Battery signal acquisition precision optimization method of BMS (Battery Management System)

Abstract

The invention relates to the technical field of battery management systems, and particularly discloses a battery signal acquisition precision optimization method of a BMS (Battery Management System), which comprises the following steps of: dynamically adjusting a covariance matrix of process noise and measurement noise by monitoring the voltage, current and temperature of a battery in real time so as to cope with the change of the working state of the battery; according to the change trend of the charge and discharge rate and the temperature, noise parameters are predicted and corrected, and the adaptability of the system in an extreme environment is improved; in combination with data from a plurality of sensors, weighted average and a self-adaptive adjustment mechanism of a Kalman filter are adopted, so that evaluation of the system on signal quality of different sensors and effective suppression of noise are enhanced; aiming at the sudden change of the battery state, the system can be ensured to quickly respond by automatically adjusting the filtering algorithm, the accurate monitoring of the battery health state is ensured, the stability and the accuracy of the battery management system can be improved, and the problem of estimation error under the extreme working condition of a traditional method is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management systems, and in particular to a method for optimizing battery signal acquisition accuracy of a BMS system. Background Art

[0002] With the rapid development of electric vehicles, energy storage systems, mobile devices and other fields, batteries, as one of the core components, undertake important energy storage and supply tasks. Battery health management, performance optimization and safety assurance have become the focus of battery management system (BMS) research. At present, battery management systems generally rely on real-time monitoring and data processing of key parameters such as battery voltage, current, temperature, etc. to ensure that the battery maintains a safe working state during the charging and discharging process. However, batteries often face various extreme working conditions in actual working processes, such as rapid charging and discharging, large temperature fluctuations, battery aging, etc. These factors may cause drastic changes in battery status, which brings huge challenges to traditional BMS systems.

[0003] Existing Kalman filters may not converge correctly under extreme conditions (such as rapid battery charging and discharging, high temperature or low temperature environment), resulting in inaccurate battery state estimation. When the battery operates in these extreme environments, the noise characteristics of the battery deviate greatly from the assumptions under standard operating conditions, causing the noise model of the Kalman filter to fail, and thus making it impossible for the system to effectively track the true state of the battery. This non-convergence may cause the battery management system to misjudge the battery's charge and discharge state, leading to serious problems such as overcharging, over-discharging, battery damage, and temperature control failure. Therefore, how to improve the adaptability and convergence of the Kalman filter in extreme environments is a major challenge in current technology. Summary of the invention

[0004] The object of the present invention is to provide a method for optimizing the battery signal acquisition accuracy of a BMS system to solve the above-mentioned problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A method for optimizing the accuracy of battery signal acquisition of a BMS system, S1: dynamically adjusting the noise covariance matrix of a Kalman filter by real-time monitoring of the battery working state including voltage, current and temperature; S2: According to the real-time changes of charge and discharge rate and temperature, the noise parameters of the Kalman filter are predicted and corrected to improve the filter's adaptability to dynamic changes of noise under extreme conditions; S3: Combine data from multiple sensors and use weighted average and Kalman filter adaptive adjustment mechanism to enhance the system's assessment of the quality of different sensor signals and effective noise suppression; S4: Extending the Kalman filter to handle nonlinear dynamics in the battery system and enhancing the accurate prediction of nonlinear factors such as fast charging and discharging and temperature changes; S5: When a sudden change in battery status is detected, the filtering algorithm is automatically adjusted to ensure that the system can respond quickly and accurately monitor the battery status in extreme environments.

[0006] As a further solution of the present invention: the noise covariance matrix of the dynamic adjustment Kalman filter specifically includes: Monitor the battery voltage, current and temperature data in real time, and dynamically construct the process noise covariance matrix. The calculation expression is: ; in, and is the constant coefficient based on battery characteristic fitting, is the current battery voltage, is the maximum battery voltage, is the battery voltage change rate, Represents the adjustment factor based on the working status of the battery. represents the current at the current moment, Indicates the current temperature. represents the process noise covariance matrix, represents the index of the time step, Represents a continuous time interval; The measurement noise covariance matrix is ​​dynamically constructed, and the calculation expression is: ; in, and is the constant coefficient fitted according to the sensor performance, represents the adjustment factor for the measurement noise model, represents the measurement noise covariance matrix; Dynamically update the covariance matrix of the Kalman filter, the calculation expression is: ; In the formula, represents the covariance matrix of the Kalman filter.

[0007] As a further solution of the present invention: the adaptability of the filter to dynamic changes of noise under extreme conditions is improved, specifically including: A training set is built based on historical noise data, a noise prediction model is constructed, and the current noise parameters are predicted in combination with a machine learning algorithm. At the same time, corrections are made by real-time monitoring of the battery's temperature change rate and charge and discharge power change trends.

[0008] As a further solution of the present invention: the specific process of the correction is: Collect historical data of the battery under different working conditions, including battery voltage, current and temperature; Perform denoising and normalization preprocessing on the collected historical data; Selecting a machine learning algorithm to establish a noise prediction model, the algorithm being a long short-term memory network; The input data is the battery historical status data; The output data are the predicted values ​​of the noise parameters; Through the machine learning model, a noise prediction formula is established to predict the noise parameters at the current moment according to the real-time data of the battery. The noise prediction formula is: ; in, represents the predicted current process noise parameter, Indicates the historical battery voltage, represents the historical battery current, Indicates the historical battery temperature, Represents the historical time index, Represents the long short-term memory network model; By real-time monitoring of the battery temperature change rate and charge and discharge power change trend, the noise prediction value is corrected. The correction formula is: ; in, represents the corrected noise parameter, represents the rate of temperature change, represents a continuous time interval, Indicates the charging and discharging power change trend, and Indicates the correction factor.

[0009] As a further solution of the present invention: the combining of data from multiple sensors and the use of weighted average and Kalman filter adaptive adjustment mechanism specifically include: Weights are assigned to signals from different sensors according to their reliability, where the weights are determined by the noise level and signal consistency of the real-time monitored sensor signals, and data fusion is achieved using a dynamic weighted average method.

[0010] As a further solution of the present invention: the processing of nonlinear dynamics in the battery system by using an extended Kalman filter specifically includes: The dynamic state variables of the battery, including state of charge, state of health, and internal impedance, are estimated through nonlinear state transfer equations to improve the estimation accuracy under nonlinear conditions.

[0011] As a further solution of the present invention: the acquisition process of the nonlinear state transfer equation is: In order to calculate the dynamic state variables of the battery, a nonlinear state transfer equation is constructed to describe the dynamic behavior of the battery under various working conditions. The specific calculation method is as follows: Based on the changes in battery current, voltage and temperature, the nonlinear state transfer equation of the state of charge is calculated. The calculation expression is: ; in, Indicates The state of charge at the moment, Indicates the collection time. Indicates the state of charge at the previous moment, Indicates The battery current at the moment, Indicates The battery voltage at the moment, The battery open circuit voltage calculated based on the current state of charge, Indicates based on Calculated battery internal impedance, Indicates the amount of change at each moment; The nonlinear state transfer equation for calculating the internal impedance is expressed as: ; in, represents the initial internal impedance of the battery, Indicates The state of charge at the moment, Indicates the preset scale factor, Represents the index that controls the rate of change of the battery's internal resistance.

[0012] As a further solution of the present invention: when a sudden change in the battery state is detected, the filtering algorithm is automatically adjusted, specifically including: By setting the error tolerance threshold, when the filter error is detected to exceed the threshold, the Kalman gain coefficient is automatically adjusted to ensure the filter convergence and system stability; The calculation process of filtering error is: The filtering error is calculated based on the difference between the measured value and the predicted value at the current moment. The calculation expression is: ; in, represents the filtering error, Indicates the collection time. Indicates The actual measured value at the moment, It is based on the previous state prediction value.

[0013] As a further solution of the present invention: the process of obtaining the Kalman gain coefficient specifically includes: If the filtering error is greater than the error tolerance threshold, the Kalman gain coefficient is adjusted, and the calculation expression is: ; In the formula, Indicates the collection time. Indicates The Kalman gain coefficient at time , represents the error covariance matrix at the previous moment, represents the filtering error at the current moment, Represents the adjustment factor.

[0014] As a further solution of the present invention: the error covariance matrix is ​​specifically updated and calculated as follows: To ensure the stability and accuracy of the filter, the error covariance matrix is ​​updated to reflect the state estimation accuracy at the current moment. The calculation expression is: ; In the formula, represents the updated error covariance matrix, Indicates The Kalman gain coefficient at time , represents the observation matrix, represents the identity matrix, represents the error covariance matrix at the previous moment, Indicates the collection time.

[0015] Beneficial effects of the present invention: (1) The present invention realizes fine optimization of battery state estimation by real-time monitoring of the battery charge and discharge rate and temperature changes, and by dynamically adjusting the noise covariance matrix of the Kalman filter. In the traditional Kalman filter method, the noise model is usually fixed, which makes the estimation accuracy of the filter susceptible to significant errors when facing drastic changes in the battery working state or the influence of the external environment. In contrast, the present invention can dynamically adjust the parameters of the filter according to the actual working conditions of the battery by acquiring the working state parameters of the battery in real time, and predicting and correcting the noise parameters based on these real-time data, so as to ensure that the system can still accurately estimate the battery state when the battery is in extreme conditions of rapid charge and discharge or temperature fluctuations. This method not only improves the robustness of the battery management system, but also enables the system to flexibly adapt to different working environments and application scenarios, greatly reducing the estimation error caused by the fixed traditional noise model, and improving the system's adaptability and stability. Therefore, the present invention can maintain high-precision battery state monitoring under variable working environments, ensure that the battery management system can operate stably under various complex working conditions, and effectively extend the battery life.

[0016] (2) The present invention realizes efficient fusion and optimization of multiple sensor data by combining dynamic weighted average and Kalman filter adaptive adjustment mechanism. Specifically, the system monitors the signal quality of each sensor in real time, and dynamically adjusts the weight of each sensor based on the noise level and consistency of the signal, so that the sensors with higher reliability occupy a larger proportion in the data fusion, thereby reducing the interference of noise and error on the system performance. This mechanism can flexibly adjust the data fusion strategy according to the real-time quality of the sensor signal to ensure the accuracy and stability of the system under various working conditions. More importantly, when the battery state suddenly changes (such as battery failure, extreme cases of drastic temperature fluctuations), the system can automatically identify this change and quickly adjust the filtering algorithm, optimize the setting of the Kalman gain, and ensure that the system can still respond quickly and accurately monitor the battery state in emergencies. This mechanism significantly improves the adaptability of the battery management system in dynamic and complex environments, not only enhancing the robustness of the system and reducing the risks caused by environmental or equipment failures, but also effectively improving the stability and accuracy of the overall system, ensuring the safe and reliable operation of the battery under various extreme conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the accompanying drawings.

[0018] Figure 1 It is a specific flow chart of a method for optimizing the battery signal acquisition accuracy of a BMS system of the present invention; Figure 2 It is a flowchart of dynamically adjusting the noise covariance matrix of the Kalman filter in the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention is a method for optimizing the battery signal acquisition accuracy of a BMS system, comprising the following steps: S1: Dynamically adjust the noise covariance matrix of the Kalman filter by real-time monitoring of the battery operating status including voltage, current and temperature; S2: According to the real-time changes of charge and discharge rate and temperature, the noise parameters of the Kalman filter are predicted and corrected to improve the filter's adaptability to dynamic changes of noise under extreme conditions; S3: Combine data from multiple sensors and use weighted average and Kalman filter adaptive adjustment mechanism to enhance the system's assessment of the quality of different sensor signals and effective noise suppression; S4: Extending the Kalman filter to handle nonlinear dynamics in the battery system and enhancing the accurate prediction of nonlinear factors such as fast charging and discharging and temperature changes; S5: When a sudden change in battery status is detected, the filtering algorithm is automatically adjusted to ensure that the system can respond quickly and accurately monitor the battery status in extreme environments.

[0021] In S1, the noise covariance matrix of the Kalman filter is dynamically adjusted by real-time monitoring of the battery operating status including voltage, current and temperature, including: The battery voltage is collected through the voltage sensor in the battery management system (BMS). Specifically, the BMS system connects multiple battery cells in parallel and uses accurate voltage sensors to monitor the voltage of each battery cell in real time. The analog signal is converted into a digital signal through an analog-to-digital converter (ADC) for subsequent processing. The sampling frequency is set according to system requirements, ranging from several times to dozens of times per second, to ensure real-time capture of voltage changes under different loads and working conditions, and to ensure the timeliness and accuracy of the data.

[0022] The battery current is collected by current sensors, and common sensor types include Hall effect sensors. When the current flows through the shunt resistor, the magnitude of the current will cause the voltage across the resistor to drop, and the BMS system calculates the current magnitude through the voltage difference. The Hall effect sensor senses the intensity of the current by detecting changes in the magnetic field. The collected current signal is also converted into a digital signal through an analog-to-digital converter (ADC), synchronized with the voltage collection data, and processed and analyzed by the BMS system.

[0023] The battery temperature is collected through temperature sensors. The BMS system deploys multiple temperature sensors at key locations of the battery pack (such as the positive and negative electrodes of the battery cells) to monitor the temperature changes of the battery in real time. The temperature signal is collected through analog signals and converted into a digital signal through an analog-to-digital converter (ADC) for processing in the BMS.

[0024] See also Figure 2 As shown, the noise covariance matrix of the dynamic adjustment Kalman filter specifically includes: Monitor the battery voltage, current and temperature data in real time, and dynamically construct the process noise covariance matrix. The calculation expression is: ; in, and is the constant coefficient based on battery characteristic fitting, is the current battery voltage, is the maximum battery voltage, is the battery voltage change rate, Represents the adjustment factor based on the working status of the battery. represents the current at the current moment, Indicates the current temperature. represents the process noise covariance matrix, represents the index of the time step, Represents a continuous time interval; The measurement noise covariance matrix is ​​dynamically constructed, and the calculation expression is: ; in, and is the constant coefficient fitted according to the sensor performance, represents the adjustment factor for the measurement noise model, represents the measurement noise covariance matrix; Dynamically update the covariance matrix of the Kalman filter, the calculation expression is: ; In the formula, represents the covariance matrix of the Kalman filter.

[0025] It should be noted that by dynamically updating the covariance matrix, the Kalman filter can adaptively adjust the Kalman gain to ensure that the system can still maintain high-precision state estimation under extreme conditions such as rapid battery charging and discharging and temperature changes. This solves the estimation error and filter convergence problems caused by the fixed noise model in traditional technologies, and enhances the robustness and adaptability of the system.

[0026] In S2, the noise parameters of the Kalman filter are predicted and corrected according to the real-time changes in the charge and discharge rate and temperature, so as to improve the adaptability of the filter to the dynamic changes of noise under extreme conditions, including: Build a training set based on historical noise data, build a noise prediction model, and use machine learning algorithms to predict current noise parameters. At the same time, make corrections by monitoring the temperature change rate and charge and discharge power change trends of the battery in real time. The specific process of the correction is as follows: Collect historical data of the battery under different working conditions, including battery voltage, current and temperature; Perform denoising and normalization preprocessing on the collected historical data; Selecting a machine learning algorithm to establish a noise prediction model, the algorithm being a long short-term memory network; The input data is the battery historical status data; The output data are the predicted values ​​of the noise parameters; Through the machine learning model, a noise prediction formula is established to predict the noise parameters at the current moment according to the real-time data of the battery. The noise prediction formula is: ; in, represents the predicted current process noise parameter, Indicates the historical battery voltage, represents the historical battery current, Indicates the historical battery temperature, Represents the historical time index, Represents the long short-term memory network model; By real-time monitoring of the battery temperature change rate and charge and discharge power change trend, the noise prediction value is corrected. The correction formula is: ; in, represents the corrected noise parameter, represents the rate of temperature change, represents a continuous time interval, Indicates the charging and discharging power change trend, and Indicates the correction factor.

[0027] It should be noted that the noise prediction model can intelligently adjust the noise prediction value according to the actual working state of the battery, improve the adaptability to the dynamic changes of noise of the battery under different working conditions, and correct the noise prediction value by real-time monitoring of the temperature change rate and charge and discharge power change trend of the battery, thereby enhancing the stability and accuracy of the system under extreme working conditions.

[0028] In S3, data from multiple sensors are combined, and the weighted average and Kalman filter adaptive adjustment mechanism are used to enhance the system's assessment of the quality of different sensor signals and effective noise suppression, including: Weights are assigned to signals from different sensors according to their reliability, where the weights are determined by the noise level and signal consistency of the real-time monitored sensor signals, and data fusion is achieved using a dynamic weighted average method.

[0029] It should be noted that the system first monitors the noise level of each sensor signal through a noise detection algorithm. Sensors with lower signal noise will receive higher weights. At the same time, by analyzing the consistency between the sensor signals, if multiple sensor signals show high consistency, these signals will also be given higher weights. Based on these weights, the data of multiple sensors are fused in a dynamic weighted average manner to ensure that the final fusion result reflects the reliability of each sensor, minimize the impact of noise and errors, and improve the accuracy and stability of data fusion. This dynamic weighting mechanism enables the system to flexibly adjust the data acquisition strategy according to the actual performance of the sensor in different working environments, thereby improving the reliability and accuracy of the overall system.

[0030] In S4, the nonlinear dynamics in the battery system are processed by extending the Kalman filter to enhance the accurate prediction of nonlinear factors such as fast charging and discharging and temperature changes, including: The dynamic state variables of the battery, including state of charge, state of health, and internal impedance, are estimated through nonlinear state transfer equations to improve the estimation accuracy under nonlinear conditions.

[0031] The acquisition process of the nonlinear state transfer equation is: In order to calculate the dynamic state variables of the battery, a nonlinear state transfer equation is constructed to describe the dynamic behavior of the battery under various working conditions. The specific calculation method is as follows: Based on the changes in battery current, voltage and temperature, the nonlinear state transfer equation of the state of charge is calculated. The calculation expression is: ; in, Indicates The state of charge at the moment, Indicates the collection time. Indicates the state of charge at the previous moment, Indicates The battery current at the moment, Indicates The battery voltage at the moment, The battery open circuit voltage calculated based on the current state of charge, Indicates based on Calculated battery internal impedance, Indicates the amount of change at each moment; The nonlinear state transfer equation for calculating the internal impedance is expressed as: ; in, represents the initial internal impedance of the battery, Indicates The state of charge at the moment, Indicates the preset scale factor, Represents the index that controls the rate of change of the battery's internal resistance.

[0032] It should be noted that: by constructing nonlinear state transfer equations, the dynamic behavior of the battery under various working conditions can be accurately described, especially the changes in battery state under the influence of nonlinear factors such as rapid charging and discharging and temperature changes. The nonlinear state transfer equation can not only effectively improve the prediction accuracy of battery health status and battery performance, but also maintain high-precision battery state estimation under extreme working conditions, enhance the stability and robustness of the system, solve the estimation error caused by ignoring nonlinear factors in traditional methods, and improve the overall performance of the battery management system.

[0033] In S5, when a sudden change in battery status is detected, the filtering algorithm is automatically adjusted to ensure that the system can respond quickly and accurately monitor the battery status in extreme environments, including: By setting the error tolerance threshold, when the filter error is detected to exceed the threshold, the Kalman gain coefficient is automatically adjusted to ensure the filter convergence and system stability; The calculation process of filtering error is: The filtering error is calculated based on the difference between the measured value and the predicted value at the current moment. The calculation expression is: ; in, represents the filtering error, Indicates the collection time. Indicates The actual measured value at the moment, It is based on the previous state prediction value.

[0034] If the filtering error is greater than the error tolerance threshold, the Kalman gain coefficient is adjusted, and the calculation expression is: ; In the formula, Indicates the collection time. Indicates The Kalman gain coefficient at time , represents the error covariance matrix at the previous moment, represents the filtering error at the current moment, Represents the adjustment factor.

[0035] To ensure the stability and accuracy of the filter, the error covariance matrix is ​​updated to reflect the state estimation accuracy at the current moment. The calculation expression is: ; In the formula, represents the updated error covariance matrix, Indicates The Kalman gain coefficient at time , represents the observation matrix, represents the identity matrix, represents the error covariance matrix at the previous moment, Indicates the collection time.

[0036] The working principle of the present invention is to monitor the voltage, current and temperature of the battery in real time, and dynamically adjust the noise covariance matrix of the Kalman filter based on these data to ensure that the filter can adapt to the noise level under different battery states, thereby improving the estimation accuracy. On this basis, the present invention further introduces real-time data based on the charge and discharge rate and temperature changes to predict and correct the noise parameters of the Kalman filter, thereby improving its adaptability to dynamic changes in noise under extreme conditions. Through the dynamic weighted average method, combined with data from multiple sensors, the system assigns weights according to the noise level and consistency of each sensor signal, and adopts an adaptive adjustment mechanism to fuse data to ensure that the system can effectively suppress noise and reduce errors under multi-sensor input. In view of the nonlinear dynamic behavior in the battery system, the present invention processes the nonlinear state transfer equation of the battery by extending the Kalman filter, accurately predicts the state of charge, health state and internal impedance of the battery, especially under the influence of nonlinear factors such as rapid charge and discharge and temperature changes, maintains high-precision estimation, and further optimizes the stability and robustness of the system.

[0037] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0038] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0039] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0040] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0041] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for optimizing the battery signal acquisition accuracy of a BMS system, characterized in that: The following steps are involved: S1: Dynamically adjust the noise covariance matrix of the Kalman filter by real-time monitoring of the battery operating status including voltage, current and temperature; S2: According to the real-time changes of charge and discharge rate and temperature, the noise parameters of the Kalman filter are predicted and corrected to improve the filter's adaptability to dynamic changes of noise under extreme conditions; S3: Combine data from multiple sensors and use weighted average and Kalman filter adaptive adjustment mechanism to enhance the system's assessment of the quality of different sensor signals and effective noise suppression; S4: Extending the Kalman filter to handle nonlinear dynamics in the battery system and enhancing the accurate prediction of nonlinear factors such as fast charging and discharging and temperature changes; S5: When a sudden change in battery status is detected, the filtering algorithm is automatically adjusted to ensure that the system can respond quickly and accurately monitor the battery status in extreme environments.

2. The method for optimizing the battery signal acquisition accuracy of a BMS system according to claim 1, characterized in that: The dynamic adjustment of the noise covariance matrix of the Kalman filter specifically includes: Monitor the battery voltage, current and temperature data in real time, and dynamically construct the process noise covariance matrix. The calculation expression is: ； in, and is the constant coefficient based on battery characteristic fitting, is the current battery voltage, is the maximum battery voltage, is the battery voltage change rate, Represents the adjustment factor based on the working status of the battery. represents the current at the current moment, Indicates the current temperature. represents the process noise covariance matrix, represents the index of the time step, Represents a continuous time interval; Dynamically construct the measurement noise covariance matrix, and the calculation expression is: ； in, and is the constant coefficient fitted according to the sensor performance, represents the adjustment factor for the measurement noise model, represents the measurement noise covariance matrix; Dynamically update the covariance matrix of the Kalman filter, the calculation expression is: ； In the formula, represents the covariance matrix of the Kalman filter.

3. The method for optimizing the battery signal acquisition accuracy of a BMS system according to claim 1, characterized in that: The improvement of the adaptability of the filter to dynamic changes of noise under extreme conditions specifically includes: A training set is built based on historical noise data, a noise prediction model is constructed, and the current noise parameters are predicted in combination with a machine learning algorithm. At the same time, corrections are made by real-time monitoring of the battery's temperature change rate and charge and discharge power change trends.

4. The method for optimizing the battery signal acquisition accuracy of a BMS system according to claim 3, characterized in that: The specific process of the correction is as follows: Collect historical data of the battery under different working conditions, including battery voltage, current and temperature; Perform denoising and normalization preprocessing on the collected historical data; Selecting a machine learning algorithm to establish a noise prediction model, the algorithm being a long short-term memory network; The input data is the battery history status data; The output data are the predicted values ​​of the noise parameters; Through the machine learning model, a noise prediction formula is established to predict the noise parameters at the current moment according to the real-time data of the battery. The noise prediction formula is: ； in, represents the predicted current process noise parameter, Indicates the historical battery voltage, represents the historical battery current, Indicates the historical battery temperature, Represents the historical time index, Represents the long short-term memory network model; By real-time monitoring of the battery temperature change rate and charge and discharge power change trend, the noise prediction value is corrected. The correction formula is: ； in, represents the corrected noise parameter, represents the rate of temperature change, represents a continuous time interval, Indicates the charging and discharging power change trend, and Indicates the correction factor.

5. The method for optimizing battery signal acquisition accuracy of a BMS system according to claim 1, characterized in that: The method of combining data from multiple sensors and utilizing weighted average and Kalman filter adaptive adjustment mechanism specifically includes: Weights are assigned to signals from different sensors according to their reliability, where the weights are determined by the noise level and signal consistency of the real-time monitored sensor signals, and data fusion is achieved using a dynamic weighted average method.

6. The method for optimizing battery signal acquisition accuracy of a BMS system according to claim 1, characterized in that: The method of processing nonlinear dynamics in the battery system by using an extended Kalman filter specifically includes: The dynamic state variables of the battery, including state of charge, state of health, and internal impedance, are estimated through nonlinear state transfer equations to improve the estimation accuracy under nonlinear conditions.

7. The method for optimizing the battery signal acquisition accuracy of a BMS system according to claim 6, characterized in that: The acquisition process of the nonlinear state transfer equation is: In order to calculate the dynamic state variables of the battery, a nonlinear state transfer equation is constructed to describe the dynamic behavior of the battery under various working conditions. The specific calculation method is as follows: Based on the changes in battery current, voltage and temperature, the nonlinear state transfer equation of the state of charge is calculated. The calculation expression is: ； in, Indicates The state of charge at the moment, Indicates the collection time. Indicates the state of charge at the previous moment, Indicates The battery current at the moment, Indicates The battery voltage at the moment, The battery open circuit voltage calculated based on the current state of charge, Indicates based on Calculated battery internal impedance, Indicates the amount of change at each moment; The nonlinear state transfer equation for calculating the internal impedance is expressed as: ； in, represents the initial internal impedance of the battery, Indicates The state of charge at the moment, Indicates the preset scale factor, Represents the index that controls the rate of change of the battery's internal resistance.

8. The method for optimizing battery signal acquisition accuracy of a BMS system according to claim 1, characterized in that: When a sudden change in the battery state is detected, the filtering algorithm is automatically adjusted, specifically including: By setting the error tolerance threshold, when the filter error is detected to exceed the threshold, the Kalman gain coefficient is automatically adjusted to ensure the filter convergence and system stability; The calculation process of filtering error is: The filtering error is calculated based on the difference between the measured value and the predicted value at the current moment. The calculation expression is: ； in, represents the filtering error, Indicates the collection time. Indicates The actual measured value at the moment, It is based on the previous state prediction value.

9. The method for optimizing battery signal acquisition accuracy of a BMS system according to claim 8, characterized in that: The process of obtaining the Kalman gain coefficient specifically includes: If the filtering error is greater than the error tolerance threshold, the Kalman gain coefficient is adjusted, and the calculation expression is: ； In the formula, Indicates the collection time. Indicates The Kalman gain coefficient at time , represents the error covariance matrix at the previous moment, represents the filtering error at the current moment, Represents the adjustment factor.

10. The method for optimizing battery signal acquisition accuracy of a BMS system according to claim 9, characterized in that: The specific update calculation method of the error covariance matrix is: To ensure the stability and accuracy of the filter, the error covariance matrix is ​​updated to reflect the state estimation accuracy at the current moment. The calculation expression is: ； In the formula, represents the updated error covariance matrix, Indicates The Kalman gain coefficient at time , represents the observation matrix, represents the identity matrix, represents the error covariance matrix at the previous moment, Indicates the collection time.

Images