature.

[0019] In one embodiment, the battery health index (BHI) is calculated using the following formula:

[0020] BHI=ω1×SOC diff +ω2×V diff +ω3×T norm +ω4×C eff +ω5×R int ;

[0021] In the formula, ω1, ω2, ω3, ω4, and ω5 are indicator weights; SOC diff is the SOC change rate; V diff is the voltage difference of single battery; T norm is the temperature normalized index; C eff is the capacity attenuation coefficient; R int is the rate of change of battery internal resistance;

[0022] in:

[0023]

[0024]

[0025] Where X7(t) is the current state of charge SOC at time t, X7(t-1) is the current state of charge SOC at time t-1; T ref is the reference temperature under normal working conditions, T max is the maximum allowable temperature; C est The estimated current maximum available capacity is calculated based on the accumulated charge and discharge data;

[0026] The remaining battery life RUL is calculated based on the battery health index BHI, using the expression:

[0027]

[0028] Where SOH = BHI·100%, and the average daily discharge is calculated from historical data.

[0029] In one embodiment, the method of optimizing and calculating the indicator weights using the salp swarm algorithm includes:

[0030] Initialize population parameters: set population size N, number of iterations T max , randomly initialize N groups of weights, expression:

[0031] W i =(ω i1 ,ω i2 ,ω i3 ,ω i4 ,ω i5 ),i=1,2,3,4,5;

[0032] Define the fitness function F(W) expression:

[0033]

[0034] Update the leader weight expression:

[0035] W new =W best +c1×r1×(W best -W);

[0036] Update the follower weight expression:

[0037]

[0038] The weights are normalized to satisfy the expression:

[0039] Repeat the steps until T is reached max or convergence conditions;

[0040] Where: i1 ,ω i2 ,ω i3 ,ω i4 ,ω i5 is the weight coefficient, which is used to balance the importance of each indicator, F(W) is the fitness function, BHI actual,t The real BHI value calculated for historical data, BHI pred,t is the BHI value calculated based on the current weight, W is the weight combination set in the current population, and W new is the latest weight combination in the current population, W best is the optimal weight combination in the current population, W i is the i-th weight combination, r1 is a random number between [0,1], c1 is the exploration step coefficient, which controls the convergence speed, c2 is the adjustment factor, which controls the follower step size, r2 is a random number between [0,1], ω k is the kth weight coefficient.

[0041] In one embodiment, the steps of establishing a correlation model between battery health status and vehicle-grid interaction behavior are as follows:

[0042] Remaining battery charge and discharge times T s , the expression is:

[0043] T s =T init ×BHI-T used ;

[0044] Where: T initis the initial rated cycle number of the battery, T used The number of charge and discharge times that have been used;

[0045] The expression of charge and discharge depth is:

[0046]

[0047] Where: P c is the depth of charge, P d is the discharge depth, P c,min The lowest charge depth of the battery (e.g. 10%), P d,min is the minimum discharge depth of the battery (e.g. 10%), α c ,α d It is the adjustment coefficient of charge and discharge depth, usually set based on experience or experimental data;

[0048] Vehicle-grid interaction time interval T c expression:

[0049] T c =T c,max -β×(1-BHI);

[0050] Where: T c,max is the maximum vehicle-grid interaction time interval (minutes), β is the vehicle-grid interaction time reduction coefficient (empirical parameter);

[0051] Charge and discharge power expression:

[0052]

[0053] Where: P max is the maximum charge and discharge power.

[0054] In one embodiment, the use of the Autoformer model to process historical charging data specifically includes:

[0055] Y t =F(X t )+ε t ;

[0056] Where: Y t is the output variable of the future vehicle-grid interaction coefficient, ε t is the prediction error, F represents the prediction function determined by the Autoformer model, X t is the input variable of the historical vehicle-grid interaction coefficient;

[0057] Y t =(α c,t+1 ,α d,t+1 ,β t+1 )

[0058] Where: α c,t+1 is the predicted charging depth adjustment coefficient, α d,t+1 is the predicted discharge depth adjustment coefficient, β t+1 is the predicted vehicle-grid interaction time reduction coefficient;

[0059] X t =(BHI t ,P c,t ,P d,t ,T c,t ,T bat,t ,P EV,t ,α c,t ,α d,t ,β t );

[0060] Of which: BHI t is the battery health index, P c,t is the current charging depth, P d,t is the current discharge depth, T c,t is the current vehicle-grid interaction time interval, T bat,t is the current battery temperature, P EV,t is the current charging and discharging power of the electric vehicle, α c,t is the current charging depth adjustment coefficient, α d,t is the current discharge depth adjustment coefficient, β t is the current vehicle-grid interaction time reduction coefficient.

[0061] In one embodiment, the specific steps of predicting the charge and discharge depth adjustment coefficient and the interaction time shortening coefficient during the vehicle-grid interaction process include:

[0062] Through the autocorrelation mechanism and recursive decomposition module of the Autoformer model, the long-term and short-term features of the input data are extracted;

[0063] The expression for generating the prediction output is:

[0064] Y t =Μ Autoformer (X t );

[0065] Where: Y t For the future vehicle network interaction coefficient, X t is the historical vehicle-grid interaction coefficient;

[0066] The analytical prediction result expression is:

[0067]

[0068] Where: α c,t+1 is the predicted charging depth adjustment coefficient, αd,t+1 is the predicted discharge depth adjustment coefficient, β t+1 is the predicted vehicle-grid interaction time reduction coefficient, M Autoformer represents the Autoformer prediction model, f Autoformer , g Autoformer , h Autoformer To be more specific, the Autoformer prediction model;

[0069] The predicted adjustment coefficient α c,t+1 , α d,t+1 , β t+1 Substituting the following formula, the expression for dynamically adjusting the charge and discharge depth and interaction time interval of the next cycle is:

[0070]

[0071] Of which: BHI t is the battery health index, P c,t+1 is the predicted depth of charge, P d,t+1 is the predicted discharge depth, T c,t+1 is the predicted vehicle-grid interaction time interval, P c,min , P d,min is the minimum charge and discharge depth, T c,max is the maximum vehicle-grid interaction time interval.

[0072] In one embodiment, the battery health index (BHI) generates a battery maintenance strategy, including the following steps:

[0073] A battery maintenance strategy is generated based on the battery health index BHI. The strategy is classified according to the BHI value as follows:

[0074]

[0075] The corresponding maintenance strategy is:

[0076] When BHI>0.8, the charging and discharging strategy remains unchanged, allowing high-frequency vehicle-grid interaction;

[0077] When 0.5≤BHI≤0.8, limit the charge and discharge depth to P c,min and P d,min , and reduce the frequency of interaction;

[0078] When BHI≤0.5, only shallow charging and discharging are allowed, and the vehicle-grid interaction is limited to the lowest frequency.

[0079] In one embodiment, adjusting vehicle-grid interaction control parameters according to the BHI includes:

[0080] Interaction time interval T c The expression is:

[0081] T c =T c,max -β×(1-BHI);

[0082] Where: BHI is the battery health index, T c is the interaction time interval, β is the interaction time adjustment coefficient, the default value range is 10-30 minutes, T c,max It is the pre-set maximum interaction time interval;

[0083] The calculation formula for charge and discharge depth is as follows:

[0084] P c =α c ×(1-BHI)+P c,min

[0085] P d =α d ×(1-BHI)+P d,min

[0086] Where: P c , P d is the charge and discharge depth, α c , α d is the charge and discharge depth adjustment coefficient, P c,min , P d,min The minimum charge and discharge depth (usually 10%-20%);

[0087] α c and α d According to BHI classification expression:

[0088]

[0089] Interaction time interval T c The specific adjustment is calculated by the following formula:

[0090]

[0091] Where: T max is the actual maximum interaction time interval.

[0092] Compared with the existing technology, the present invention has the following beneficial effects: by integrating multiple innovative technologies, the present invention significantly improves the overall performance of electric vehicle charging and vehicle-grid interaction systems. First, a battery health index (BHI) is constructed, and the salp swarm algorithm is used to optimize the indicator weights, thereby more accurately assessing the battery health status, detecting potential faults in advance, and effectively extending the battery life. Second, a correlation model between battery health status and vehicle-grid interaction behavior is established to achieve dynamic matching of battery status and interaction strategy, avoid damage to the battery caused by excessive charging and discharging, and improve grid stability and efficiency. In addition, an autoformer model is used to process historical charging data, predict the charge and discharge depth adjustment coefficient and the interaction time reduction coefficient, optimize charging plans, and reduce waiting time and energy waste. In terms of user experience, accurate battery health assessment and vehicle-grid interaction control provide users with more reliable charging services, improving charging efficiency and convenience. At the same time, by optimizing vehicle-grid interaction strategies and charging prediction capabilities, electric vehicles can be more effectively utilized as distributed energy storage resources to balance grid load and improve energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0094] in:

[0095] Figure 1 Schematic diagram of the method flow of the present invention;

[0096] Figure 2 This is a functional diagram of an electric vehicle charging information collection device according to an embodiment of the present invention;

[0097] Figure 3 Schematic diagram of the connection of the acquisition device in an embodiment of the present invention. DETAILED DESCRIPTION

[0098] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0099] like Figure 1-Figure 3FIG. 1 is an embodiment of the present invention, which provides a battery health check and vehicle-grid interaction control method based on SSA and Autoformer, including the following steps:

[0100] 1. Electric vehicle charging information collection

[0101] Step 1: Install the charging information collection device

[0102] A dedicated charging information collection device is installed inside the charging pile. This device is like a "smart" box that can collect various information about electric vehicle batteries in real time. It mainly consists of three parts:

[0103] CAN transceiver terminal: This is the "ear" of the device. Through the CAN communication interface on the charging gun (just like the USB interface we commonly use, but specifically for internal communication in the car), it can "hear" the charging information sent by the electric car. This information is transmitted in the form of a CAN message, which is similar to the email we send, but in a more standardized format.

[0104] The main control chip: This is the "brain" of the device, responsible for receiving and parsing the messages "heard" by the CAN transceiver terminals. The main control chip is like a smart translator, able to translate the "code"-like messages sent by the electric vehicle into formatted data packets that we can understand.

[0105] WIFI module: This is the "mouth" of the device. It sends the data packets translated by the main control chip to our host computer (which can be understood as the computer that controls the entire system) through the wireless WIFI network. In this way, we can view the charging status of the electric vehicle battery in real time on the computer.

[0106] Step 2: Collect battery charging information

[0107] When an electric vehicle starts charging, the charging information collection device starts working. It collects the following important battery charging information:

[0108] Battery type (for example, lithium or lead-acid)

[0109] Battery capacity (how much charge the battery can store)

[0110] Battery rated voltage (voltage when the battery is working normally)

[0111] Charging voltage measurement value (current charging voltage)

[0112] Charging current measurement value (current during charging)

[0113] Maximum single cell voltage (the voltage of the battery with the highest voltage in the battery pack)

[0114] Current state of charge (SOC, indicating the percentage of battery power remaining)

[0115] Estimated remaining charging time (how long it will take to fully charge)

[0116] Cumulative charging time (how long has it been charged)

[0117] Maximum battery temperature (the maximum temperature of the battery during charging)

[0118] This information is like a comprehensive "physical examination" of the battery, allowing us to clearly understand the various states of the battery during the charging process.

[0119] Step 3: Optimize indicator weights using the Salp Swarm Algorithm

[0120] To more accurately assess the battery's health status, we need to find an optimal set of indicator weights. This is where an optimization algorithm called the Salp Swarm Algorithm (SSA) comes in.

[0121] The SSA algorithm is an optimization algorithm based on swarm intelligence. Just like a group of salps (a type of marine creature) searching for food together, it continuously tries and adjusts to find the optimal solution. In this case, the SSA algorithm helps us find a set of indicator weights that enables the BHI to most accurately reflect the battery's health status.

[0122] Specifically, the SSA algorithm:

[0123] Initialize population parameters: Just like a group of salps randomly distributed in the ocean, we randomly initialize a set of indicator weights as the initial population.

[0124] Define a fitness function: Just as we set a goal for the salp (such as finding the most food), we define a fitness function to evaluate the weighting of each set of indicators. The fitness function compares the calculated BHI with the actual health of the battery and generates a score.

[0125] Updating leader and follower weights: In a salp swarm, there are always some "leaders" that lead the pack. In the SSA algorithm, we update the positions of the leader (optimal metric weight) and followers (other metric weights) based on the results of the fitness function, gradually moving the entire swarm toward the optimal solution.

[0126] Through continuous iteration and optimization, the SSA algorithm will eventually find an optimal set of indicator weights, enabling BHI to most accurately assess the battery's health status.

[0127] 3. Establishment of a Correlation Model for Vehicle-Internet Interaction Behavior

[0128] Step 1: Analyze the relationship between battery charge and discharge depth, interaction interval, and interaction depth. After establishing the battery health index (BHI), we can begin to build a correlation model between battery health and vehicle-grid interaction behavior. This model acts as a bridge, connecting the battery health status and the interaction between the electric vehicle and the grid.

[0129] First, we need to analyze the relationship between the battery's charge / discharge depth, interaction interval, and interaction depth. The charge / discharge depth refers to the degree of change in the battery's charge during the charge / discharge process; the interaction interval refers to the time interval between the electric vehicle and the grid (such as charging or discharging); and the interaction depth refers to the amount of energy interacting between the electric vehicle and the grid.

[0130] By analyzing these relationships, we can understand what charging and discharging strategies and interactive behaviors should be adopted when the battery is in different health states to ensure the safety and life of the battery.

[0131] Step 2: Determine the interaction strategy between electric vehicles and the grid

[0132] Based on the above analysis, we can determine the interaction strategy between electric vehicles and the power grid. This strategy will take into account multiple factors, such as:

[0133] The remaining number of charge and discharge cycles of the battery (how many charge and discharge cycles the battery can still undergo)

[0134] Depth of charge and discharge (amount of charge per charge and discharge)

[0135] Vehicle-grid interaction time interval (the time interval for interaction between electric vehicles and the power grid)

[0136] Charging and discharging power (the amount of electric power interacting between electric vehicles and the power grid)

[0137] By comprehensively considering these factors, we can develop an interactive strategy that can ensure battery safety and lifespan while maximizing the utilization of grid resources.

[0138] 4. Charging, discharging and interactive prediction

[0139] Step 1: Use the Autoformer model to process historical charging data

[0140] To more accurately predict future charge and discharge depths and interaction intervals, we need to employ an advanced forecasting model—the Autoformer model. The Autoformer model is a deep learning model specifically designed for processing time series data. It automatically extracts both long-term and short-term features from the data, enabling more accurate predictions.

[0141] First, we collect a large amount of historical charging data, including battery charge and discharge depth, interaction time interval, battery health index (BHI), etc. Then, we input this data into the Autoformer model, allowing the model to automatically learn the rules and patterns in the data.

[0142] Step 2: Predict the charge and discharge depth adjustment coefficient and interaction time reduction coefficient

[0143] After the Autoformer model has learned historical data, it can be used to predict future charge-discharge depth adjustment coefficients and interaction time reduction coefficients. These coefficients act like "adjustment instructions" for future charge-discharge strategies and interaction behaviors, dynamically adjusting based on battery health and grid demand.

[0144] Specifically, the Autoformer model extracts long-term and short-term features from input data through an autocorrelation mechanism and a recursive decomposition module, then generates a prediction output. This prediction output indicates the appropriate charge and discharge depth and interaction interval to ensure battery safety and grid stability within a specific timeframe.

[0145] Step 3: Dynamically adjust the charge and discharge depth and interaction time interval for the next cycle

[0146] Based on the Autoformer model's predictions, we can dynamically adjust the charge / discharge depth and interaction interval for the next cycle. If we predict a decline in battery health, we can appropriately reduce the charge / discharge depth or extend the interaction interval to protect battery safety and lifespan. Conversely, if we predict an increase in grid demand, we can appropriately increase the charge / discharge depth or shorten the interaction interval to maximize grid resource utilization.

[0147] 5. Battery Maintenance and Vehicle-Network Interaction Control

[0148] Step 1: Generate a battery maintenance strategy based on the battery health index (BHI)

[0149] Finally, based on the Battery Health Index (BHI), we can generate a corresponding battery maintenance strategy. The maintenance strategy will be classified according to the BHI value and take different maintenance measures. For example:

[0150] If the BHI value is high (indicating good battery health), we can keep the normal charge and discharge strategy unchanged.

[0151] If the BHI value is low (indicating a decrease in battery health), we can limit the depth of charge and discharge and reduce the frequency of interaction to protect the safety and life of the battery.

[0152] If the BHI value is very low (indicating a serious decline in battery health), we may only be able to allow the battery to be shallowly charged and discharged, and limit vehicle-grid interaction to the lowest frequency, or even stop the interaction.

[0153] Step 2: Adjust vehicle-network interaction control parameters according to maintenance strategy

[0154] Based on the generated maintenance strategy, we can adjust the control parameters for vehicle-grid interaction. These parameters include interaction interval and charge / discharge depth. By adjusting these parameters, we can implement the battery maintenance strategy and ensure battery safety and lifespan.

[0155] Step 3: Control the behavior of electric vehicles in grid interaction

[0156] Ultimately, based on the adjusted vehicle-grid interaction control parameters, we can control the EV's interaction with the grid. This includes adjusting charging and discharging power and setting interaction intervals. Through precise control, we can achieve safe and efficient interaction between EVs and the grid, protecting battery safety and lifespan while maximizing grid resource utilization.

[0157] In summary, in the description of the present invention in this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the described specific features, structures, materials, or characteristics may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and features of different embodiments or examples, unless they are mutually inconsistent.

[0158] Any process or method description in the flowchart or otherwise described herein can be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process. The scope of the preferred embodiments of the present application includes additional implementations in which the functions may be performed in a different order than shown or discussed, including in a substantially simultaneous manner or in a reverse order depending on the functions involved.

[0159] The above description is merely a specific embodiment 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 modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A battery health check and vehicle-grid interaction control method based on SSA and Autoformer, characterized in that: The method comprises the following steps: Step 1: Collect battery charging information through the electric vehicle charging information collection device, and transmit the battery charging information to the host computer through the WIFI module; Step 2: Construct the battery health index (BHI) and use the salp swarm algorithm to optimize the indicator weights to evaluate the battery health status. The battery health index is calculated based on the predetermined health assessment indicators and health warning information is generated. Step 3: Based on the battery health index (BHI), a correlation model between battery health status and vehicle-grid interaction behavior is established to analyze the relationship between battery charge and discharge depth, interaction time interval, and interaction depth, and determine the interaction strategy between the electric vehicle and the grid; Step 4: Use the Autoformer model to process historical charging data and predict the charge and discharge depth adjustment coefficient and interaction time reduction coefficient during the vehicle-grid interaction process; Step 5: Generate a battery maintenance strategy based on the battery health index BHI and control the behavior of the electric vehicle in participating in the grid interaction, including adjusting the charging and discharging power and setting the interaction time interval.

2. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 1 is characterized in that: In step 1, the electric vehicle charging information collection device is integrated into the charging pile, including: CAN transceiver terminal, used to obtain electric vehicle charging information in the form of CAN messages through the CAN communication interfaces S+ and S- of the charging gun; The main control chip receives and parses the message through the CAN transceiver, generates a formatted data packet, and communicates with the WIFI module through the SCI serial port; Power supply motherboard, used to power the CAN transceiver, main control chip and WIFI module; The expression of the battery charging information is: X C =[X1,X2,X3,X4,X5,X6,X7,X8,X9,X 10 ] Where: X c The collected battery charging information, X1 is the battery type, X2 is the battery capacity, X3 is the battery rated voltage, X4 is the charging voltage measurement value, X5 is the charging current measurement value, X6 is the highest single cell voltage, X7 is the current state of charge SOC, X8 is the estimated remaining charging time, X9 is the cumulative charging time, X 10 The maximum battery temperature.

3. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 2 is characterized in that: The battery health index BHI is calculated by the following formula: BHI=ω1×SOC diff +ω2×V diff +ω3×T norm +ω4×C eff +ω5×R int ; In the formula, ω1, ω2, ω3, ω4, and ω5 are indicator weights; SOC diff is the SOC change rate; V diff is the voltage difference of single battery; T norm is the temperature normalized index; C eff is the capacity attenuation coefficient; R int is the rate of change of battery internal resistance; in: Where X7(t) is the current state of charge (SOC) at time t, and X7(t-1) is the current state of charge (SOC) at time t-1. T ref is the reference temperature under normal working conditions, T max is the maximum allowable temperature; C est The estimated current maximum available capacity is calculated based on the accumulated charge and discharge data; The remaining battery life RUL is calculated based on the battery health index BHI, using the expression: Where SOH = BHI·100%, and the average daily discharge is calculated from historical data.

4. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 3 is characterized in that: The method of optimizing and calculating the indicator weights by using the salp swarm algorithm includes: Initialize population parameters: set population size N, number of iterations T max , randomly initialize N groups of weights, expression: W i =(ω i1 ,oh i2 ,oh i3 ,oh i4 ,oh i5 ),i=1,2,3,4,5; Define the fitness function F(W) expression: Update the leader weight expression: IN new =In best +c1×r1×(W best -IN); Update the follower weight expression: The weights are normalized to satisfy the expression: Repeat the steps until T is reached max or convergence conditions; Where: i1 ,ω i2 ,ω i3 ,ω i4 ,ω i5 is the weight coefficient, which is used to balance the importance of each indicator, F(W) is the fitness function, BHI actual,t The real BHI value calculated for historical data, BHI pred,t is the BHI value calculated based on the current weight, W is the weight combination set in the current population, and W new is the latest weight combination in the current population, W best is the optimal weight combination in the current population, W i is the i-th weight combination, r1 is a random number between [0,1], c1 is the exploration step coefficient, which controls the convergence speed, c2 is the adjustment factor, which controls the follower step size, r2 is a random number between [0,1], ω k is the kth weight coefficient.

5. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 1 is characterized in that: In step 3, the steps for establishing a correlation model between battery health status and vehicle-grid interaction behavior are as follows: Remaining battery charge and discharge times T s , the expression is: T s =T init ×BHI-T used ; Where: T init is the initial rated cycle number of the battery, T used The number of charge and discharge times that have been used; The expression of charge and discharge depth is: Where: P c is the depth of charge, P d is the depth of discharge, P c,min The lowest charge depth of the battery (e.g. 10%), P d,min is the minimum discharge depth of the battery (e.g. 10%), α c ,α d is the adjustment coefficient of the charge and discharge depth, which is usually set based on experience or experimental data; the vehicle-grid interaction time interval T c expression: T c =T c,max -β×(1-BHI); Where: T c,max is the maximum vehicle-grid interaction time interval (minutes), β is the vehicle-grid interaction time reduction coefficient (empirical parameter); Charge and discharge power expression: Where: P max is the maximum charge and discharge power.

6. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 5 is characterized in that: The Autoformer model is used to process historical charging data, specifically including: Y t =F(X t )+ε t ; Where: Y t is the output variable of the future vehicle-grid interaction coefficient, ε t is the prediction error, F represents the prediction function determined by the Autoformer model, X t is the input variable of the historical vehicle-grid interaction coefficient; Y t ＝(α c,t+1 ,α d,t+1 ,β t+1 ) Where: α c,t+1 is the predicted charging depth adjustment coefficient, α d,t+1 is the predicted discharge depth adjustment coefficient, β t+1 is the predicted vehicle-grid interaction time reduction coefficient; X t =(BHI t ,P c,t ,P d,t ,T c,t ,T bat,t ,P EV,t ,α c,t ,α d,t ,β t ); Of which: BHI t is the battery health index, P c,t is the current charging depth, P d,t is the current discharge depth, T c,t is the current vehicle-grid interaction time interval, T bat,t is the current battery temperature, P EV,t is the current charging and discharging power of the electric vehicle, α c,t is the current charging depth adjustment coefficient, α d,t is the current discharge depth adjustment coefficient, β t is the current vehicle-grid interaction time reduction coefficient.

7. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 1 is characterized in that: The specific steps of predicting the charge and discharge depth adjustment coefficient and the interaction time reduction coefficient during the vehicle-grid interaction process include: extracting long-term and short-term features of input data through the autocorrelation mechanism and recursive decomposition module of the Autoformer model; The expression for generating the prediction output is: Y t =M Autoformer (X t ); Where: Y t For the future vehicle network interaction coefficient, X t is the historical vehicle-grid interaction coefficient; The analytical prediction result expression is: Where: α c,t+1 is the predicted charging depth adjustment coefficient, α d,t+1 is the predicted discharge depth adjustment coefficient, β t+1 is the predicted vehicle-grid interaction time reduction coefficient, M Autoformer represents the Autoformer prediction model, f Autoformer , g Autoformer , h Autoformer To be more specific, the Autoformer prediction model; The predicted adjustment coefficient α c,t+1 , α d,t+1 , β t+1 Substitute the following formula to dynamically adjust the charge and discharge depth and interaction time of the next cycle The interval expression is: Of which: BHI t is the battery health index, P c,t+1 is the predicted depth of charge, P d,t+1 is the predicted discharge depth, T c,t+1 is the predicted vehicle-grid interaction time interval, P c,min , P d,min is the minimum charge and discharge depth, T c,max is the maximum vehicle-grid interaction time interval.

8. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 1 is characterized in that: In step 5, the battery health index BHI generates a battery maintenance strategy, including the following steps: generating a battery maintenance strategy based on the battery health index BHI, and the strategy is classified according to the BHI value as follows: The corresponding maintenance strategy is: When BHI>0.8, the charging and discharging strategy remains unchanged, allowing high-frequency vehicle-grid interaction; When 0.5≤BHI≤0.8, limit the charge and discharge depth to P c,min and P d,min , and reduce the frequency of interaction; When BHI≤0.5, only shallow charging and discharging are allowed, and the vehicle-grid interaction is limited to the lowest frequency.

9. The battery health check and vehicle-grid interaction control method based on SSA and Autoformer according to claim 8 is characterized in that: Adjust vehicle-grid interaction control parameters based on BHI, including: Interaction time interval T c The expression is: T c =T c,max -β×(1-BHI); Where: BHI is the battery health index, T c is the interaction time interval, β is the interaction time adjustment coefficient, T c,max It is the pre-set maximum interaction time interval; The expression of charge and discharge depth is: P c =α c ×(1-BHI)+P c,min P d =α d ×(1-BHI)+P d,min Where: P c , P d is the depth of charge and discharge, α c , α d is the charge and discharge depth adjustment coefficient, P c,min , P d,min is the minimum charge and discharge depth; α c and α d According to BHI classification expression: Interaction time interval T c The specific adjustment is calculated by the following formula: Where: T max is the actual maximum interaction time interval.

Images