Battery safety monitoring method and system for predicting battery voltage-temperature based on extended Kalman filtering

Abstract

The invention discloses a battery safety monitoring method and system for predicting battery voltage-temperature based on extended Kalman filtering, and belongs to the technical field of batteries, and a processing module included in the system is configured to execute the battery safety monitoring method. The method comprises the following steps: acquiring a voltage measurement value, a current measurement value and a temperature measurement value in real time during operation of a battery through a battery sensor; constructing a battery electric model and a battery electric heating model; based on an extended Kalman filtering algorithm, a battery voltage-temperature prediction model is established in combination with the battery electric model and the battery electric heating model; generating a battery voltage estimated value and a battery temperature estimated value by using the battery voltage-temperature prediction model; calculating a residual error between a battery voltage measurement value and an estimation value and a residual error between a battery temperature measurement value and an estimation value; the residual error is compared with a preset safety threshold value, and the safety state of the battery is judged; and when the residual error exceeds a safety threshold and lasts for a preset time, triggering an alarm and executing corresponding safety measures.

Description

Technical Field

[0001] This invention belongs to the field of battery technology, specifically relating to a battery safety monitoring method and system based on extended Kalman filtering to predict battery voltage-temperature. Background Technology

[0002] With the rapid development of new energy technologies, energy storage systems have been widely used in power systems, electric vehicles, and renewable energy grid connection. As the core component of an energy storage system, the operational safety of the battery directly affects the stability and reliability of the entire system. However, during charging and discharging, batteries may cause safety hazards due to overcharging, over-discharging, internal short circuits, thermal runaway, etc., even leading to serious accidents such as fires or explosions. Therefore, real-time and accurate monitoring of the battery's operating status is crucial.

[0003] Currently, battery safety monitoring primarily relies on threshold judgment methods for parameters such as voltage, current, and temperature. For example, when the battery voltage exceeds a set upper or lower threshold, or the temperature exceeds a safe range, the system will trigger an alarm or take protective measures. However, this type of method has the following limitations:

[0004] Lag: Monitoring methods based on fixed thresholds can usually only trigger warnings after a fault has occurred or is about to occur, making it difficult to detect potential hidden dangers early.

[0005] Insufficient sensitivity: Early abnormalities in the battery may only manifest as minor changes in voltage or temperature, and fixed thresholds are insufficient to detect such subtle abnormalities, leading to missed or false alarms.

[0006] Reliance on experience thresholds: Safety thresholds are often set based on historical data or experimental experience, which makes it difficult to adapt to dynamic factors such as battery aging and environmental changes, thus affecting the accuracy of monitoring.

[0007] In recent years, model-based battery state estimation methods have gradually attracted attention. For example, Kalman filtering (KF) and its improved algorithms (such as extended Kalman filtering, EKF) have been used for battery state estimation (e.g., SOC, SOH). However, existing research mainly focuses on estimating battery performance parameters, while how to utilize residual analysis between model predictions and measured data to improve the timeliness and accuracy of battery safety monitoring remains an urgent problem to be solved. Summary of the Invention

[0008] In view of this, the purpose of this invention is to solve the problems existing in the existing battery safety monitoring, and to propose a battery safety monitoring method and system based on extended Kalman filter prediction of battery voltage-temperature. It can predict the battery voltage-temperature through a battery voltage-temperature prediction model and combine it with residual analysis technology to achieve early warning of potential battery safety hazards, thereby improving the safety and reliability of energy storage systems.

[0009] To achieve the above objectives, the technical solution provided by the present invention is as follows:

[0010] According to a first aspect of the present invention, a battery safety monitoring method based on extended Kalman filtering to predict battery voltage-temperature is proposed, comprising the following steps:

[0011] S1: Real-time acquisition of voltage, current, and temperature measurements during battery operation via battery sensors;

[0012] S2: Construct a battery electrical model and a battery electrothermal model; the battery electrical model adopts a second-order RC equivalent circuit model to describe the dynamic characteristics of the battery terminal voltage; the battery electrothermal model describes the dynamic temperature change of the battery by combining the Joule heating effect and the first-order thermal balance equation.

[0013] S3: Based on the extended Kalman filter algorithm, a battery voltage-temperature prediction model is established by combining the battery electrical model and the battery electrothermal model;

[0014] S4: Generate battery voltage and battery temperature estimates using the battery voltage-temperature prediction model;

[0015] S5: Calculate the residual between the measured battery voltage and the estimated battery voltage, and the residual between the measured battery temperature and the estimated battery temperature;

[0016] S6: Compare the obtained residual with the preset safety threshold to determine the battery safety status;

[0017] S7: When the residual exceeds the safety threshold for a preset time, an alarm is triggered and corresponding safety measures are implemented.

[0018] Furthermore, the second-order RC equivalent circuit model includes an ohmic internal resistance connected in series, a first parallel RC network, and a second parallel RC network. The expression for the second-order RC equivalent circuit model is as follows:

[0019]

[0020] Where t is the battery running time, I is the battery output current, and U... C1 U is the electrochemical polarization voltage of the battery. C2R0 is the concentration polarization voltage of the battery, R1 and C1 are the electrochemical polarization resistance and capacitance of the battery, respectively, and R2 and C2 are the concentration polarization resistance and capacitance of the battery, respectively. oc U is the battery open-circuit voltage. L This refers to the battery terminal voltage. The rate of change of the electrochemical polarization voltage of the battery. This represents the rate of change of the concentration polarization voltage of the battery.

[0021] Furthermore, the expression for the battery electrothermal model is as follows:

[0022]

[0023] Where T is the battery surface temperature; t is the battery operating time; and I is the battery output current. T represents the rate of change of battery surface temperature; mc is the effective heat capacity of the battery, in J / ℃; R is the battery heating resistance; hA is the battery heat transfer coefficient, in W / ℃; mc, R, and hA are all related to the battery's state of charge (SOC), and are updated in real time using a lookup table method as the SOC changes; T ∞ This is the ambient temperature of the battery operation, with a default value of 25℃.

[0024] Furthermore, the process of establishing the battery voltage-temperature prediction model in step S3 is as follows:

[0025] By combining the parameters of the battery electrical model and the battery electrothermal model, an extended Kalman filter state-space equation is constructed, including the state equation and the observation equation.

[0026] The state equation is expressed as follows:

[0027]

[0028] The expression for the observation equation is:

[0029]

[0030] In the formula, Q is the maximum capacity of the battery; k is the estimated capacity of the battery at time k.

[0031] The state variables include the battery's state of charge (SOC), battery surface temperature (T), and battery electrochemical polarization voltage (U). C1 and the concentration polarization voltage U of the battery C2 The observed values ​​include the battery terminal voltage U. L and battery surface temperature T;

[0032] The extended Kalman filter algorithm is used to iteratively update the state variables, enabling synchronous prediction of voltage and temperature.

[0033] Furthermore, the method for determining the security threshold includes:

[0034] Statistically analyze the residual data of the battery under normal conditions, and calculate its expected value and variance of Gaussian distribution;

[0035] The interval is taken as the normal range of the residual.

[0036] Furthermore, the preset time is 2 seconds.

[0037] Furthermore, the safety measures in step S7 include:

[0038] Check the battery sensors and battery status;

[0039] Choose between repair or continue operation mode based on the inspection results.

[0040] Furthermore, the parameters of the battery electrical model and the battery electrothermal model are obtained by fitting the voltage and temperature data during battery operation using the least squares method.

[0041] Furthermore, the battery safety monitoring method based on extended Kalman filtering to predict battery voltage-temperature also includes:

[0042] The parameters of the battery voltage-temperature prediction model are updated in real time to adapt to changes in battery state. According to a second aspect of the invention, a battery safety monitoring system based on extended Kalman filtering for predicting battery voltage-temperature is proposed, comprising:

[0043] The sensor module is used to collect battery voltage, battery current and battery temperature data in real time;

[0044] The processing module is configured to execute the battery safety monitoring method based on extended Kalman filter prediction of battery voltage-temperature.

[0045] The alarm module is used to trigger an alarm when an anomaly is detected.

[0046] Compared with existing technologies, the present invention proposes a battery safety monitoring method and system based on extended Kalman filtering to predict battery voltage-temperature, which has the following advantages:

[0047] The battery safety monitoring method and system proposed in this invention, based on extended Kalman filtering for predicting battery voltage and temperature, first acquires real-time battery data, then estimates battery voltage and temperature using extended Kalman filtering based on battery electrical and electrothermal models. Finally, residuals are generated using the measured and estimated battery voltage values, and residuals are generated using the measured and estimated battery temperature values. These residuals are then compared with predefined safety thresholds. If abnormal parameters are detected, the battery sensors and battery status are checked promptly to prevent battery safety issues. Attached Figure Description

[0048] Figure 1 The flowchart shows a battery safety monitoring method based on extended Kalman filtering to predict battery voltage-temperature.

[0049] Figure 2 This is a structural diagram of a second-order RC model for a lithium-ion battery.

[0050] Figure 3 This is a structural diagram of the electrothermal model of a lithium-ion battery;

[0051] Figure 4 This is a probability density distribution curve of battery safety parameters;

[0052] Figure 5 The graph shows the battery voltage monitoring results for US06 under normal operating conditions at 25℃.

[0053] Figure 6 This is a diagram showing the safety status assessment of the battery voltage under normal operating conditions at 25℃ (US06).

[0054] Figure 7 The image shows the temperature monitoring results of the US06 battery under normal operating conditions at 25℃.

[0055] Figure 8 This is a diagram for assessing the temperature safety status of a US06 battery under normal operating conditions at 25℃.

[0056] Figure 9 This is a graph showing the temperature change of the battery under abnormal operating conditions (US06, 25℃).

[0057] Figure 10 Characteristic diagram of abnormal battery temperature interference signal;

[0058] Figure 11 The graph shows the battery temperature monitoring results under abnormal operating conditions of 25℃ (US06).

[0059] Figure 12 This is a diagram for determining the battery temperature safety status under the abnormal operating condition of 25℃ (US06).

[0060] Figure 13 A partial feature diagram of battery temperature monitoring under abnormal operating conditions (US06) at 25℃.

[0061] Figure 14 This is a graph showing the battery voltage change under abnormal operating conditions (US06) at 25℃.

[0062] Figure 15 Characteristic diagram of abnormal battery voltage interference signal;

[0063] Figure 16 The graph shows the battery voltage monitoring results under abnormal operating conditions of US06 at 25℃.

[0064] Figure 17This is a diagram showing the battery voltage safety status under the abnormal operating condition of US06 at 25℃.

[0065] Figure 18 This is a partial feature diagram of the US06 battery under abnormal operating conditions at 25℃. Detailed Implementation

[0066] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, this invention is not limited to the following embodiments, and specific implementation methods can be determined according to the technical solutions of this invention and actual circumstances. To avoid obscuring the essence of this invention, well-known methods, processes, and procedures are not described in detail.

[0067] This invention proposes a battery safety monitoring method based on extended Kalman filtering to predict battery voltage and temperature. It utilizes a battery voltage-temperature prediction model to obtain the residuals between battery voltage and temperature, which are then compared with safety thresholds for more efficient and timely detection of potential battery safety hazards. The method comprises three parts: battery data collection, battery voltage and temperature estimation, and battery safety status assessment. Figure 1 As shown, real-time battery data is first acquired by battery sensors (battery voltage sensor, battery current sensor, and battery temperature sensor). Then, based on the battery electrical model and battery electrothermal model, the extended Kalman filter is used to estimate the battery voltage and battery temperature. Finally, residuals are generated using the measured battery voltage value and the estimated battery voltage value, and residuals are generated using the measured battery temperature value and the estimated battery temperature value. After generating the residuals, they are compared with the set safety thresholds. If abnormal parameters are found, the battery sensors and battery status are checked in time to prevent battery safety problems.

[0068] like Figure 1 As shown, the battery safety monitoring method based on extended Kalman filtering to predict battery voltage-temperature includes the following steps:

[0069] S1: Based on the target battery configuration, the battery sensor collects the battery voltage, battery current and battery temperature in real time, and obtains the battery voltage measurement value, battery current measurement value and battery temperature measurement value; it should be noted that the battery sensor is existing technology, and will not be described in detail here.

[0070] S2: Determine the battery electrical model and battery electrothermal model. The battery electrical model adopts the equivalent circuit model of the battery, which is a second-order RC model. The second-order RC model is a commonly used equivalent circuit model for electrochemical energy storage systems (e.g., lithium batteries). The second-order RC model consists of the following topology: a circuit composed of an ohmic internal resistance (series resistance) and two parallel RC networks cascaded together. Specifically, the connection method is as follows: first, a pure resistive element (R0) representing the ohmic internal resistance is used as a series element, then the first parallel RC network (R1∥C1) is connected in series, and finally the second parallel RC network (R2∥C2) is connected in series. The overall topology is represented as: R0 + (R1∥C1) + (R2∥C2), and the structure is as follows. Figure 2 As shown, this model simulates the rapid polarization dynamics of the battery using a first RC network and characterizes the slow diffusion effect using a second RC network, thus establishing a voltage response characteristic coupled with multiple time constants. Compared to the traditional first-order RC model, the second-order topology used in this invention can simultaneously capture the fast and slow dynamic behaviors of the electrochemical system, achieving an optimal engineering balance between model complexity and estimation accuracy. In particular, this second-order RC model effectively solves the time-varying nonlinearity problem of voltage prediction in automotive and energy storage applications through its dual-time-constant design, providing battery management systems with full-state voltage observation capabilities that include ohmic voltage drop, polarization loss, and diffusion effects. This invention preferably applies this second-order RC model to the online state estimation of lithium batteries, achieving dynamic prediction and monitoring of the terminal voltage by real-time calculation of the relaxation process of the two RC networks.

[0071] The relationship between the voltage curve and the parameters of the second-order RC model is shown in Equation (1):

[0072]

[0073] Where t is the battery running time, I is the battery output current, and U... C1 for Figure 2 The C1 voltage, i.e., the electrochemical polarization voltage of the battery, U C2 for Figure 2 In the equation, C2 is the concentration polarization voltage of the battery, R0 is the ohmic internal resistance of the battery, R1 and C1 are the electrochemical polarization resistance and capacitance of the battery, respectively, and R2 and C2 are the concentration polarization resistance and capacitance of the battery, respectively. U oc U is the battery open-circuit voltage. L This refers to the battery terminal voltage. The rate of change of voltage C1 The rate of change of voltage C2.

[0074] To simulate the temperature rise of a battery during discharge, this invention combines a battery thermal model with a battery electrical model to form a battery electrothermal model, such as... Figure 3As shown in the figure, its parameters are correlated with the battery SOC (SOC is the state of charge of the battery, which refers to the percentage of the current remaining charge to the total battery capacity) to more accurately estimate the battery temperature. The battery electrothermal model is shown in formula (2). The battery electrothermal model adopted is based on a simple first-order thermal balance equation. By introducing the Joule heating term, it effectively describes the electro-thermal coupling relationship, and has good physical interpretability while maintaining low computational complexity. By modifying the heat capacity and heat dissipation coefficient parameters in the model, it can adapt to changes in different SOC, aging states and environmental conditions, and is suitable for embedding in the battery safety monitoring system for temperature prediction and supervision.

[0075]

[0076] Where T is the battery surface temperature; t is the battery operating time; and I is the battery output current. denoted as , where is the rate of change of battery surface temperature; mc is the effective heat capacity of the battery, in J / ℃; R is the battery heating resistance; hA is the battery heat transfer coefficient, in W / ℃; mc, R, and hA are all related to the change in the battery's state of charge (SOC); T ∞ This is the battery operating ambient temperature, with a default value of 25℃.

[0077] S3: Establish a battery voltage-temperature prediction model

[0078] The Kalman filter algorithm is a recursive filtering solution specifically designed for discrete system problems. Essentially, the Kalman filter algorithm is a series of mathematical equations used for computation, providing an effective recursive solution for the least squares method. The Kalman filter algorithm has many functions and advantages; even when the modeling system is uncertain, it can still calculate the system's state variables, including changes in state values ​​that have occurred, are occurring, and will occur. Equation (3) represents a series of equations obtained after discretizing a continuous model.

[0079]

[0080] Where x k It is a state variable, x k+1 It is the state variable to be obtained next time, u k Input variable, y k It is an output variable, A k It is the input matrix, B k It is the control input matrix, C k It is the observation matrix, D k It is the output value transfer matrix, ω k It is process noise, v k This measures noise. ω k and v kIt satisfies a normal distribution and consists of mutually independent Gaussian white noise, as shown in Equation (4).

[0081]

[0082] in For ω k and The mathematical expectation of the product, It is the transpose of process noise, Q ω It is ω k The covariance matrix, For v k and The mathematical expectation of the product, It is the transpose of the noise measurement, R v It is v k The covariance matrix, Cov(ω) k ,v k ) is ω k With v k The covariance.

[0083] Formula (5) is the detailed calculation formula for the Kalman filter algorithm, where These are prior estimates of the state variables. These are the posterior estimates of the state variables. It is the posterior estimate of the state variable from the previous iteration. It is the prior estimate of the error covariance matrix. It is the posterior estimate of the error covariance matrix. It is the posterior estimate of the previous error covariance matrix, Kg k I is the Kalman filter gain, and I0 is the identity matrix. It is the transpose of the input matrix. It is the transpose of the observation matrix.

[0084]

[0085] The extended Kalman filter is a Kalman filter that linearizes the current mean and covariance. For nonlinear systems, this invention uses the partial derivatives of the state equation and the observation equation to linearize the current estimate, that is, it is necessary to use Taylor series expansion to approximate the calculation formula. Formula (6) is its linear discrete state equation. Substituting it into formula (5) yields formula (7), which is the extended Kalman filter.

[0086]

[0087] Where f(x) k ,u k ) is the state transition function, g(x) k ,uk ) is a measurement function. It is the state estimate at time k, x k+1 It is the system state variable at time k+1. It is the state estimate at time k based on the information at time k. It is the state estimate at time k based on the information at time k-1. H is the state estimate at time k-1 based on information from time k-1. k It is the input matrix at time k-1. It is the transpose of the input matrix at time k-1. It is the state transition function based on the state estimate at time k. It is the state transition function based on the state estimate at time k-1. It is a measurement function based on the state estimate at time k-1.

[0088] To accurately estimate battery voltage and temperature, this invention uses an extended Kalman filter (EKF) combined with battery model parameters to establish a battery voltage-temperature prediction model. The battery model parameters are obtained through least squares fitting experiments on the voltage and temperature during battery operation. Combining the second-order RC model voltage formula (1) with the electrothermal model temperature formula (2) yields the state equation and observation equation of the EKF, as shown in formulas (8) and (9).

[0089]

[0090] in, It represents the relationship between open-circuit voltage and battery SOC, k is the estimated value of the battery at time k, and Q is the maximum capacity of the battery.

[0091] The residuals generated by comparing the estimated battery voltage and temperature values ​​obtained from the battery electrical and electrothermal models with the measured battery voltage and temperature values ​​are used as battery safety monitoring parameters. The safety threshold for determining whether a fault has occurred is calculated using statistical methods, as follows.

[0092] The calculation of safety thresholds for safety monitoring parameters requires a statistical evaluation of the residuals between predicted and measured values ​​under fault-free conditions. Since the residuals between predicted and measured values ​​include estimation errors and random errors in system measurements, they can be determined to conform to a Gaussian distribution X~N(μ,σ). 2 As shown in formula (10), the probability distribution is as follows: Figure 4 .

[0093]

[0094] Where X is a random variable that follows a normal distribution, x is the value of the random variable X, μ is the mean (expected value), which determines the central location of the distribution, and σ is the standard deviation, which measures the dispersion of the data. 2 (where σ is the variance). By statistically analyzing the residual data of batteries under normal conditions, we can calculate their Gaussian distribution expectation μ and variance σ. 2 The probability distribution is shown in Table 1. Since there are brief jumps during battery discharge that affect the accuracy of the estimation, (μ-5σ, μ+5σ) is defined as the normal range for its residuals. Residuals exceeding this range and remaining for more than 2 seconds are considered abnormal.

[0095] Table 1. Probability of Gaussian Distribution Curve Interval

[0096]

[0097] A battery safety monitoring system based on extended Kalman filtering to predict battery voltage-temperature includes:

[0098] The sensor module is used to collect battery voltage, battery current and battery temperature data in real time;

[0099] The processing module is configured to execute the battery safety monitoring method based on extended Kalman filter prediction of battery voltage-temperature.

[0100] The alarm module is used to trigger an alarm when an anomaly is detected.

[0101] Specific application examples

[0102] Using data from a 2.9Ah Panasonic 18650PF battery at the University of Wisconsin-Madison, we first established battery electrical and electrothermal models. Then, we performed extended Kalman filter estimation on the battery temperature and voltage under the US06 operating condition at 25°C (US06 is one of the fuel economy test conditions established by the US Environmental Protection Agency (EPA) and the Society of Automotive Engineers (SAE), belonging to the Supplemental Federal Test Procedure (SFTP), used to simulate the impact of aggressive driving behavior on vehicle fuel consumption and emissions). The variance and expected value were calculated, and the results are shown in Table 2. Since the battery was in normal operating condition at this time, the calculated data were used as the battery safety threshold. Figure 5 and Figure 6 Battery voltage safety monitoring corresponding to 25℃ US06 operating condition. Figure 7 and Figure 8 Battery temperature safety monitoring corresponding to the 25℃ US06 operating condition. Figure 5 The following graph shows the battery voltage monitoring results under normal operating conditions at 25℃ for US06. Figure 6 This diagram shows the battery voltage safety status assessment under normal operating conditions at 25℃ (US06). Figure 7 The graph shows the battery temperature monitoring results under normal operating conditions at 25℃ for US06. Figure 8 The diagram shows the battery temperature safety status assessment under normal operating conditions at 25℃ (US06). Figure 5 , Figure 6 , Figure 7 and Figure 8 This includes estimates of battery voltage and temperature under normal conditions, as well as assessments of battery safety.

[0103] Table 2. Probability of Gaussian Distribution Curve Interval

[0104]

[0105] After obtaining the threshold parameters of the battery safety monitoring system, in order to verify the reliability of the battery safety monitoring system, interference was introduced into the voltage and temperature under the 25℃ US06 operating condition. Figure 9 and Figure 10 At t = 4010.75s, a 500s interference with an amplitude of 1℃ was added to the temperature, and through... Figure 11 , Figure 12 and Figure 13 As can be seen, the system can promptly identify temperature anomalies and issue timely safety warnings. Figure 14 and Figure 15 At t = 4010.75s, an interference with an amplitude of 1V for 500s was applied to the voltage, and through... Figure 16 , Figure 17 and Figure 18 As can be seen, the system can promptly identify voltage anomalies and issue timely safety warnings.

Claims

1. A battery safety monitoring method for predicting battery voltage-temperature based on extended Kalman filter, characterized in that, The method comprises the following steps: S1: collecting voltage, current and temperature measurement values of the battery in real time through a battery sensor; S2: constructing a battery electrical model and a battery electrical-thermal model; the battery electrical model adopts a second-order RC equivalent circuit model to describe the dynamic characteristics of the battery terminal voltage; the battery electrical-thermal model describes the dynamic change of the battery temperature by combining the Joule heat effect and a first-order thermal balance equation; S3: establishing a battery voltage-temperature prediction model based on an extended Kalman filtering algorithm and in combination with the battery electrical model and the battery electrical-thermal model; S4: generating a battery voltage estimation value and a battery temperature estimation value by using the battery voltage-temperature prediction model; S5: calculating the residual error of the battery voltage measurement value and the battery voltage estimation value, and the residual error of the battery temperature measurement value and the battery temperature estimation value; S6: comparing the obtained residual error with a preset safety threshold to determine the battery safety state; S7: when the residual error exceeds the safety threshold for a preset time, triggering an alarm and performing corresponding safety measures. 2.The battery safety monitoring method of predicting battery voltage-temperature based on extended Kalman filter according to claim 1, characterized in that, The second-order RC equivalent circuit model comprises an ohmic internal resistance, a first parallel RC network and a second parallel RC network connected in series, and the expression of the second-order RC equivalent circuit model is: where t is the battery operation time, I is the battery output current, U C1 is the battery's electrochemical polarization voltage, U C2 is the battery's concentration polarization voltage, R0 is the battery's ohmic internal resistance, R1 and C1 are the battery's electrochemical polarization resistance and capacitance, respectively, R2 and C2 are the battery's concentration polarization resistance and capacitance, respectively, U oc is the battery's open-circuit voltage, U L is the battery's terminal voltage, is the rate of change of the battery's electrochemical polarization voltage, is the rate of change of the battery's concentration polarization voltage. 3.The battery safety monitoring method of predicting battery voltage-temperature based on extended Kalman filter according to claim 2, characterized in that, The expression of the battery electrical-thermal model is: Wherein, T is the battery surface temperature; t is the battery running time; I is the battery output current; is the battery surface temperature change rate; mc is the effective heat capacity of the battery, unit J / ℃; R is the battery heating resistance; hA is the battery heat transfer coefficient, unit W / ℃; mc, R and hA are all related to the state of charge SOC of the battery, which are updated in real time by table lookup method with the change of the state of charge SOC of the battery; T ∞ is the battery running environment temperature, which is taken as 25℃ by default.

4. The battery safety monitoring method of predicting battery voltage-temperature based on extended Kalman filter according to claim 3, characterized in that, The process of establishing the battery voltage-temperature prediction model in step S3 is: constructing the state space equation of the extended Kalman filtering in combination with the parameters of the battery electrical model and the battery electrical-thermal model, including a state equation and an observation equation; The expression of the state equation is: The expression of the observation equation is: In the formula, Q is the maximum capacity of the battery; k is the operation estimation at time k of the battery; wherein the state variables comprise the state of charge SOC of the battery, the battery surface temperature T, the electrochemical polarization voltage U of the battery C1 and the concentration polarization voltage U of the battery C2 , and the observation values comprise the battery terminal voltage U L and the battery surface temperature T; iteratively updating the state quantity based on the extended Kalman filtering algorithm to realize the synchronous prediction of the voltage and the temperature.

5. The method of claim 1, wherein the battery safety monitoring method based on the extended Kalman filter predicting the battery voltage-temperature is characterized by, The determination method of the safety threshold comprises: statistically analyzing the residual error data under the normal state of the battery to calculate the mathematical expectation and the variance of the Gaussian distribution; regarding the interval as the normal range of the residual error.

6. The method of claim 1, wherein the battery safety monitoring method based on the extended Kalman filter predicting the battery voltage-temperature is characterized by, The preset time is 2 seconds.

7. The method of claim 1, wherein the battery safety monitoring method based on extended Kalman filter prediction of battery voltage-temperature is characterized by, The safety measures in step S7 comprise: checking the battery sensor and the battery state; selecting the repair or the continuous operation mode according to the checking result.

8. The battery safety monitoring method of predicting battery voltage-temperature based on extended Kalman filter according to claim 4, characterized in that, The parameters of the battery electrical model and the battery electrical-thermal model are obtained by parameter fitting of the voltage and temperature data in the battery operation process through the least square method.

9. The method of claim 1, wherein the battery safety monitoring method based on extended Kalman filter prediction of battery voltage-temperature is characterized by, The method further comprises: updating the parameters of the battery voltage-temperature prediction model in real time to adapt to the change of the battery state.

10. A battery safety monitoring system for predicting battery voltage-temperature based on extended Kalman filter, characterized in that, It comprises: a sensor module for collecting battery voltage, battery current and battery temperature data in real time; a processing module configured to perform the battery safety monitoring method for predicting the battery voltage-temperature based on the extended Kalman filtering according to any one of claims 1-9; an alarm module for triggering an alarm when an abnormality is detected.

Images