Battery charging protection method and system based on intelligent control, and storage medium

Abstract

The application relates to the technical field of battery protection control, and discloses a battery charging protection method and system based on intelligent control and a storage medium. The method comprises the following steps: obtaining a state value by collecting and calculating battery parameters; constructing a neural network model based on the state value to output charging parameters; generating a charging strategy according to the charging parameters; inputting the charging strategy into a control circuit for execution; obtaining a result by executing multi-level security protection; and updating the neural network model to optimize control parameters according to the result. The application constructs a charging control model based on a back propagation neural network, combines real-time battery state monitoring, self-adaptive charging strategy generation, a multi-level security protection mechanism and a self-learning optimization function, realizes accurate control and all-round protection of the battery charging process, effectively prolongs the service life of the battery and improves charging safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery protection control, in particular to a battery charging protection method and system based on intelligent control and a storage medium. BACKGROUND

[0002] With the rapid development of electric vehicles, renewable energy storage systems and portable electronic devices, the performance and life of batteries as key energy storage elements have become important factors restricting the further development of these technologies. Traditional battery charging methods mainly use constant current-constant voltage (CC-CV) charging mode, which is simple to operate, but lacks adaptive adjustment capability for different battery states and environmental conditions. Some improved charging methods such as pulse charging and multi-stage charging have improved charging efficiency and battery life to some extent, but still use fixed charging parameters and preset thresholds, which cannot be dynamically optimized according to the real-time state and aging degree of the battery. In recent years, with the development of artificial intelligence technology, battery management systems based on fuzzy logic, neural networks and other intelligent algorithms have been gradually applied to battery charging control, but most systems only focus on optimizing charging efficiency or charging time, and lack consideration of battery safety protection and life extension.

[0003] The main problems existing in the prior art battery charging technology include: first, the lack of comprehensive monitoring and accurate evaluation of battery parameters leads to the inability to accurately grasp the battery state during charging, increasing the risk of overcharging and overdischarging; second, the charging strategy generally uses static parameter setting, which cannot be adaptively adjusted according to the dynamic change characteristics of the battery and environmental conditions, affecting charging efficiency and safety; third, most charging systems lack effective safety protection mechanisms, when abnormal conditions occur, either completely rely on hardware protection to cut off charging, or only have simple parameter adjustment, and cannot achieve multi-level protection; fourth, as the battery ages, its internal characteristics change, but the existing charging system rarely considers the impact of battery aging on charging parameters, making it difficult to adapt to the management needs of the entire life cycle of the battery; finally, the existing system usually lacks self-learning ability, and cannot summarize experience from historical charging data and optimize the charging strategy, making it difficult to continuously improve the charging control. SUMMARY

[0004] The present application provides a battery charging protection method and system based on intelligent control and a storage medium, which is used to build a charging control model based on a backpropagation neural network, combine real-time battery state monitoring, adaptive charging strategy generation, multi-level safety protection mechanism and self-learning optimization function, realize accurate control and comprehensive protection of the battery charging process, effectively prolong the service life of the battery and improve the charging safety.

[0005] In a first aspect, the application provides a battery charging protection method based on intelligent control, which comprises: collecting parameters of voltage, current, temperature and internal resistance of a battery to obtain battery parameter data, and performing calculation and processing on the battery parameter data to obtain a charging state value, a health state value and a power state value; constructing a back propagation neural network model based on the battery parameter data, the charging state value, the health state value and the power state value to obtain an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage and a charging risk level; generating a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level to obtain a charging strategy; inputting the charging strategy into a power electronic control circuit for charging control to obtain a charging control result; executing a multi-level safety protection mechanism based on the battery parameter data, the charging strategy and the charging control result to obtain a safety protection result; and performing reinforcement learning update on the back propagation neural network model according to the safety protection result and charging cycle data to obtain optimized battery charging protection control parameters.

[0006] In a second aspect, the application provides a battery charging protection system based on intelligent control, which comprises:

[0007] A collection module for collecting parameters of voltage, current, temperature and internal resistance of a battery to obtain battery parameter data, and performing calculation and processing on the battery parameter data to obtain a charging state value, a health state value and a power state value;

[0008] A construction module for constructing a back propagation neural network model based on the battery parameter data, the charging state value, the health state value and the power state value to obtain an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage and a charging risk level;

[0009] A generation module for generating a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level to obtain a charging strategy;

[0010] A control module for inputting the charging strategy into a power electronic control circuit for charging control to obtain a charging control result;

[0011] An execution module for executing a multi-level safety protection mechanism based on the battery parameter data, the charging strategy and the charging control result to obtain a safety protection result;

[0012] An updating module is configured to perform reinforcement learning updating on the back propagation neural network model according to the safety protection result and charging cycle data, so as to obtain optimized battery charging protection control parameters.

[0013] In a third aspect, a battery charging protection device based on intelligent control is provided, which comprises a memory and at least one processor, the memory storing instructions; the at least one processor invokes the instructions in the memory, so that the battery charging protection device based on intelligent control executes the battery charging protection method based on intelligent control as described above.

[0014] In a fourth aspect, a computer readable storage medium is provided, which stores instructions, and when the instructions are run on a computer, the computer executes the battery charging protection method based on intelligent control as described above.

[0015] The technical scheme provided in the application has the following remarkable and beneficial effects. Firstly, the application of the back propagation neural network model changes the battery charging control from the traditional fixed parameter mode to the adaptive intelligent control mode. The nonlinear mapping capability of the neural network can accurately capture the complex relationship between the battery parameters, and output optimized charging control parameters, including the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage, and the charging risk level, thereby significantly improving the accuracy and adaptability of the charging control. Secondly, the adaptive generation technology of the segmented charging curve dynamically adjusts the parameters of the pre-charging, constant current charging, constant voltage charging, and trickle charging stages according to the real-time state of the battery, taking into account the charging efficiency and battery safety, and effectively reducing the damage to the battery during the charging process. Thirdly, the precise cooperation of the power electronic control circuit and the charging strategy realizes the precise execution of the charging process through high-precision closed-loop control, thereby ensuring the effective implementation of the charging strategy. Fourthly, the implementation of the multi-level safety protection mechanism builds a comprehensive safety protection system at the software and hardware levels, which can quickly respond to various abnormal situations, from slight parameter adjustment to emergency charging interruption, and provides gradient protection measures, thereby significantly improving the safety of the charging process. Finally, the reinforcement learning update mechanism enables the system to have self-learning and optimization capabilities. By continuously accumulating charging experience and optimizing the neural network model, the system performance continuously improves over time, and can also adapt to the aging characteristics of the battery, thereby providing optimized charging management for the entire life cycle of the battery. The characteristics of the back propagation neural network algorithm in the application make a great contribution to the scheme. The multi-layer neuron structure can learn the complex nonlinear relationship in the battery charging process. The ReLU activation function improves the expression ability of the model. The L2 regularization method prevents overfitting, thereby ensuring the generalization ability of the model. The parameter optimization based on gradient descent enables the model to continuously adjust to adapt to changes in battery characteristics. These algorithm characteristics together form an adaptive, accurate, and safe intelligent battery charging control system, which solves the problems of fixed parameters, insufficient safety protection, and poor adaptability in traditional charging methods. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0017] Figure 1 An embodiment of the battery charging protection method based on intelligent control in the embodiments of the application is shown in the figure.

[0018] Figure 2 Figure 1 is a schematic diagram of an embodiment of a battery charging protection system based on intelligent control in the present application;

[0019] Figure 3 Figure 1 is a schematic diagram of an embodiment of a battery charging protection system based on intelligent control in the present application; DETAILED DESCRIPTION

[0020] The present application provides a battery charging protection method and system based on intelligent control, and a storage medium. In the specification and claims of the present application and the above-described drawings, the terms "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the term "comprise" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present application is described below. Please refer to Figure 1 An embodiment of the battery charging protection method based on intelligent control in the present application includes the following steps.

[0022] Step S101, collecting parameters of voltage, current, temperature and internal resistance of the battery to obtain battery parameter data, and performing calculation and processing on the battery parameter data to obtain charging state value, health state value and power state value;

[0023] Step S102, constructing a back propagation neural network model based on the battery parameter data, the charging state value, the health state value and the power state value to obtain an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage and a charging risk level;

[0024] Step S103, generating a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level to obtain a charging strategy;

[0025] Step S104, inputting the charging strategy into a power electronic control circuit for charging control to obtain a charging control result;

[0026] Step S105, executing a multi-level safety protection mechanism based on the battery parameter data, the charging strategy and the charging control result to obtain a safety protection result;

[0027] Step S106, according to the security protection result and the charging cycle data, the back propagation neural network model is updated by reinforcement learning, and the optimized battery charging protection control parameter is obtained.

[0028] It can be understood that the execution subject of the present application can be a battery charging protection system based on intelligent control, and can also be a terminal or a server, which is not limited here. The server is taken as an example for description in the embodiments of the present application.

[0029] Specifically, the voltage, current, temperature and internal resistance of the battery are collected by the high-precision data acquisition circuit. In this process, a 16-bit ADC converter is used to collect data at a sampling frequency of 100 Hz, ensuring the accuracy and real-time performance of the data. The temperature sensor uses a PT100 type thermistor, and the current acquisition uses a Hall current sensor. The collected raw data is processed by a Butterworth low-pass filter with a cutoff frequency of 10 Hz to filter out high-frequency noise interference and form battery parameter data. Then, the improved ampere-hour metering method combined with Kalman filtering algorithm is used to calculate the state of charge value. This method integrates the charging and discharging current and combines the open circuit voltage correction, and the accuracy can reach ±2%. The health state value is calculated by a multi-parameter fusion method, which comprehensively considers the battery capacity attenuation rate, internal resistance increase rate and charging and discharging efficiency change rate to generate a health score of 0-100%. The power state value is calculated based on the current temperature of the battery, the state of charge value and the internal resistance to calculate the maximum allowed charging power. Based on the obtained battery parameter data, the state of charge value, the health state value and the power state value, a back propagation neural network model is constructed. The model adopts a four-layer structure, including an input layer, two hidden layers and an output layer. The input layer contains 10 neurons corresponding to the key parameters in the previous step. The first hidden layer contains 16 neurons, and the second hidden layer contains 8 neurons, both of which use ReLU activation function to enhance the non-linear expression ability of the network. The output layer contains 4 neurons, which respectively output the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage and the charging risk level. The neural network is trained by the back propagation algorithm, the loss function selects the mean square error, the optimizer uses the Adam algorithm, and the training data set is composed of historical charging records. L2 regularization method is used to prevent overfitting during training, and the regularization coefficient is set to 0.0001. The early stopping mechanism is introduced.

[0030] A segmented charging curve is generated according to the output optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage and the charging risk level. The charging process is divided into a pre-charging phase, a constant current charging phase, a constant voltage charging phase and a trickle charging phase. The pre-charging phase charging current is set to 0.1 times the maximum allowed charging current threshold, the constant current charging phase uses the optimal charging current value but does not exceed the maximum allowed threshold, the constant voltage charging phase voltage is constant at the charging termination voltage value, and the trickle charging phase adopts pulse charging mode with a frequency of 1 kHz. For different charging risk levels, the system automatically adjusts the parameters of each phase to form a complete charging strategy. The charging strategy is input to the power electronic control circuit for charging control. The control circuit takes a digital signal processor as the core, and combines a field programmable gate array to realize high-speed signal processing and PWM generation. The power conversion circuit adopts a full-bridge LLC resonant conversion topology, and the main switch tube uses a silicon carbide MOSFET. The control circuit realizes closed-loop control through a PI control algorithm, and automatically switches the control mode according to the charging phase: the pre-charging and constant current charging phases use current closed-loop control, the constant voltage charging phase uses voltage closed-loop control, and the trickle charging phase controls the power switch tube through the PWM signal.

[0031] A multi-level safety protection mechanism is executed based on the battery parameter data, the charging strategy and the charging control result. The multi-level safety protection includes two levels of software protection and hardware protection. The software protection layer performs state evaluation every 100 ms, including charging current abnormality detection, charging voltage abnormality detection, temperature abnormality detection and charging state value change rate abnormality detection. When the charging current exceeds the maximum allowed charging current threshold, an current abnormality alarm is triggered; when the charging voltage exceeds the charging termination voltage, a voltage abnormality alarm is triggered; when the temperature exceeds the set threshold, the system automatically adjusts the charging current or stops charging according to the temperature level. The hardware protection layer realizes overcurrent protection, overvoltage protection and overtemperature protection through the power electronic control circuit, with a response time less than 10 μs. According to the safety protection result and the charging cycle data, the back propagation neural network model is updated through reinforcement learning. The safety protection result and the charging cycle data are combined into a training sample set, corresponding to the battery response characteristics under different charging conditions. After the data is normalized to eliminate the dimension influence, it is input into the back propagation neural network model, the difference between the actual charging effect and the expected effect is calculated, and the model error is obtained. Based on the model error, the network weight parameters are adjusted through gradient descent, and the network parameters are updated. The battery capacity and internal resistance in the battery parameter data are analyzed in time sequence, the capacity attenuation trend and the internal resistance change trend are calculated, and the updated network parameters are adjusted for aging compensation, and applied to the prediction of control parameters in the next charging cycle.

[0032] For example, in a specific battery charging protection application scenario, when the battery voltage is 3.2V, the current is 2A, the surface temperature is 25℃, and the internal resistance is 50mΩ, the stable battery parameter data is obtained after filtering processing. The state of charge value is calculated to be 30% by ampere-hour metering method combined with Kalman filter algorithm, the state of health value is evaluated to be 85% by multi-parameter fusion method, and the power state value is calculated to be 100W based on the current temperature and internal resistance. After these data are input into the back propagation neural network model, the output optimal charging current is 2.5A, the maximum allowed charging current threshold is 3A, the charging termination voltage is 4.2V, and the charging risk level is low risk. Accordingly, the segmented charging curve is generated: the pre-charging phase current is 0.3A until the battery voltage reaches 3.8V, the constant current charging phase current is 2.5A until the battery voltage reaches 4.1V, the constant voltage charging phase voltage is constant at 4.2V until the current drops to 0.25A, and then the trickle charging phase uses 1kHz pulse charging. During the charging process, the multi-level safety protection mechanism continues to monitor, and when the battery temperature rises to 45℃, the charging current is automatically reduced to 2A until the temperature returns to normal. After the charging is completed, the data of the entire charging process is used for reinforcement learning update of the neural network model, and the optimized model can more accurately predict the parameters required for the next charging, ensuring the safety and maximizing the life of the battery charging.

[0033] In the embodiments of the present application, by collecting the parameters of the voltage, current, temperature and internal resistance of the battery, and combining the state of charge value, the state of health value and the state of power value to construct a back propagation neural network model, a number of significant beneficial effects are achieved. First, the application of the back propagation neural network model changes the battery charging control from the traditional fixed parameter mode to the adaptive intelligent control mode. The nonlinear mapping capability of the neural network can accurately capture the complex relationship between the battery parameters, and output optimized charging control parameters, including the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level, which significantly improves the accuracy and adaptability of the charging control. Second, the adaptive generation technology of the segmented charging curve dynamically adjusts the parameters of the pre-charging, constant current charging, constant voltage charging and trickle charging stages according to the real-time state of the battery, taking into account the charging efficiency and battery safety, and effectively reducing the damage to the battery during the charging process. Third, the precise cooperation of the power electronic control circuit and the charging strategy realizes the precise execution of the charging process through high-precision closed-loop control, ensuring the effective implementation of the charging strategy. Fourth, the implementation of the multi-level safety protection mechanism builds a comprehensive safety protection system at the software and hardware levels, which can quickly respond to various abnormal situations, from slight parameter adjustment to emergency charging interruption, providing gradient protection measures and significantly improving the safety of the charging process. Finally, the reinforcement learning update mechanism enables the system to have self-learning optimization capability, which continuously accumulates charging experience and optimizes the neural network model, and the system performance continuously improves with the use time, while being able to adapt to the aging characteristics of the battery, providing optimized charging management for the entire life cycle of the battery. The characteristics of the back propagation neural network algorithm in the present application make a great contribution to the scheme. The multi-layer neuron structure can learn the complex nonlinear relationship in the battery charging process, the ReLU activation function improves the expression ability of the model, the L2 regularization method prevents overfitting, and ensures the generalization ability of the model, and the parameter optimization based on gradient descent enables the model to continuously adjust to adapt to the changes in battery characteristics. These algorithm characteristics together form an adaptive, accurate and safe intelligent battery charging control system, solving the problems of fixed parameters, insufficient safety protection and poor adaptability in traditional charging methods.

[0034] In a specific embodiment, the process of step S101 can specifically include the following steps:

[0035] The voltage, current, temperature and internal resistance of the battery are acquired by a high-precision data acquisition circuit to obtain original analog signals;

[0036] The original analog signals are input into a 16-bit ADC converter for 100Hz frequency digitization processing to obtain battery parameter data;

[0037] The battery parameter data are filtered by a Butterworth low-pass filter with a 10Hz cutoff frequency to obtain filtered battery parameter data;

[0038] The filtered battery parameter data is input into the improved ampere-hour metering method combined with the Kalman filtering algorithm for state of charge calculation to obtain a state of charge value;

[0039] The filtered battery parameter data is analyzed by a multi-parameter fusion method to obtain a health state value;

[0040] The temperature, state of charge value and internal resistance in the filtered battery parameter data are used to calculate the maximum power to obtain a power state value.

[0041] Specifically, the voltage, current, temperature and internal resistance of the battery are acquired by a high-precision data acquisition circuit to obtain raw analog signals. In this process, the voltage acquisition is usually measured by using a precision differential amplifier, and the measurement range is 0-5V; the current acquisition uses a Hall current sensor, which can non-contact measure the current value in the range of 0-100A; the temperature acquisition uses a PT100 type thermistor, and the measurement range is -20℃ to 80℃; the internal resistance acquisition is measured by a small signal alternating current excitation method, that is, a small signal alternating current is applied to the battery, and the internal resistance value is calculated by measuring the voltage response. These acquisition elements are connected to the data acquisition circuit through shielded wires to reduce environmental interference. The raw analog signals are input into a 16-bit ADC converter for 100Hz frequency digitization processing to obtain battery parameter data. The resolution of the 16-bit ADC converter is 2^16=65536 different digital values, which makes the voltage measurement accuracy reach the level of 0.1mV. The sampling frequency of 100Hz is sufficient to capture the dynamic changes of the battery parameters, and at the same time, it will not produce excessive data processing burden. The digitized data is stored in the form of an ordered array, containing sampling time stamp, voltage value, current value, temperature value and internal resistance value.

[0042] The battery parameter data is filtered by a Butterworth low-pass filter with a 10Hz cutoff frequency to remove noise, and the filtered battery parameter data is obtained. The Butterworth low-pass filter is a filter with maximum flat amplitude-frequency characteristics, which can effectively filter out high-frequency noise above 10Hz and retain the actual change trend of the battery parameters. The filtered data is smoother, reducing the data fluctuations caused by measurement noise, and providing a stable data basis for subsequent parameter calculation. The filtered battery parameter data is input into the improved ampere-hour metering method combined with the Kalman filtering algorithm for state of charge calculation to obtain a state of charge value. The improved ampere-hour metering method is a method of calculating the charge change by integrating the battery charging and discharging current, and correcting the SOC combined with the open circuit voltage. The Kalman filtering algorithm is a recursive estimation algorithm that continuously optimizes the SOC estimation value through the prediction-correction process to reduce the cumulative error. In actual calculation, the calculation of the state of charge value follows the following formula:

[0043]

[0044] wherein, Pmax represents the maximum allowed charging power (power state value), Pmax, temp represents the maximum charging power based on temperature limit, Pmax, soc represents the maximum charging power based on state of charge value limit, Pmax, r represents the maximum charging power based on internal resistance limit. These three parameters are calculated by the following ways respectively: Pmax, t is determined according to the difference between current temperature and safe temperature range, the closer the temperature is to the upper limit, the smaller it is; Pmax, soc is determined according to the state of charge value, the higher the state of charge value is, the smaller it is, to prevent overcharging; Pmax, r is determined according to the internal resistance value, the larger the internal resistance is, the smaller it is, to prevent overheating.

[0045] The filtered battery parameter data is analyzed for battery health by a multi-parameter fusion method to obtain a health state value. The multi-parameter fusion method comprehensively considers the battery capacity decay rate, internal resistance increase rate and charge-discharge efficiency change rate, and calculates the health state value by weighted average. The battery capacity decay rate is calculated by comparing the ratio of the current maximum charging capacity to the initial capacity; the internal resistance increase rate is calculated by comparing the ratio of the current internal resistance to the initial internal resistance; and the charge-discharge efficiency change rate is calculated by measuring the energy loss in the charge-discharge cycle. After normalization, these parameters are given different weight coefficients for weighted average to obtain a health state value in the range of 0-100%. Based on the temperature, state of charge value and internal resistance in the filtered battery parameter data, the maximum power is calculated to obtain a power state value. The power state value represents the maximum charging power that the current battery can safely accept, and is an important constraint condition for charging control. By comprehensively considering the influence of temperature on battery safety, the limitation of state of charge on overcharging and the influence of internal resistance on heating, the minimum power value under the restriction of the three is selected as the final power state value, to ensure that the charging process is always within the safe range.

[0046] For example, when the battery temperature is 35℃, the state of charge value is 80%, and the internal resistance is 60mΩ, the maximum charging power P_t under the temperature limit is first calculated as 120W (because the critical temperature 40℃ has not been reached), the maximum charging power P_s under the state of charge value limit is 80W (because the state of charge has reached 80% and the charging power needs to be reduced), and the maximum charging power P_i under the internal resistance limit is 100W (because the internal resistance value is large and the heat needs to be controlled). According to the principle of taking the minimum value of the three, the final calculated power state value is 80W, which means that the current battery charging power should not exceed 80W to ensure charging safety and prolong battery life. In this way, through accurate collection and scientific calculation of battery parameters, a reliable data basis is provided for subsequent charging control decisions based on the back propagation neural network model.

[0047] In a specific embodiment, the process of performing step S102 can specifically include the following steps:

[0048] The battery parameter data, the state of charge value, the state of health value, and the power state value are organized as input parameters into 10 input neurons, a four-layer neural network structure including an input layer, two hidden layers, and an output layer is constructed, and a back propagation neural network structure model is obtained;

[0049] The first hidden layer of the back propagation neural network structure model is set to have 16 neurons and a ReLU activation function, the second hidden layer of the back propagation neural network structure model is set to have 8 neurons and a ReLU activation function, and the output layer of the back propagation neural network structure model is set to have 4 neurons, thereby obtaining an initial back propagation neural network model;

[0050] The historical charging record data is input into the initial back propagation neural network model, and the back propagation training is performed through a mean square error loss function and an Adam optimizer, thereby obtaining an initial training model;

[0051] The initial training model is subjected to overfitting suppression through an L2 regularization method with a regularization coefficient of 0.0001, thereby obtaining a back propagation neural network model;

[0052] The back propagation neural network model is applied to real-time battery data processing to predict charging control parameters, thereby obtaining an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage, and a charging risk level.

[0053] Specifically, the battery parameter data, the state of charge value, the state of health value and the state of power value are organized as input parameters into 10 input neurons to construct a four-layer neural network structure including an input layer, two hidden layers and an output layer, and to obtain a back propagation neural network structure model. The 10 input neurons include the battery voltage, the current, the surface temperature, the ambient temperature, the internal resistance value, the state of charge value, the state of health value, the state of power value, the charging time and the historical charging cycle number. These parameters collectively reflect the current state and the historical use of the battery, and provide comprehensive input information for the neural network. The design of the input layer fully considers the key influencing factors in the battery charging process, and each input parameter is normalized to unify the parameters of different dimensions to the interval [0, 1], so as to avoid that some parameters with large factor values dominate the network learning process. The first hidden layer of the back propagation neural network structure model is set to 16 neurons and a ReLU activation function, the second hidden layer of the back propagation neural network structure model is set to 8 neurons and a ReLU activation function, and the output layer of the back propagation neural network structure model is set to 4 neurons, to obtain an initial back propagation neural network model. The expression of the ReLU (Rectified Linear Unit) activation function is f(x) = max(0, x), which is simpler to calculate than the traditional sigmoid or tanh activation function, and does not have the gradient vanishing problem when the input is positive, which helps to speed up the network training process. The 16 neurons of the first hidden layer enable the network to have sufficient complexity to capture the nonlinear relationship between the input parameters, and the 8 neurons of the second hidden layer further extract high-level features. The 4 neurons of the output layer correspond to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level respectively, and directly output the key parameters required for charging control.

[0054] The historical charging record data is input into the initial back propagation neural network model, and the back propagation training is performed through the mean square error loss function and the Adam optimizer to obtain an initial training model. The historical charging record data contains a large number of charging processes and results of batteries under different conditions, totaling about 100,000 samples, of which 80% are used for training and 20% are used for verification. The mean square error (MSE) loss function calculates the average of the sum of squares of the difference between the predicted value and the actual value, is suitable for regression problems, and can quantify the accuracy of the model prediction. The Adam optimizer is an adaptive learning rate optimization algorithm that combines the advantages of the momentum method and RMSProp, with an initial learning rate of 0.001, which can automatically adjust the learning rate according to the gradient changes to accelerate convergence and jump out of local optimization. The training process uses batch processing with 64 samples per batch to reduce memory usage and improve training speed. During the training process, the network weights are updated by calculating the gradient and back propagation, continuously reducing the prediction error, and gradually optimizing the model performance. The initial training model is subjected to L2 regularization with a regularization coefficient of 0.0001 to suppress overfitting, obtaining a back propagation neural network model. L2 regularization adds a penalty term of the sum of squares of the weights to the loss function to constrain the weight value, preventing the model from overfitting to the training data. The regularization coefficient of 0.0001 controls the regularization strength, balancing the fitting ability and generalization ability. At the same time, an early stopping mechanism is introduced during the training process, which stops training when the validation set error does not decrease for 50 consecutive iterations, avoiding the decline in generalization ability caused by overtraining. The model after L2 regularization has an accuracy of more than 95% on the validation set, a sensitivity of 98% for charging risk prediction, and a specificity of 96%, indicating that the model can correctly identify normal charging conditions and accurately predict potential risks.

[0055] The back propagation neural network model is applied to real-time battery data processing to predict charging control parameters, obtaining the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage, and the charging risk level. In the real-time application process, the current voltage, current, temperature, and other parameters of the battery are first collected, combined with the charging state value, health state value, and power state value calculated in the previous step, and then normalized and input into the trained neural network model. The network forward propagation calculates the weighted sum and activation value of each layer of neurons, and the final output layer has four neurons that output the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage, and the charging risk level. The optimal charging current value represents the current value that can ensure charging speed and protect the battery under current conditions; the maximum allowed charging current threshold is the upper limit of the current set from a safety perspective and should not be exceeded; the charging termination voltage is the charging cutoff voltage dynamically adjusted based on the current state of the battery; and the charging risk level is divided into four levels: low risk, medium risk, high risk, and extremely high risk, which are used to guide the adjustment of subsequent charging strategies.

[0056] For example, when charging management is performed on a group of lithium-ion batteries, the battery voltage is 3.6V, the current is 0A (not yet charging), the surface temperature is 30°C, the ambient temperature is 25°C, the internal resistance is 45mΩ, the state of charge value is 40%, the state of health value is 85%, the power state value is 120W, the charging duration is 0 minutes, and the historical charging cycle number is 200 times. After normalization processing, these data are input into the neural network, and through the nonlinear transformation of two hidden layers, the optimal charging current value is 2.2A, the maximum allowed charging current threshold is 3.5A, the charging termination voltage is 4.15V, and the charging risk level is low risk. This shows that the current battery state is good, and the charging current of 2.2A can be used, but it should not exceed 3.5A, and the charging voltage should not exceed 4.15V. These parameters directly guide the generation of subsequent charging strategies, ensuring that the charging process is efficient and safe, and fully embodying the core value of the battery charging protection method based on intelligent control. As the battery state changes, such as temperature rise and increase of state of charge value, the neural network will adjust the output parameters in real time, dynamically optimize the charging process, effectively prolong the service life of the battery and improve the charging safety.

[0057] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0058] Comparing and analyzing the optimal charging current value and the maximum allowed charging current threshold to obtain a safe charging current value;

[0059] Based on the safe charging current value and the charging risk level, a pre-charging phase is divided, and a charging control applied to a state of charge value of the battery below a specified threshold is obtained, to obtain pre-charging phase parameters;

[0060] Based on the safe charging current value and the charging risk level, a constant current charging phase is divided, and a charging control applied to a state of charge value of the battery below a specified threshold is obtained, to obtain constant current charging phase parameters;

[0061] Based on the charging termination voltage, a constant voltage charging phase is divided, and a charging control applied to a state of charge value of the battery below a specified threshold is obtained, to obtain constant voltage charging phase parameters;

[0062] Based on a specified proportion of the safe charging current value, a trickle charging phase is divided, and a pulse type charging control applied to a charging current below a specified threshold is obtained, to obtain trickle charging phase parameters;

[0063] The pre-charging phase parameters, the constant current charging phase parameters, the constant voltage charging phase parameters and the trickle charging phase parameters are combined into a charging current-time curve, a charging voltage-time curve and a PWM control signal, to obtain a charging strategy.

[0064] Specifically, the optimal charging current value and the maximum allowed charging current threshold are compared and analyzed to obtain a safe charging current value. The comparison and analysis process adopts a simple and intuitive minimum selection method, that is, taking the smaller value of the optimal charging current value and the maximum allowed charging current threshold as the safe charging current value. The purpose of such design is to take into account both charging efficiency and safety. When the optimal charging current value output by the neural network is less than the safety threshold, the optimal value is directly used; when the optimal value exceeds the safety threshold, the safety threshold is used as the reference to ensure that the charging current will not harm the battery. Based on the safe charging current value and the charging risk level, a pre-charging phase is divided, and pre-charging phase parameters are obtained for charging control applied when the battery charging state value is below a specified threshold. The pre-charging phase is mainly aimed at severely discharged batteries, and is started when the charging state value is below 10%. In this phase, the charging current is set to 0.1 times the safe charging current value, and small current charging can effectively avoid overheating problems caused by excessive internal resistance, while reducing the impact on the active material of the battery. The pre-charging phase continues until the battery voltage reaches 90% of the rated voltage, and the pre-charging phase parameters include the charging current value, the starting condition and the termination condition. For different charging risk levels, the pre-charging current will also be adjusted accordingly: the low-risk state maintains the original parameters; in the medium-risk state, the pre-charging current is further reduced to 0.08 times the safe charging current value; in the high-risk state, it is reduced to 0.05 times; and in the extremely high-risk state, the charging is suspended.

[0065] Based on the safe charging current value and the charging risk level, a constant current charging phase is divided, and constant current charging phase parameters are obtained for charging control applied before the battery voltage reaches a specified proportion of the charging termination voltage. The constant current charging phase is the main phase of battery charging, and the charging current in this phase is constant at the safe charging current value, and the battery voltage gradually rises with the charging process. The constant current charging phase continues until the battery voltage reaches 98% of the charging termination voltage, at which time the battery has charged about 80% of the capacity. The constant current charging phase parameters include the constant charging current value, the starting condition and the termination condition. Similarly, according to the charging risk level, the safe charging current value is dynamically adjusted: in the low-risk state, the original safe charging current value is maintained; in the medium-risk state, the charging current is reduced to 80% of the safe charging current value; in the high-risk state, it is reduced to 50%; and in the extremely high-risk state, the charging is suspended.

[0066] The constant-voltage charging phase is divided based on the charging termination voltage, and parameters of the constant-voltage charging phase are obtained for charging control applied to the battery voltage reaching the charging termination voltage. The constant-voltage charging phase keeps the charging voltage constant at the charging termination voltage, and the charging current gradually decreases as the battery charging proceeds. The main purpose of this phase is to safely charge the battery to near full state of charge, avoiding overvoltage damage to the battery. The constant-voltage charging phase continues until the charging current decreases to 0.1 times the safe charging current value, at which time the battery capacity has reached about 95%. The constant-voltage charging phase parameters include the constant charging voltage value, the start condition and the termination condition. The processing under different charging risk levels is as follows: the charging termination voltage given by the neural network is used in the low-risk state; in the medium-risk state, the termination voltage is reduced by 0.05V; in the high-risk state, the termination voltage is reduced by 0.1V; and in the extremely high-risk state, the charging is suspended.

[0067] The trickle charging phase is divided based on a specified proportion of the safe charging current value, and parameters of the trickle charging phase are obtained for pulse-type charging control applied to the charging current decreasing to a specified threshold. The trickle charging phase adopts pulse charging, with a pulse width modulation frequency of 1kHz and a duty cycle dynamically adjusted according to the battery temperature, ranging from 10% to 30%. This way can ensure sufficient charging while avoiding damage to the battery caused by long-time small current charging. The duration of the trickle charging phase is determined according to the battery health state value, and the lower the health state value, the shorter the trickle charging time, so as to reduce further damage to the aged battery. The trickle charging phase parameters include the pulse frequency, the duty cycle, the duration and the termination condition. Under different charging risk levels, the processing of the trickle charging is as follows: the normal execution in the low-risk state; the duration is shortened to 80% of the normal duration in the medium-risk state; the duration is shortened to 50% in the high-risk state; and the trickle charging phase is directly skipped in the extremely high-risk state.

[0068] The pre-charging phase parameters, the constant-current charging phase parameters, the constant-voltage charging phase parameters and the trickle charging phase parameters are combined into a charging current-time curve, a charging voltage-time curve and a PWM control signal to obtain a charging strategy. This process connects the discrete parameter points of each phase into a continuous curve through a time series interpolation algorithm, ensuring smooth transition of the charging process. The charging current-time curve describes the charging current size at each time point, the charging voltage-time curve describes the charging voltage level at each time point, and the PWM control signal is mainly used for pulse control in the trickle charging phase. The three sets of data together constitute a complete charging strategy, providing detailed control instructions for the subsequent power electronic control circuit.

[0069] For example, a set of lithium-ion batteries is controlled to charge, assuming that the optimal charging current value output by the back propagation neural network model is 2.5 A, the maximum allowed charging current threshold is 2.2 A, the charging termination voltage is 4.2 V, and the charging risk level is medium risk. First, the safe charging current value is determined to be 2.2 A (taking the minimum value) through comparison and analysis. Since the battery state of charge value is 5%, which is lower than the pre-charging threshold of 10%, the pre-charging phase is entered, and the charging current is set to 0.08 times the safe charging current value (because of medium risk), that is, 0.176 A. The pre-charging continues until the battery voltage rises to 3.78 V (90% of the rated voltage 4.2 V). Then, the constant current charging phase is entered, and the charging current is set to 80% of the safe charging current value (because of medium risk), that is, 1.76 A, and continues until the battery voltage reaches 4.116 V (98% of the termination voltage 4.2 V). Then, the constant voltage charging phase is entered, and the charging voltage is kept constant at 4.15 V (the termination voltage minus 0.05 V because of medium risk), and continues until the charging current drops to 0.176 A (0.1 times 1.76 A). Finally, the trickle charging phase is entered, and pulse charging with a frequency of 1 kHz and a duty cycle of 20% is adopted, and the duration is 80% of the normal value (because of medium risk). The entire charging process is output in the form of a charging current-time curve, a charging voltage-time curve, and a PWM control signal, constituting a complete charging strategy to guide the power electronic control circuit to accurately control the charging process and ensure efficient charging of the battery under safe conditions.

[0070] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0071] Input the charging current-time curve, the charging voltage-time curve, and the PWM control signal in the charging strategy into the power electronic control circuit with a digital signal processor as the core to obtain control instruction data;

[0072] Perform high-speed signal processing and PWM generation on the control instruction data through a field programmable gate array to obtain power switch control signals;

[0073] Drive the full-bridge LLC resonant conversion topology using silicon carbide MOSFET based on the power switch control signals to obtain accurate charging current and charging voltage outputs;

[0074] Perform closed-loop control on the charging current through a PI algorithm to obtain a current control loop;

[0075] Perform closed-loop control on the charging voltage through a PI algorithm to obtain a voltage control loop;

[0076] According to the automatic switching of the current control loop and the voltage control loop in the charging stage, the current control loop is used in the pre-charging and constant current stage, the voltage control loop is used in the constant voltage stage, and the PWM direct control is used in the trickle stage, so as to obtain the charging control result.

[0077] Specifically, the charging current-time curve, the charging voltage-time curve and the PWM control signal in the charging strategy are input to the power electronic control circuit with a digital signal processor as the core to obtain control instruction data. The digital signal processor (DSP) is a microprocessor specially used for digital signal processing, has high-speed computing capability and a special instruction set, and is suitable for real-time control applications. In the battery charging control, the DSP uses a TMS320F28335 chip with a main frequency of 150 MHz and 12-bit ADC conversion accuracy to quickly process the input charging strategy curve. The DSP first reads the charging current-time curve, the charging voltage-time curve and the PWM control signal data stored in the memory, looks up the corresponding target charging current, charging voltage and PWM parameters according to the current time point, and forms control instruction data. The control instruction data includes the accurate current value or voltage value to be output at present, and the PWM frequency and duty cycle information used in the trickle charging stage. The control instruction data is processed at high speed by a field programmable gate array (FPGA) to generate power switch control signals. The FPGA is an integrated circuit that can be programmed according to the internal logic structure, has high parallel processing capability and precise timing control capability. In the charging control system, the FPGA uses a Xilinx Spartan-6 series with a clock frequency of 100 MHz to receive the control instruction data sent by the DSP and process the high-speed signals. The FPGA realizes an accurate PWM generation module that can generate PWM signals with a frequency range from several Hz to several MHz and a duty cycle resolution of 0.1%. For the pre-charging and constant current charging stages, the FPGA generates corresponding PWM signals according to the output of the current control loop; for the constant voltage charging stage, the FPGA generates PWM signals according to the output of the voltage control loop; and for the trickle charging stage, the FPGA generates pulse signals directly according to the PWM parameters provided by the charging strategy. These PWM signals are collectively referred to as power switch control signals, which directly control the on and off of the switch tubes in the subsequent power conversion circuit. Based on the power switch control signals, a full-bridge LLC resonant conversion topology with silicon carbide MOSFET is driven to obtain accurate charging current and charging voltage output. The full-bridge LLC resonant conversion topology is a high-efficiency and low-noise power conversion circuit that includes a full-bridge switching circuit, an LLC resonant network and an output rectification and filtering circuit. The silicon carbide (SiC) MOSFET is a new type of wide-bandgap semiconductor power device that has lower on-resistance, higher switching frequency and better temperature characteristics than traditional silicon-based MOSFETs. In the charging system, the power switch control signals drive the silicon carbide MOSFET through an optoelectronic isolation drive circuit to control the on and off time of the silicon carbide MOSFET, thereby adjusting the output charging current and voltage. The LLC resonant network is composed of a resonant inductor, a magnetizing inductor and a resonant capacitor, which can realize soft switching, reduce switching loss and improve conversion efficiency.The switching frequency can be adjusted in the range of 100 kHz-500 kHz, and the output power is controlled by changing the switching frequency or PWM duty cycle. The output end is rectified by a Schottky diode, and a low ESR electrolytic capacitor and a ceramic capacitor are connected in parallel to filter the high-frequency square wave voltage into a stable DC voltage and current to supply the battery charging.

[0078] The charging current is closed-loop controlled by a PI algorithm to obtain a current control loop. PI (Proportional-Integral) control is a classic feedback control algorithm that precisely controls the system output by adjusting the proportional and integral coefficients. In the current control loop, the actual output current is first measured by a Hall current sensor and compared with the target current value in the control instruction data to calculate the current error. The PI controller calculates the control quantity according to the current error, the proportional term directly responds to the current error, and the integral term accumulates the historical error and eliminates the static error. The proportional coefficient of the current loop control is set to 0.05, the integral coefficient is set to 0.01, the control bandwidth is 10 kHz, which can quickly respond to current changes and maintain stability. The PI controller output is subjected to amplitude limiting and sent to the FPGA to convert into a PWM signal, which in turn adjusts the on-time of the power switch to form a closed-loop control, ensuring that the actual charging current is consistent with the target current. The charging voltage is closed-loop controlled by a PI algorithm to obtain a voltage control loop. The voltage control loop is similar to the current control loop, except that the control object changes from current to voltage. First, the actual output voltage is measured by a high-precision voltage dividing sampling circuit and compared with the target voltage value in the control instruction data to calculate the voltage error. The voltage PI controller calculates the control quantity according to the voltage error, the proportional coefficient of the voltage loop is set to 0.1, the integral coefficient is set to 0.02, and the control bandwidth is 1 kHz, which is lower than that of the current loop, because voltage changes are usually slower than current changes. The PI controller output is also subjected to amplitude limiting and sent to the FPGA to convert into a PWM signal, which adjusts the on-time of the power switch to form a voltage closed-loop control, ensuring that the actual charging voltage is consistent with the target voltage.

[0079] According to the automatic switching of the current control loop and the voltage control loop in the charging stage, the current control loop is used in the pre-charging and constant current charging stage, the voltage control loop is used in the constant voltage charging stage, and the PWM direct control is used in the trickle charging stage to obtain the charging control result. Different control methods are needed in different charging stages. The DSP automatically determines the current charging stage and switches the corresponding control loop according to the charging strategy and the real-time monitored battery state. In the pre-charging and constant current charging stage, the battery voltage has not reached the charging termination voltage, and the main target is to maintain constant current charging, so the current control loop is used to control the actual charging current to the target value. When the battery voltage rises close to the charging termination voltage, the system automatically switches to the constant voltage charging stage, and the voltage control loop is used to keep the charging voltage constant at the charging termination voltage, and the charging current naturally decreases with the change of the internal electrochemical state of the battery. When the charging current decreases to a specified threshold, the trickle charging stage is entered, and the closed-loop control is no longer used, but the PWM control signal defined in the charging strategy is directly used to charge the battery in a pulse manner until the charging is completed. The control results of the whole process include the actual output charging current, the charging voltage and the battery state data, which are recorded in real time and fed back to the safety protection mechanism for monitoring the safety of the charging process and adjusting the control strategy.

[0080] For example, when charging a group of large-capacity lithium batteries, the charging strategy is first sent to the DSP. The DSP reads that the current is in the pre-charging stage, and the target current is 0.5A, and converts this data into a control instruction and sends it to the FPGA. The FPGA generates the corresponding PWM signal, and the duty cycle is initially set to 10%, which drives the four silicon carbide MOSFET switch tubes in the full-bridge LLC resonant circuit. The actual output current measured by the Hall current sensor is 0.48A, which has an error of 0.02A from the target value of 0.5A. The current PI controller calculates the control amount, increases the PWM duty cycle to 10.5%, and makes the actual current reach the target value of 0.5A. As the battery voltage rises to 3.7V, the system determines that it enters the constant current charging stage, the target current increases to 2A, and the control loop still uses current control. When the battery voltage reaches 4.1V, the system switches to the constant voltage charging stage, and the voltage control loop is enabled to keep the charging voltage constant at 4.2V, while monitoring the gradual decrease of the charging current. When the charging current decreases to 0.2A, the system enters the trickle charging stage, and the 1kHz, 20% duty cycle PWM direct control is used to complete the final charging process. The whole charging process is accurately controlled by the power electronic control circuit, ensuring the safety and efficiency of the charging.

[0081] In a specific embodiment, the process of performing step S105 can specifically include the following steps:

[0082] Real-time monitoring is performed on the charging current, charging voltage, temperature and state of charge values in the battery parameter data, state evaluation is performed every specified time, and an abnormal state detection result is obtained;

[0083] The charging current is compared with the maximum allowed charging current threshold value, and when the charging current is greater than the maximum allowed charging current threshold value, an abnormal current alarm is triggered, and an abnormal current detection result is obtained;

[0084] The charging voltage is compared with the charging termination voltage, and when the charging voltage is greater than the charging termination voltage by a specified proportion, an abnormal voltage alarm is triggered, and an abnormal voltage detection result is obtained;

[0085] The temperature is judged by multiple thresholds, when the temperature exceeds the first threshold, the charging current is reduced, when the temperature exceeds the second threshold, the charging current is further reduced, and when the temperature exceeds the third threshold, the charging is immediately stopped, and a temperature abnormality protection result is obtained;

[0086] The abnormal current detection result, the abnormal voltage detection result and the temperature abnormality protection result are combined and processed to construct a multi-level protection mechanism of software protection layer and hardware protection layer, and a fault handling scheme is obtained;

[0087] Based on the fault handling scheme, the charging control result is adjusted, when a recoverable fault is detected, the charging parameters are adjusted to continue charging, when an unrecoverable fault is detected, the charging is terminated and an alarm signal is sent, and a safety protection result is obtained.

[0088] Specifically, the charging current, charging voltage, temperature and state of charge values in the battery parameter data are monitored in real time, and state evaluation is performed every specified time to obtain abnormal state detection results. During real-time monitoring, the sampling frequency is set to 10 Hz, that is, the battery parameter data is collected every 100 milliseconds. After data collection is completed, the latest collected data is stored in a circular buffer, and the last 30 seconds of historical data are retained for trend analysis. The state evaluation adopts a combination of threshold judgment method and change rate detection method, the threshold judgment is used to detect whether the parameters exceed the safe range, and the change rate detection is used to identify the abnormal change trend of the parameters. An abnormal state detection result is generated after each state evaluation, which contains the state flag bit and abnormal level of each parameter, the flag bit 0 represents normal, 1 represents warning, 2 represents abnormal, and 3 represents serious abnormality. The charging current is compared with the maximum allowed charging current threshold to determine whether the charging current is abnormal. When the charging current is greater than the maximum allowed charging current threshold, an abnormal current alarm is triggered, and a current abnormality detection result is obtained. The charging current abnormality detection sets a three-level judgment mechanism: when the actual charging current exceeds 105% but does not exceed 110% of the maximum allowed charging current threshold, a first-level warning is triggered, the warning event is recorded but the charging continues; when it exceeds 110% but does not exceed 120%, a second-level warning is triggered, and the charging current is automatically reduced to 95% of the maximum allowed value; when it exceeds 120%, a third-level warning is triggered, and the charging is immediately interrupted and an alarm is given. The current abnormality detection result contains the judgment result flag bit and the abnormal level, and records the duration and maximum overshoot amplitude of the current overshoot, which is used for subsequent fault diagnosis.

[0089] The charging voltage is compared with the charging termination voltage to determine whether the charging voltage is abnormal. When the charging voltage is greater than the charging termination voltage by a specified proportion, a voltage abnormality alarm is triggered, and a voltage abnormality detection result is obtained. The charging voltage abnormality detection also adopts a three-level judgment mechanism: when the actual charging voltage exceeds 102% but does not exceed 105% of the charging termination voltage, a first-level warning is triggered, the warning event is recorded but the charging continues; when it exceeds 105% but does not exceed 110%, a second-level warning is triggered, and it is immediately switched to constant voltage mode to control the voltage at the charging termination voltage value; when it exceeds 110%, a third-level warning is triggered, and the charging is immediately interrupted and an alarm is given. The voltage abnormality detection result contains the judgment result flag bit and the abnormal level, as well as the duration and maximum overshoot amplitude of the voltage overshoot.

[0090] The temperature data is subjected to multi-level threshold judgment, the charging current is reduced when the temperature exceeds a first threshold, the charging current is further reduced when the temperature exceeds a second threshold, and the charging is immediately stopped when the temperature exceeds a third threshold, to obtain a temperature anomaly protection result. The temperature multi-level threshold judgment mechanism sets four temperature threshold points: T1 = 40°C, T2 = 45°C, T3 = 50°C, and T4 = 55°C, which correspond to different protection measures. When the battery temperature exceeds T1 but is lower than T2, the charging current is reduced to 80% of the safe charging current value; when the temperature exceeds T2 but is lower than T3, the charging current is further reduced to 50% of the safe charging current value; when the temperature exceeds T3 but is lower than T4, the charging current is reduced to 20% of the safe charging current value; and when the temperature exceeds T4, the charging is immediately stopped and an alarm is triggered. The temperature anomaly protection result includes a temperature judgment result flag, an anomaly level, and a recommended charging current adjustment value.

[0091] The current anomaly detection result, the voltage anomaly detection result, and the temperature anomaly protection result are combined and processed to construct a multi-level protection mechanism of the software protection layer and the hardware protection layer, and a fault handling scheme is obtained. The combined processing adopts a priority ordering and logical merging manner, first, each detection result is ordered from high to low according to the anomaly level, and when multiple detections exist anomalies at the same time, the highest level processing measure is adopted. The software protection layer is mainly based on a safety monitoring program running in the DSP, with a millisecond-level response time, and is responsible for handling most abnormal situations; the hardware protection layer is based on a hardware comparator and a protection circuit in the power electronic control circuit, with a microsecond-level response time, and serves as the last safeguard of the software protection layer. The two form a complete multi-level protection mechanism, ensuring that abnormal situations can be responded to in a timely manner under any circumstances. The fault handling scheme is divided into two categories: recoverable faults and non-recoverable faults, the former can be recovered to normal charging through parameter adjustment, and the latter needs to interrupt charging and wait for manual intervention. Based on the fault handling scheme, the charging control result is adjusted, when a recoverable fault is detected, the charging parameters are adjusted to continue charging, and when a non-recoverable fault is detected, the charging is terminated and an alarm signal is sent, to obtain a safety protection result. For recoverable faults, such as slight temperature rise or current fluctuation, measures such as adjusting the charging current, switching the charging mode, or prolonging the charging time are taken to enable the charging process to continue but remain within a safe range. The specific adjustment strategy is guided by the recommended parameters in the fault handling scheme, such as reducing the charging current to a specific percentage of the original value, switching to the constant voltage stage in advance, or prolonging the trickle charging time. For non-recoverable faults, such as severe overheating, continuous overcurrent, or internal short circuit of the battery, the power output circuit is immediately turned off, the charging current is cut off, and an alarm signal is sent through an audible and visual alarm, while detailed fault information, including fault type, occurrence time, and related parameter values, is recorded for subsequent analysis. The safety protection result includes the adjusted charging parameter value after processing, the protection state flag, and the detailed fault record.

[0092] For example, when charging a set of vehicle-mounted power batteries, first, the battery parameter data is continuously collected by high-precision sensors. It is assumed that during the charging process, it is monitored that the battery temperature gradually rises from the initial 25°C to 42°C, exceeding the first temperature threshold T1 (40°C). The temperature anomaly protection mechanism immediately determines that the current belongs to the first level of temperature anomaly, and generates a temperature anomaly protection result, including a temperature anomaly flag bit 1 and a recommended charging current adjustment coefficient 0.8. At the same time, the current and voltage are normal, and the abnormal flag bits are all 0. The multi-level protection mechanism combines and processes the three detection results, since only the temperature has a first-level anomaly, so the final fault handling scheme is "mild temperature anomaly, belongs to recoverable fault, and the charging current needs to be reduced to 80% of the original value". Based on this scheme, the battery charging protection system adjusts the original charging current 2.5A to 2.0A (2.5A x 0.8), and continues to monitor the temperature change trend. After adjustment, the battery temperature rises slowly and stabilizes at 43°C, without triggering higher-level temperature thresholds. During the whole process, the charging can continue, and the temperature is controlled by actively reducing the charging current, ensuring the safety of the charging process, and reflecting the adaptive protection capability of the battery charging protection method based on intelligent control.

[0093] In a specific embodiment, the process of performing step S106 can specifically include the following steps:

[0094] Combine the safety protection result and the charging cycle data into a training sample set, corresponding to the battery response characteristics under different charging conditions, to obtain back propagation neural network model update data;

[0095] Normalize the back propagation neural network model update data to eliminate the dimension effect, to obtain standardized training data;

[0096] Input the standardized training data into the back propagation neural network model, calculate the loss function value according to the difference between the actual charging effect and the expected charging effect, to obtain the model error;

[0097] Based on the model error, adjust the weight parameters of the back propagation neural network model by gradient descent, to obtain the updated network parameters;

[0098] Perform time series analysis on the battery capacity and internal resistance in the battery parameter data, calculate the capacity attenuation trend and internal resistance change trend, to obtain battery aging characteristic data;

[0099] Based on the battery aging characteristic data, compensate and adjust the updated network parameters, and apply them to the prediction of control parameters in the next charging cycle, including the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage, and the optimization calculation of the charging risk level, to obtain the optimized battery charging protection control parameters.

[0100] Specifically, the safety protection results and the charging cycle data are combined into a training sample set, corresponding to the battery response characteristics under different charging conditions, to obtain the back propagation neural network model update data. The safety protection results include the processing records of various abnormal conditions, parameter adjustment records, and final charging effect evaluation during the charging process; the charging cycle data includes complete charging process records, such as time series data of charging current, voltage, temperature, charging duration, charging efficiency, and other information. These two types of data are matched and integrated according to the time stamp to form a structured training sample set. Each sample contains input features (initial battery state, charging conditions, control parameters) and output labels (actual charging effect, safety risk rating). The sample set is divided into multiple batches, each batch containing samples from different battery states and charging conditions to ensure that the model can learn the optimal control strategy under various conditions. The back propagation neural network model update data is normalized to eliminate the dimension effect and obtain standardized training data. Normalization is a standard preprocessing step in machine learning, aiming to unify different dimensional features into the same numerical range to avoid certain features dominating the training process due to their large values. A commonly used normalization method is the min-max normalization, which maps feature values to the [0, 1] interval. For continuous numerical features such as battery voltage, current, and temperature, first determine the historical maximum and minimum values of each feature, then map the original values to the normalized interval through linear transformation. For example, battery voltage usually varies between 2.5V and 4.2V, after normalization, 2.5V is mapped to 0 and 4.2V is mapped to 1, and the intermediate values are proportionally mapped. For discrete features such as charging risk levels, One-Hot Encoding is used to convert them into binary feature vectors. After normalization, all features are in the same numerical magnitude, which is beneficial for efficient training of the neural network.

[0101] The standardized training data is input into the back propagation neural network model, and the loss function value is calculated according to the difference between the actual charging effect and the expected charging effect to obtain the model error. In the forward propagation stage of the back propagation neural network model, the weighted input and activation value of each layer are calculated in turn. For the 10 neurons in the input layer, the normalized input features are directly received; for the two hidden layers, each neuron first calculates the weighted sum of the input, and then obtains the output through the ReLU activation function; the 4 neurons in the output layer also calculate the weighted sum, but do not apply the activation function, and directly output the predicted values. These predicted values are compared with the actual label values (actual optimal charging parameters) to calculate the loss function value. The loss function uses Mean Squared Error (MSE), which is the average of the squared difference between the predicted value and the actual value. The model error includes not only the overall loss value, but also the individual error of each output neuron, which is used to guide the subsequent parameter update.

[0102] The weight parameters of the backpropagation neural network model are adjusted by gradient descent based on the model error to obtain updated network parameters. Gradient descent is a commonly used parameter optimization algorithm in machine learning. By calculating the partial derivative (gradient) of the loss function with respect to each parameter, the parameters are updated in the opposite direction of the gradient to reduce the loss value. During backpropagation, the gradient of each layer's parameters is calculated layer by layer from the output layer. First, the partial derivative of the output layer error with respect to the weighted input of the output layer is calculated. Then, the gradient of the output layer weight is calculated according to the chain rule. Then, the error term of the hidden layer is calculated, and the gradient of the hidden layer weight is calculated. Finally, all parameters are updated using the Adam optimization algorithm. The Adam algorithm combines the advantages of the momentum method and RMSprop, and can adaptively adjust the learning rate to accelerate convergence and avoid local optima. After each parameter update, the model performance is evaluated using the validation set. If the performance decreases, the learning rate is reduced or the training is stopped early to prevent overfitting.

[0103] The battery capacity and internal resistance in the battery parameter data are analyzed in time series to calculate the capacity decay trend and internal resistance change trend to obtain battery aging characteristic data. Time series analysis is a method for studying the change law of data over time, and plays an important role in battery aging analysis. First, the battery capacity and internal resistance data of each charging period are extracted from the historical charging records to form two time series. For the capacity time series, the capacity change rate between adjacent charging periods is calculated, and the capacity decay curve is fitted, which usually presents an approximate exponential decay characteristic. For the internal resistance time series, the change rate is also calculated and the internal resistance growth curve is fitted, which usually presents an approximate linear or exponential growth characteristic. Combining the two curves, a battery aging characteristic model is established, which can predict the change trend of battery capacity and internal resistance in future charging periods. The aging characteristic data includes key indicators such as current capacity retention rate, internal resistance growth rate, and predicted remaining cycle number.

[0104] The updated network parameters are compensated and adjusted based on battery aging characteristic data, applied to the prediction of control parameters in the next charging cycle, including the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage, and the optimization calculation of the charging risk level, to obtain the optimized battery charging protection control parameters. The compensation adjustment is a process of adaptive correction of the neural network model according to the battery aging characteristics. Specifically, for the aged battery, a more conservative charging strategy is needed. First, the maximum allowed charging current threshold is linearly reduced according to the capacity decay rate, and the current threshold is reduced by 5% for every 10% reduction in capacity. Second, the optimal charging current value is adjusted according to the internal resistance growth rate, and the optimal charging current value is reduced by 10% for every 20% increase in internal resistance, to reduce the heat generated during charging. Third, the charging termination voltage is fine-tuned according to the comprehensive aging condition, and for the severely aged battery, the termination voltage is appropriately reduced, such as reducing the termination voltage by 0.5% for every 100 cycles. Finally, the determination standard of the charging risk level is adjusted according to the aging condition, so that the aged battery is more likely to trigger higher level of risk warning. These compensation adjustments directly affect the output layer of the neural network model, and the output results are dynamically adjusted without changing the network structure.

[0105] For example, when intelligent charging management is performed on a group of electric bicycle lithium batteries, a complete charging cycle is first completed, and the battery parameter data and safety protection results during the whole process are recorded. The data shows that the battery had a slight temperature anomaly during charging, and the safety protection mechanism reduced the charging current from the initial 2A to 1.6A, successfully controlling the temperature, and the final charging capacity was 10Ah. Comparing these data with the expected charging effect (target temperature range, charging time, charging capacity, etc.), it is found that the actual temperature control effect is good, but the charging time is slightly longer than expected. These information is organized into training samples, and after normalization, it is input into the neural network model. The model calculates the mean square error between the predicted value and the actual value as 0.015, and calculates the gradient of each layer parameter through the back propagation algorithm. The network parameters are updated using the Adam optimizer with a learning rate of 0.001, and converge after 20 iterations. At the same time, by analyzing the historical charging data, it is found that the battery has completed 200 charge and discharge cycles, the capacity has decreased from the initial 12Ah to 10Ah, and the decay rate is 16.7%; the internal resistance has increased from the initial 25mΩ to 35mΩ, and the growth rate is 40%. Based on these aging characteristic data, the output of the neural network model is compensated and adjusted: the maximum allowed charging current threshold is reduced from 2.5A to 2.3A (8.3% reduction), the optimal charging current value is reduced from 2A to 1.8A (10% reduction), and the charging termination voltage is fine-tuned from 4.2V to 4.18V. These optimized battery charging protection control parameters are applied to the next charging cycle, taking into account the actual response characteristics of the battery and adapting to the aging condition of the battery, achieving precise control and safety protection of the charging process.

[0106] The battery charging protection method based on intelligent control in the embodiments of the present application is described above, and the battery charging protection system based on intelligent control in the embodiments of the present application is described below. Please refer to Figure 2 An embodiment of the battery charging protection system based on intelligent control in the embodiments of the present application includes:

[0107] The acquisition module is configured to acquire parameters of voltage, current, temperature and internal resistance of the battery to obtain battery parameter data, and to calculate and process the battery parameter data to obtain a charging state value, a health state value and a power state value.

[0108] The construction module is configured to construct a back propagation neural network model based on the battery parameter data, the charging state value, the health state value and the power state value to obtain an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage and a charging risk level.

[0109] The generation module is configured to generate a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level to obtain a charging strategy.

[0110] The control module is configured to input the charging strategy into a power electronic control circuit for charging control to obtain a charging control result.

[0111] The execution module is configured to execute a multi-level security protection mechanism based on the battery parameter data, the charging strategy and the charging control result to obtain a security protection result.

[0112] The update module is configured to perform reinforcement learning update on the back propagation neural network model according to the security protection result and charging cycle data to obtain optimized battery charging protection control parameters.

[0113] Through the cooperation of the above-mentioned components, through the parameter collection of the voltage, current, temperature and internal resistance of the battery, combined with the state of charge value, the state of health value and the state of power value to construct a back propagation neural network model, a plurality of significant beneficial effects are realized. First, the application of the back propagation neural network model changes the battery charging control from the traditional fixed parameter mode to the adaptive intelligent control mode, the nonlinear mapping capability of the neural network can accurately capture the complex relationship between the battery parameters, and output the optimized charging control parameters, including the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage and the charging risk level, which significantly improves the accuracy and adaptability of the charging control. Secondly, the adaptive generation technology of the segmented charging curve dynamically adjusts the parameters of the pre-charging, constant current charging, constant voltage charging and trickle charging stages according to the real-time state of the battery, taking into account the charging efficiency and battery safety, effectively reducing the damage to the battery during the charging process. Thirdly, the precise cooperation of the power electronic control circuit and the charging strategy realizes the precise execution of the charging process through high-precision closed-loop control, ensuring the effective implementation of the charging strategy. Fourthly, the implementation of the multi-level safety protection mechanism builds a comprehensive safety protection system at the software and hardware levels, which can quickly respond to various abnormal situations, from slight parameter adjustment to emergency charging interruption, providing gradient protection measures, which significantly improves the safety of the charging process. Finally, the reinforcement learning update mechanism enables the system to have self-learning optimization capability, by continuously accumulating charging experience and optimizing the neural network model, the system performance continuously improves with the use time, and it can also adapt to the battery aging characteristics, providing optimized charging management for the whole life cycle of the battery. The characteristics of the back propagation neural network algorithm in the present application make a great contribution to the scheme, its multi-layer neuron structure can learn the complex nonlinear relationship in the battery charging process, the ReLU activation function improves the expression ability of the model, the L2 regularization method prevents overfitting, and ensures the generalization ability of the model, and the parameter optimization based on gradient descent enables the model to continuously adjust to adapt to the changes of battery characteristics. These algorithm characteristics together form an adaptive, accurate and safe intelligent battery charging control system, which solves the problems of fixed parameters, insufficient safety protection and poor adaptability in traditional charging methods.

[0114] The above Figure 2 The intelligent control based battery charging protection system in the embodiment of the present application is described in detail from the perspective of modular functional entities, and the intelligent control based battery charging protection device in the embodiment of the present application is described in detail from the perspective of hardware processing.

[0115] Figure 3is a structural schematic diagram of a battery charging protection device based on intelligent control provided by an embodiment of the present application. The battery charging protection device based on intelligent control 300 can have great differences due to different configurations or performances, and can include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, one or more storage media 330 (for example, one or more mass storage device ends) storing application programs 333 or data 332. The memory 320 and the storage medium 330 can be temporary storage or persistent storage. The programs stored in the storage medium 330 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the battery charging protection device based on intelligent control 300. Further, the processor 310 can be configured to communicate with the storage medium 330 and execute a series of instruction operations in the storage medium 330 on the battery charging protection device based on intelligent control 300 to realize the steps of the battery charging protection method based on intelligent control described above.

[0116] The battery charging protection device based on intelligent control 300 can also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input / output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that the battery charging protection device based on intelligent control 300 can also include other components, and the components shown in the figure are not exhaustive. Figure 3 The structure of the battery charging protection device based on intelligent control shown in the figure does not constitute a limitation on the battery charging protection device based on intelligent control provided by the present application, and can include more or fewer components than shown in the figure, or combine certain components, or different component arrangements.

[0117] The present application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium, and can also be a volatile computer readable storage medium. The computer readable storage medium has instructions stored therein, and when the instructions are run on a computer, the computer executes the steps of the battery charging protection method based on intelligent control.

[0118] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0119] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a smart control-based battery charging protection device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0120] The above-described embodiments are merely used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A battery charging protection method based on intelligent control, characterized in that, The method comprises the following steps: S1, collecting parameters of voltage, current, temperature and internal resistance of the battery to obtain battery parameter data, and performing calculation and processing on the battery parameter data to obtain state of charge value, state of health value and state of power value; S2, constructing a back propagation neural network model based on the battery parameter data, the state of charge value, the state of health value and the state of power value to obtain an optimal charging current value, a maximum allowable charging current threshold, a charging termination voltage and a charging risk level; S3, generating a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level to obtain a charging strategy; S4, inputting the charging strategy into a power electronic control circuit for charging control to obtain a charging control result; S5, executing a multi-level security protection mechanism based on the battery parameter data, the charging strategy and the charging control result to obtain a security protection result; S6, updating the back propagation neural network model based on the security protection result and the charging cycle data to obtain optimized battery charging protection control parameters; S2 comprises: organizing the battery parameter data, the state of charge value, the state of health value and the state of power value as input parameters into 10 input neurons to construct a four-layer neural network structure containing an input layer, two hidden layers and an output layer to obtain a back propagation neural network structure model; setting 16 neurons and ReLU activation functions for the first hidden layer of the back propagation neural network structure model, setting 8 neurons and ReLU activation functions for the second hidden layer of the back propagation neural network structure model, and setting 4 neurons for the output layer of the back propagation neural network structure model to obtain an initial back propagation neural network model; inputting historical charging record data into the initial back propagation neural network model to perform back propagation training through a mean square error loss function and an Adam optimizer to obtain an initial training model; performing overfitting suppression on the initial training model through L2 regularization and introducing an early stopping mechanism to obtain the back propagation neural network model; applying the back propagation neural network model to real-time battery data processing to predict charging control parameters to obtain the optimal charging current value, the maximum allowable charging current threshold, the charging termination voltage and the charging risk level; S6 includes: combining the security protection result and the charging cycle data into a training sample set, corresponding to the battery response characteristics under different charging conditions, obtaining back propagation neural network model update data; normalizing the back propagation neural network model update data to eliminate the influence of dimension, obtaining standardized training data; inputting the standardized training data into the back propagation neural network model, calculating the loss function value according to the difference between the actual charging effect and the expected charging effect, obtaining the model error; based on the model error, the weight parameters of the back propagation neural network model are adjusted by gradient descent, obtaining the updated network parameters; the battery capacity and internal resistance in the battery parameter data are analyzed in time sequence, the capacity attenuation trend and the internal resistance change trend are calculated, and the battery aging characteristic data are obtained; based on the battery aging characteristic data, the updated network parameters are compensated and adjusted, and are applied to the control parameter prediction of the next charging cycle, including the optimization calculation of the optimal charging current value, the maximum allowed charging current threshold, the charging termination voltage and the charging risk level, and the optimized battery charging protection control parameters are obtained.

2. The intelligent control based battery charging protection method according to claim 1, wherein, S1 includes: The voltage, current, temperature and internal resistance of the battery are acquired by a high-precision data acquisition circuit to obtain original analog signals; the original analog signals are input into a 16-bit ADC converter for 100Hz frequency digitization processing to obtain battery parameter data; the battery parameter data are filtered by a Butterworth low-pass filter with a 10Hz cutoff frequency to remove noise and obtain filtered battery parameter data; The filtered battery parameter data are input into an improved ampere-hour metering method combined with Kalman filtering algorithm for charging state calculation to obtain a charging state value; The filtered battery parameter data are analyzed by a multi-parameter fusion method to obtain a health state value; The temperature, charging state value and internal resistance in the filtered battery parameter data are used for maximum power calculation to obtain a power state value. 3.The smart control based battery charging protection method of claim 1, wherein, S3 includes: comparing and analyzing the optimal charging current value and the maximum allowed charging current threshold to obtain a safe charging current value; based on the safe charging current value and the charging risk level, a pre-charging stage is divided, and the charging control applied to the battery charging state value below the specified threshold is obtained to obtain pre-charging stage parameters; based on the safe charging current value and the charging risk level, a constant current charging stage is divided, and the charging control applied to the battery voltage reaching the specified proportion of the charging termination voltage is obtained to obtain constant current charging stage parameters; based on the charging termination voltage, a constant voltage charging stage is divided, and the charging control applied to the battery voltage reaching the charging termination voltage is obtained to obtain constant voltage charging stage parameters; based on the specified proportion of the safe charging current value, a trickle charging stage is divided, and the pulse type charging control applied to the charging current falling below the specified threshold is obtained to obtain trickle charging stage parameters; the pre-charging stage parameters, the constant current charging stage parameters, the constant voltage charging stage parameters and the trickle charging stage parameters are combined into a charging current-time curve, a charging voltage-time curve and a PWM control signal to obtain a charging strategy.

4. The smart control based battery charging protection method of claim 3, wherein, S4 includes: The charging current-time curve, the charging voltage-time curve and the PWM control signal in the charging strategy are input into a power electronic control circuit with a digital signal processor as the core to obtain control instruction data; the control instruction data is processed by a field programmable gate array to generate a power switch control signal; the full-bridge LLC resonant conversion topology using silicon carbide MOSFET is driven based on the power switch control signal to obtain accurate charging current and charging voltage outputs; The charging current is controlled by a PI algorithm to obtain a current control loop; The charging voltage is controlled by a PI algorithm to obtain a voltage control loop; the current control loop and the voltage control loop are automatically switched according to the charging stage; the current control loop is used in the pre-charging and constant-current stages; the voltage control loop is used in the constant-voltage stage; and the PWM direct control is used in the trickle stage to obtain a charging control result.

5. The intelligent control based battery charging protection method of claim 1, wherein, S5 includes: real-time monitoring of the charging current, the charging voltage, the temperature and the charging state value in the battery parameter data; performing state evaluation every specified time to obtain an abnormal state detection result; comparing the charging current with the maximum allowable charging current threshold value to determine whether the charging current is greater than the maximum allowable charging current threshold value, and triggering a current abnormality alarm when the charging current is greater than the maximum allowable charging current threshold value to obtain a current abnormality detection result; comparing the charging voltage with the charging termination voltage to determine whether the charging voltage is greater than the charging termination voltage by a specified proportion, and triggering a voltage abnormality alarm when the charging voltage is greater than the charging termination voltage by the specified proportion to obtain a voltage abnormality detection result; performing multi-level threshold value judgment on the temperature, and reducing the charging current when the temperature exceeds a first threshold value, further reducing the charging current when the temperature exceeds a second threshold value, and immediately stopping charging when the temperature exceeds a third threshold value to obtain a temperature abnormality protection result; combining the current abnormality detection result, the voltage abnormality detection result and the temperature abnormality protection result to construct a multi-level protection mechanism of a software protection layer and a hardware protection layer to obtain a fault handling scheme; adjusting the charging control result based on the fault handling scheme, adjusting the charging parameters to continue charging when a recoverable fault is detected, and terminating charging and issuing an alarm signal when an unrecoverable fault is detected to obtain a safety protection result.

6. A battery charging protection system based on intelligent control, characterized in that, The battery charging protection system based on intelligent control is used to implement the battery charging protection method based on intelligent control. The acquisition module is used to acquire the voltage, the current, the temperature and the internal resistance of the battery to obtain battery parameter data, and to calculate and process the battery parameter data to obtain a charging state value, a health state value and a power state value; The construction module is used to construct a back propagation neural network model based on the battery parameter data, the charging state value, the health state value and the power state value to obtain an optimal charging current value, a maximum allowable charging current threshold value, a charging termination voltage and a charging risk level; The generation module is used to generate a segmented charging curve according to the optimal charging current value, the maximum allowable charging current threshold value, the charging termination voltage and the charging risk level to obtain a charging strategy; The control module is used to input the charging strategy into a power electronic control circuit for charging control to obtain a charging control result. The execution module is configured to execute a multi-level security protection mechanism based on the battery parameter data, the charging strategy, and the charging control result, and obtain a security protection result; The updating module is configured to perform reinforcement learning updating on the back propagation neural network model according to the security protection result and charging cycle data, and obtain optimized battery charging protection control parameters.

7. A battery charging protection device based on intelligent control, characterized in that, The computer program is run on the processor, and the processor executes the computer program to implement the intelligent control-based battery charging protection method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is run on the processor, and the processor executes the computer program to implement the intelligent control-based battery charging protection method in any one of claims 1 to 5.

Images