Solar power conversion system scheduling method based on AI decision

Abstract

The invention relates to the field of intelligent battery management and charging scheduling, in particular to a solar battery conversion system scheduling method based on AI decision, which comprises the following steps of: S1, acquiring operation parameter data and external operation data of each battery in a battery conversion system, and preprocessing the operation parameter data and the external operation data; and S2, inputting the preprocessed operation parameter data and external operation data as input data into a pre-trained AI charging scheduling model, and outputting prediction parameters representing the health change trend of each battery. According to the invention, the AI-driven charging scheduling system is combined with the particle swarm optimization algorithm, so that the system can carry out accurate charging power and sequence adjustment in real time according to the health state of the battery, the charging demand and the external environment condition, thereby not only optimizing the charging process, but also maximizing the charging efficiency, effectively reducing the waste of energy, and improving the charging efficiency. And it is ensured that energy of each unit is efficiently utilized.

Description

Technical Field

[0001] This invention relates to the field of intelligent battery management and charging scheduling, and in particular to a scheduling method for solar battery swapping systems based on AI decision-making. Background Technology

[0002] With the increasing prevalence of electric vehicles and renewable energy storage systems, battery charging management technology is becoming increasingly important. In particular, solar cell battery swapping systems can significantly improve charging efficiency through intelligent scheduling and optimization of the charging process. Existing charging management systems typically rely on fixed charging strategies and cannot dynamically respond to real-time battery health status and changes in the external environment, thus leaving room for optimization in terms of charging efficiency and battery life. The application of AI technology, especially particle swarm optimization-based charging scheduling, can dynamically optimize based on real-time data, making the charging process more efficient.

[0003] Existing charging scheduling systems often struggle to make precise adjustments in real time based on battery health, charging demand, and external environment, resulting in low charging efficiency and significant energy waste. Furthermore, fixed charging sequences and power allocation strategies fail to fully utilize the battery's charging potential and cannot prevent overcharging or overheating, leading to shortened battery life. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides a solar battery swapping system scheduling method based on AI decision-making, which aims to improve the problem that existing technologies are unable to dynamically optimize in real time based on battery health status, charging demand and external environment.

[0005] In a first aspect, the present invention provides the following technical solution: a solar battery swapping system scheduling method based on AI decision-making, comprising: S1. Collect the operating parameter data of each battery in the battery swapping system and external operating data, and preprocess the operating parameter data and external operating data; S2. The preprocessed operating parameter data and external operating data are used as input data and fed into the pre-trained AI charging scheduling model to output predicted parameters that characterize the health change trend of each battery. S3. Evaluate the health status of each battery based on the predicted parameters and generate charging priority parameters corresponding to each battery. S4. Based on the charging priority parameters, implement a differentiated charging scheduling strategy for the batteries in the charging state. S5. Based on the preset optimization objective function, the charging power and charging sequence of each battery are adjusted in real time during the charging process; S6. During the charging process, monitor the temperature and charging status of each battery in real time, and control the battery charging power to smoothly convert to the target charging power.

[0006] Preferably, in step S1, the data preprocessing step includes: Anomaly detection is performed on the collected battery operating parameter data, and abnormal data that exceeds the preset reasonable range is removed; Complete the incomplete operating parameter data; The operating parameter data and external operating data are normalized to meet the input requirements of the AI ​​charging scheduling model.

[0007] Preferably, in step S2, the step of outputting the predicted parameters characterizing the health change trend of each battery includes: Construct a model input vector that includes battery operating status information and external operating condition information; The model input vector is input into the pre-trained AI charging scheduling model; The AI ​​charging scheduling model outputs predictive parameters that characterize the health change trends of each battery.

[0008] Preferably, in step S3, the step of generating charging priority parameters corresponding to each battery includes: Calculate the health status index of each battery based on the predicted parameters; The batteries are sorted according to the aforementioned health status indicators; Based on the sorting results, assign corresponding charging priority parameters to each battery.

[0009] Preferably, in step S4, the step of implementing a differentiated charging scheduling strategy for the battery in the charging state includes: The charging start order of each battery is determined according to the charging priority parameters; Different initial charging power is assigned to batteries with different charging priorities; Control each battery to charge according to its corresponding charging sequence and charging power.

[0010] Preferably, in step S5, the step of presetting the objective function includes: Set an optimization objective function with the goal of maximizing charging efficiency; Determine the charging requirements, charging sequence, and charging power constraints for each battery; By combining the above objectives and constraints, a multi-objective optimization model is constructed.

[0011] Preferably, the step of solving the optimization objective function includes: Multiple particles are randomly generated, each representing a charging scheduling scheme, including charging sequence and power parameters; The fitness function is defined based on charging time and efficiency factors. The charging strategy of each particle is evaluated, and the historical optimal solution and global optimal solution of each particle are updated. According to the rules of particle swarm optimization, the speed and position of particles are adjusted by using historical optimal solutions and global optimal solutions to continuously optimize the charging scheduling strategy. When the stopping condition is met, the globally optimal solution is applied as the final charging scheduling scheme, and the charging process is adjusted in real time.

[0012] Preferably, in step S6, the step of controlling the smooth transition of the battery's charging power to the target charging power includes: Based on the battery's current charging status, health status, and temperature information, a charging power adjustment strategy is designed. Fuzzy control methods are used to ensure a smooth transition of charging power to the target charging power; Monitor changes in battery status during charging and dynamically adjust control parameters.

[0013] Secondly, this invention provides the following technical solution: a solar battery swapping system scheduling system based on AI decision-making, comprising: The data acquisition and preprocessing module acquires the operating parameter data of each battery in the battery swapping system as well as external operating data, and preprocesses the operating parameter data and external operating data. The health trend prediction module takes pre-processed operating parameter data and external operating data as input data, inputs them into the pre-trained AI charging scheduling model, and outputs prediction parameters that characterize the health change trend of each battery. The health assessment and priority generation module assesses the health status of each battery based on the predicted parameters and generates charging priority parameters corresponding to each battery. The differentiated charging scheduling module executes a differentiated charging scheduling strategy for batteries in the charging state according to the charging priority parameters. The dynamic optimization control module adjusts the charging power and charging sequence of each battery in real time during the charging process based on a preset optimization objective function. The charging power control module monitors the temperature and charging status of each battery in real time during the charging process, and controls the battery's charging power to smoothly transition to the target charging power.

[0014] The present invention has the following beneficial effects: 1. In this invention, through an AI-driven charging scheduling system combined with a particle swarm optimization algorithm, the system can make precise adjustments to the charging power and sequence in real time based on the battery's health status, charging needs, and external environmental conditions. This not only optimizes the charging process but also maximizes charging efficiency, effectively reduces energy waste, and ensures that the energy of each unit is used efficiently.

[0015] 2. In this invention, the system uses fuzzy control to smoothly adjust the charging power based on the real-time health status, charging status and temperature of the battery, so as to avoid overheating or overcharging of the battery during the charging process. This precise control can reduce battery degradation, slow down the aging rate, and thus extend the overall service life and performance retention time of the battery.

[0016] 3. In this invention, the system assigns a charging priority to each battery and adjusts the charging power according to the battery's health status and charging needs, ensuring that charging resources are optimally configured. High-priority batteries are charged first, while low-priority batteries are charged according to the remaining resources, thereby avoiding waste of resources and improving overall charging efficiency. Attached Figure Description

[0017] Figure 1 This is a flowchart of the AI-based decision-making scheduling method for solar battery swapping systems proposed in this invention. Figure 2 This is a system architecture diagram of the solar battery swapping system scheduling system based on AI decision-making proposed in this invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1 In a first embodiment of the present invention, the present invention provides a solar battery swapping system scheduling method based on AI decision-making, such as... Figure 1 As shown, it includes the following steps: S1. Collect the operating parameter data of each battery in the battery swapping system and external operating data, and preprocess the operating parameter data and external operating data; Furthermore, in S1, the data preprocessing steps include: detecting outliers in the collected battery operating parameter data and removing outliers that exceed the preset reasonable range; completing incomplete operating parameter data; and normalizing the operating parameter data and external operating data to meet the input requirements of the AI ​​charging scheduling model.

[0020] Specifically, firstly, outlier detection is performed on the collected battery operating parameter data. The purpose of outlier detection is to remove data that exceeds a preset reasonable range. These reasonable ranges are usually based on typical battery operating conditions and empirical values. For example, the battery temperature should be between -20℃ and 60℃, and the state of charge (SOC) should be between 0% and 100%. If the data exceeds these ranges, it is considered outlier and needs to be removed from the dataset. This can be achieved using a threshold-based outlier detection algorithm, as shown in the following formula: ; in, For the first Data points, and These represent the lower and upper limits of the data, respectively. After outliers are removed, the remaining data can be used for subsequent analysis.

[0021] Secondly, for missing data, completion techniques are used. Completion methods can include interpolation or mean-based imputation. The most common completion method is the nearest neighbor method, which uses the average of the two adjacent valid data points to fill in the missing value. Specifically, if the battery temperature data is missing at a certain moment, the data from the previous moment is used. and the data value at the next time step Calculate their average to fill in the missing values: ; This method ensures data consistency and avoids inaccurate analysis caused by missing data.

[0022] Then, all battery operating parameter data and external operating data are normalized. Normalization transforms data with different dimensions to the same scale, avoiding bias during model training. A common normalization method is the min-max normalization method, which transforms the data... The value within the interval, for each data point The normalization formula is as follows: ; in, The original data values, and These are the minimum and maximum values ​​of the feature data, respectively. This is the normalized data.

[0023] Finally, external operational data such as climate data and solar power generation should be normalized in the same way. Different characteristics may have different effects, so it is necessary to unify the data format and range to ensure that the data input into the AI ​​model is appropriate.

[0024] Through the above processing, the data becomes more consistent and standardized, meeting the input requirements of AI models. Specifically, the processed data can be used to train AI charging scheduling models, improving the model's prediction accuracy and decision-making ability. By removing outliers, filling in missing data, and normalizing the data, the quality of the input data can be ensured, the model training effect can be improved, and scheduling errors caused by data problems can be avoided.

[0025] S2. The preprocessed operating parameter data and external operating data are used as input data and fed into the pre-trained AI charging scheduling model to output predicted parameters that characterize the health change trend of each battery. Furthermore, in S2, the step of outputting the prediction parameters characterizing the health change trend of each battery includes: constructing a model input vector containing battery operating state information and external operating condition information; inputting the model input vector into the pre-trained AI charging scheduling model; and outputting the prediction parameters characterizing the health change trend of each battery through the AI ​​charging scheduling model.

[0026] Specifically, firstly, a model input vector is constructed that includes battery operating status information and external operating condition information. The model input vector should include various parameters related to battery health, such as the battery's State of Health (SOH), State of Charge (SOC), temperature, cycle count, and load status. Simultaneously, external operating conditions, such as ambient temperature, solar power generation, and time, should also be considered. These external factors may affect the battery's charging efficiency and health status. Assuming these parameters are... Then the model input vector It can be represented as: ; in, Indicating battery operating status or external environmental factors One parameter, and It is a vector containing all input features.

[0027] Next, the model input vector The input is fed into a pre-trained AI charging scheduling model. Assuming this AI model is a deep learning-based regression model, its goal is to predict battery health trends, such as changes in SOH and SOC over a future period. The deep learning model can be trained using multiple hidden layers and activation functions to learn the complex relationship between the input and prediction parameters. Let the model be... Its output is a prediction parameter for the battery health trend. ,Right now: ; in, It is the output of the AI ​​model, representing the predicted parameters of battery health change trends. It may include multiple sub-parameters, such as the battery's SOH, SOC, and temperature. These parameters can characterize the battery's health and charging status. Specific prediction parameters can be modeled using deep learning methods such as multilayer perceptron (MLP) or long short-term memory (LSTM) networks.

[0028] The output of this AI model can reveal the health trend of each battery over a future period. These predicted parameters can be used as a basis for subsequent scheduling, providing decision support for different batteries' charging priorities and charging power allocation.

[0029] By inputting preprocessed battery operating data and external environment data into the AI ​​model, the health status and charging trend of the battery can be predicted in real time. The prediction results can provide accurate information for the scheduling system, guide the setting of battery charging priorities and the adjustment of charging strategies, improve charging efficiency and extend battery life.

[0030] S3. Evaluate the health status of each battery based on the predicted parameters and generate charging priority parameters corresponding to each battery. Furthermore, in S3, the step of generating charging priority parameters corresponding to each battery includes: calculating the health status index of each battery based on the predicted parameters; sorting each battery according to the health status index; and assigning corresponding charging priority parameters to each battery according to the sorting results.

[0031] Specifically, firstly, the health status index of each battery is calculated based on the predicted parameters. This includes battery health-related information such as SOH, SOC, and temperature. This information can be used to calculate battery health indicators. For example, SOH can be used to assess the remaining battery life, SOC to assess the current state of charge, and temperature to assess the impact of the operating environment on the battery's health. The health indicators are set as follows: This indicator can be calculated by combining multiple parameters such as the battery's SOH, SOC, and temperature, as shown in the following formula: ; in, , , They represent the first The battery's state of health (SOH), state of charge (SOC), and temperature (T) are as follows: These are weighting coefficients, which represent the degree of influence of each parameter on the battery's health status. Weighting coefficients can be set using historical data or empirical rules.

[0032] Next, the batteries will be sorted according to their health status indicators. The higher the value, the better the battery's health; conversely, the lower the value, the worse the health. Therefore, the health status index can be used to determine the battery's overall health. Sort all batteries, assuming the batteries are... and The health status indicators are respectively and The sorting rules are as follows: ; All batteries were categorized by health status indicators Sort the batteries from highest to lowest to obtain the sorted sequence. ,in This represents the total number of batteries.

[0033] Finally, charging priority parameters are assigned to each battery based on the sorting results. Directly related to health status indicators, batteries in better health are charged first. Based on the ranking results, charging priority parameters can be assigned to each battery. Batteries with higher charging priority correspond to smaller values; for example, the healthiest battery is assigned a higher charging priority. The number of batteries is increased sequentially until the least healthy battery is assigned the maximum value. The formula is as follows: ; in, Indicates the first The charging priority parameters for each battery. For batteries In this sorting sequence, batteries in poor health are given lower priority, while batteries in better health are charged first.

[0034] By assessing and prioritizing battery health, a reasonable allocation of charging priorities can be ensured. Batteries in better health are charged first, while batteries in poorer health are charged after other batteries are charged. This optimizes the charging process, improves the overall efficiency of the system, and extends battery life.

[0035] S4. Based on the charging priority parameters, implement differentiated charging scheduling strategies for batteries in the charging state. Furthermore, in S4, the steps of implementing a differentiated charging scheduling strategy for batteries in the charging state include: determining the charging start order of each battery according to the charging priority parameter; allocating different initial charging power to batteries with different charging priorities; and controlling each battery to charge according to the corresponding charging order and charging power.

[0036] Specifically, firstly, the charging start order of each battery is determined according to the charging priority parameter. This is obtained through battery health assessment; higher-priority batteries should be charged first, assuming... For the first The charging priority parameters for each battery are used to sort all batteries according to their charging priority, and the charging start order is determined accordingly. It can be determined using the following formula: ; in, Indicates battery Its position in the charging sequence Indicates battery In the charging priority ranking, the higher the priority, the smaller the battery. The value will be prioritized and charged first.

[0037] Next, different initial charging powers are assigned to batteries with different charging priorities. The charging power should be allocated according to the battery's charging priority. Batteries with higher priority should be allocated higher initial charging power. The charging power for each battery can be calculated using the following formula: ; in, Indicates battery Initial charging power, This is the maximum value of the charging priority, representing the priority parameter of the battery that receives the least priority for charging. Total number of batteries For the first The charging priority parameter for each battery is determined by this formula, which ensures that batteries with higher priority are charged... Smaller batteries can distribute more charging power.

[0038] Finally, control each battery to charge according to its corresponding charging sequence and charging power. During the charging process, the charging sequence of the batteries is adjusted accordingly. and initial charging power Adjust the charging process, assuming the battery The maximum charging power limit during charging is: Actual charging power Need to and The charging power is adjusted accordingly, and the formula for controlling the charging power is as follows: ; in, Indicates battery Actual charging power, This sets an upper limit for charging power, ensuring that the battery's maximum charging power capacity is not exceeded. The battery will then charge according to the charging sequence. and actual charging power Charge sequentially.

[0039] By employing a differentiated charging scheduling strategy, a reasonable charging sequence and charging power are allocated to each battery based on its charging priority. Higher-priority batteries receive more charging resources to ensure they are fully charged in the shortest possible time, while lower-priority batteries are charged using idle resources. This maximizes the overall charging efficiency of the system, avoids resource waste during charging, and reduces the impact on battery health.

[0040] S5. Based on the preset optimization objective function, the charging power and charging sequence of each battery are adjusted in real time during the charging process; Furthermore, in S5, the steps for setting the optimization objective function include: setting the optimization objective function with the goal of maximizing charging efficiency; determining the charging requirements, charging sequence, and charging power constraints for each battery; and combining the above objectives and constraints to construct a multi-objective optimization model.

[0041] Furthermore, the steps for solving the objective function include: randomly generating multiple particles, each representing a charging schedule scheme, including charging order and power parameters; defining a fitness function based on charging time and efficiency factors, evaluating the charging strategy of each particle, and updating the historical optimal solution and global optimal solution for each particle; adjusting the speed and position of particles according to the rules of particle swarm optimization, using the historical optimal solution and global optimal solution to continuously optimize the charging schedule strategy; and when the stopping condition is met, applying the global optimal solution as the final charging schedule scheme and adjusting the charging process in real time.

[0042] Specifically, first, an optimization objective function is set, the goal of which is to maximize charging efficiency. Charging efficiency can usually be represented by the ratio of battery charging power to charging time. The objective function for charging efficiency is then defined. The calculation formula is as follows: ; in, Indicates the first The charging power of each battery. Indicates the first The charging time of each battery. It is the maximum charging power of each battery. The objective function aims to improve the overall charging efficiency of the system by maximizing the utilization of charging power per unit time.

[0043] Then, determine the charging requirements, charging sequence, and charging power constraints for each battery. The charging requirements of a battery can be determined based on its current state of charge (SOC), and the upper limit of the charging power is determined by the battery's maximum power handling capacity. The charging sequence is determined by the charging priority parameter. The constraints on charging power and sequence are as follows: ; By combining the above objectives and constraints, a multi-objective optimization model is constructed. This model can be built by considering both charging efficiency and battery safety. The ultimate goal is to balance efficiency and battery health to obtain an optimal charging scheduling scheme. The objective function of this model can be expressed as: ; in, and These are the weighting coefficients for charging efficiency and battery safety targets, respectively. This is a battery safety assessment function, typically calculated based on temperature, SOC, and battery aging level.

[0044] Next, the particle swarm optimization algorithm is used to solve the objective function. The particle swarm optimization algorithm searches for the optimal solution by simulating the movement of a swarm of particles. First, multiple particles are randomly generated, each representing a charging schedule scheme, including charging order and charging power parameters. The nth particle is then set... One particle is ,in Indicates battery The charging sequence, Indicates battery The charging power.

[0045] Then, a fitness function is defined based on charging time and efficiency factors. The charging strategy for each particle is evaluated, and the historical optimal solution and global optimal solution for each particle are updated. The fitness function can comprehensively consider charging efficiency and battery health, and the specific formula is as follows: ; in, It is the first The total objective function value of each particle It is the first The charging time of each battery.

[0046] During particle swarm optimization, the velocity and position of particles are adjusted based on historical and global optimal solutions according to the particle swarm optimization rules, continuously optimizing the charging scheduling strategy. The particle update rules are as follows: in, It is the first The speed of each particle It is the first The position of each particle determines the charging schedule. and These are the historical optimal solution and the global optimal solution, respectively. and It is the acceleration constant. and It is a random number.

[0047] When the stopping condition is met, such as reaching the maximum number of iterations or the fitness function value converging, the global optimal solution is applied as the final charging schedule, and the charging process is adjusted in real time.

[0048] By using particle swarm optimization algorithm, the charging power and charging sequence of the battery are adjusted in real time to optimize charging efficiency while ensuring battery safety. By continuously optimizing the charging strategy, the efficiency of the charging system can be maximized, the battery life can be extended, and the system can dynamically respond to changes in the environment and battery status during the charging process.

[0049] S6. During the charging process, monitor the temperature and charging status of each battery in real time, and control the battery charging power to smoothly convert to the target charging power. Furthermore, in S6, the steps for controlling the smooth transition of the battery's charging power to the target charging power include: designing a charging power adjustment strategy based on the battery's current charging state, health state, and temperature information; using a fuzzy control method to ensure a smooth transition of the charging power to the target charging power; and monitoring changes in the battery's state during charging and dynamically adjusting the control parameters.

[0050] Specifically, the battery's temperature and charging status are monitored in real time, and the charging power is controlled to smoothly transition to the target charging power, ensuring the safety and efficiency of the charging process. The key to this step is designing a charging power adjustment strategy based on the battery's charging status, health status, and temperature information, and using fuzzy control methods for a smooth transition.

[0051] First, based on the battery's current state of charge, state of health, and temperature information, a charging power adjustment strategy is designed. The battery's charging power should take into account current health parameters such as State of Health (SOH), State of Charge (SOC), and temperature to avoid overcharging or excessive battery temperature. The battery's charging power adjustment strategy is set as follows: The calculation formula is as follows: ; in, To achieve the target charging power, It is an adjustment function that adjusts the charging power based on the battery's state of health (SOH), state of charge (SOC), and temperature (T). The design of this function needs to ensure that when the battery's state of health is poor or the temperature is high, the charging power is reduced; when the battery's state of health is good and the temperature is moderate, the charging power can approach the target value.

[0052] Then, a fuzzy control method is used to ensure a smooth transition of the charging power to the target charging power. Fuzzy control is a control method based on linguistic variables and rules, capable of handling uncertainties and nonlinear relationships. During the charging process, fuzzy rules are set to adjust the rate of change of the charging power based on the battery's temperature, charging state, and health status. Assuming the difference between the current charging power and the target charging power is... The adjustment speed of charging power can be determined through fuzzy control rules: ; Fuzzy control rules are based on the difference in charging power. The adjustment amount is determined by the battery's state of health (SOH), state of charge (SOC), and temperature (T). The fuzzy control algorithm can handle different charging adjustment conditions by defining multiple fuzzy sets, such as small, medium, and large. For example, when... When the battery is large and the battery temperature is high, the adjustment rate will be reduced to avoid overheating caused by fast charging.

[0053] Next, the battery's state changes during charging are monitored, and control parameters are dynamically adjusted. During charging, the battery's temperature and SOC change, so these parameters need to be monitored in real time, and the charging strategy adjusted based on the real-time status. Assuming at time... The battery temperature and SOC are respectively and We can update the charging power using the following dynamic adjustment formula: ; in, It is the charging power at the next moment. It is the amount of charging power change adjusted according to fuzzy control. In this way, the charging power can be smoothly adjusted as the battery status changes, avoiding risks caused by excessive temperature or poor battery status.

[0054] By monitoring the battery's temperature, SOC, and health status in real time, and using fuzzy control methods to smoothly adjust the charging power, this control strategy ensures smooth power changes during the charging process, avoiding overcharging, overheating, or other battery damage issues. This approach improves charging efficiency while ensuring battery safety and extending its lifespan.

[0055] Example 2: Charging scheduling systems often fail to make precise adjustments in real time based on battery health, charging demand, and external environment, resulting in low charging efficiency and significant energy waste. To address these issues, this invention provides an AI-based decision-making scheduling system for solar-powered battery swapping systems, the structure of which is as follows: Figure 2 As shown, it includes: The data acquisition and preprocessing module acquires the operating parameter data of each battery in the battery swapping system as well as external operating data, and preprocesses the operating parameter data and external operating data. The health trend prediction module takes pre-processed operating parameter data and external operating data as input data, inputs them into the pre-trained AI charging scheduling model, and outputs prediction parameters that characterize the health change trend of each battery. The health assessment and priority generation module assesses the health status of each battery based on the predicted parameters and generates charging priority parameters corresponding to each battery. The differentiated charging scheduling module executes a differentiated charging scheduling strategy for batteries in the charging state according to the charging priority parameters. The dynamic optimization control module adjusts the charging power and charging sequence of each battery in real time during the charging process based on a preset optimization objective function. The charging power control module monitors the temperature and charging status of each battery in real time during the charging process, and controls the battery's charging power to smoothly transition to the target charging power.

[0056] Specifically, the data acquisition and preprocessing module is responsible for collecting real-time operating parameter data of each battery in the battery swapping system, such as battery health status (SOH), charging status (SOC), temperature, etc., as well as external environmental data, such as solar power generation and external temperature. It also ensures the accuracy and consistency of the data through preprocessing operations such as outlier detection, missing data completion, and normalization.

[0057] The health trend prediction module inputs preprocessed data into a pre-trained AI charging scheduling model, uses machine learning algorithms to predict the health change trends of each battery, and outputs prediction parameters to provide a basis for subsequent health assessments.

[0058] The health assessment and priority generation module evaluates the health status of each battery based on the predicted parameters and generates charging priority parameters, prioritizing the charging of batteries with high health status.

[0059] The differentiated charging scheduling module formulates differentiated charging strategies based on charging priority parameters to ensure that high-priority batteries are charged first and allocated more charging resources.

[0060] The dynamic optimization control module, based on a preset optimization objective function, adjusts the charging power and charging sequence of each battery in real time to ensure that the charging process achieves the optimal balance between maximizing efficiency and battery safety.

[0061] The charging power control module monitors the temperature and charging status of each battery in real time during the charging process, and uses a fuzzy control algorithm to smoothly adjust the charging power to avoid overheating or overcharging, ensuring a safe and efficient charging process.

[0062] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A solar battery swapping system scheduling method based on AI decision-making, characterized in that, include: S1. Collect the operating parameter data of each battery in the battery swapping system and external operating data, and preprocess the operating parameter data and external operating data; S2. The preprocessed operating parameter data and external operating data are used as input data and fed into the pre-trained AI charging scheduling model to output predicted parameters that characterize the health change trend of each battery. S3. Evaluate the health status of each battery based on the predicted parameters and generate charging priority parameters corresponding to each battery. S4. Based on the charging priority parameters, implement a differentiated charging scheduling strategy for the batteries in the charging state. S5. Based on the preset optimization objective function, the charging power and charging sequence of each battery are adjusted in real time during the charging process; S6. During the charging process, monitor the temperature and charging status of each battery in real time, and control the battery charging power to smoothly convert to the target charging power.

2. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S1, the data preprocessing steps include: Anomaly detection is performed on the collected battery operating parameter data, and abnormal data that exceeds the preset reasonable range is removed; Complete the incomplete operating parameter data; The operating parameter data and external operating data are normalized to meet the input requirements of the AI ​​charging scheduling model.

3. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S2, the step of outputting the predicted parameters characterizing the health change trend of each battery includes: Construct a model input vector that includes battery operating status information and external operating condition information; The model input vector is input into the pre-trained AI charging scheduling model; The AI ​​charging scheduling model outputs predictive parameters that characterize the health change trends of each battery.

4. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S3, the step of generating charging priority parameters corresponding to each battery includes: Calculate the health status index of each battery based on the predicted parameters; The batteries are sorted according to the aforementioned health status indicators; Based on the sorting results, assign corresponding charging priority parameters to each battery.

5. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S4, the step of implementing a differentiated charging scheduling strategy for the battery in the charging state includes: The charging start order of each battery is determined according to the charging priority parameters; Different initial charging power is assigned to batteries with different charging priorities; Control each battery to charge according to its corresponding charging sequence and charging power.

6. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S5, the step of presetting the objective function for optimization includes: Set an optimization objective function with the goal of maximizing charging efficiency; Determine the charging requirements, charging sequence, and charging power constraints for each battery; By combining the above objectives and constraints, a multi-objective optimization model is constructed.

7. The solar battery swapping system scheduling method based on AI decision-making according to claim 6, characterized in that, The steps for solving the optimization objective function include: Multiple particles are randomly generated, each representing a charging scheduling scheme, including charging sequence and power parameters; The fitness function is defined based on charging time and efficiency factors. The charging strategy of each particle is evaluated, and the historical optimal solution and global optimal solution of each particle are updated. According to the rules of particle swarm optimization, the speed and position of particles are adjusted by using historical optimal solutions and global optimal solutions to continuously optimize the charging scheduling strategy. When the stopping condition is met, the globally optimal solution is applied as the final charging scheduling scheme, and the charging process is adjusted in real time.

8. The solar battery swapping system scheduling method based on AI decision-making according to claim 1, characterized in that, In step S6, the step of smoothly converting the battery's charging power to the target charging power includes: Based on the battery's current charging status, health status, and temperature information, a charging power adjustment strategy is designed. Fuzzy control methods are used to ensure a smooth transition of charging power to the target charging power; Monitor the battery's state changes during charging and dynamically adjust control parameters.

9. A solar battery swapping system scheduling system based on AI decision-making, characterized in that, The solar battery swapping system scheduling method based on AI decision-making as described in any one of claims 1-8 includes: The data acquisition and preprocessing module acquires the operating parameter data of each battery in the battery swapping system as well as external operating data, and preprocesses the operating parameter data and external operating data. The health trend prediction module takes pre-processed operating parameter data and external operating data as input data, inputs them into the pre-trained AI charging scheduling model, and outputs prediction parameters that characterize the health change trend of each battery. The health assessment and priority generation module assesses the health status of each battery based on the predicted parameters and generates charging priority parameters corresponding to each battery. The differentiated charging scheduling module executes a differentiated charging scheduling strategy for batteries in the charging state according to the charging priority parameters. The dynamic optimization control module adjusts the charging power and charging sequence of each battery in real time during the charging process based on a preset optimization objective function. The charging power control module monitors the temperature and charging status of each battery in real time during the charging process, and controls the battery's charging power to smoothly transition to the target charging power.

Images