CN-121973630-A - Control method for multi-mode cooperative lithium battery protection circuit

CN121973630ACN 121973630 ACN121973630 ACN 121973630ACN-121973630-A

Abstract

The application provides a control method of a multi-mode cooperative lithium battery protection circuit. The method comprises the steps of collecting battery pack voltage, current, surface temperature and relay state signals in real time, constructing a standardized time sequence tensor through preprocessing, inverting the core temperature of each single battery based on an electric-thermal-aging coupling model, combining a joint extended Kalman filter with the real-time core temperature, dynamically calibrating an aging rate index, constructing a dynamic adjacency matrix through voltage correlation and temperature gradient, utilizing a lightweight graph convolution network to identify topology weakness risks, dynamically switching a protection mode in a multi-mode logic state machine according to the core temperature, the aging index, risk scores and SOC differences, generating a control instruction, finally adjusting the time sequence of a power tube through PWM phase compensation, and driving an equalization circuit through a distributed model predictive control algorithm to realize the protection of a lithium battery system with safety, service life and performance collaborative optimization.

Inventors

HE HUI
DONG LIANG
ZHANG SHENGBING

Assignees

深圳市山河动力电子有限公司

Dates

Publication Date: 20260505
Application Date: 20260211

Claims (10)

1. A method for controlling a multi-mode coordinated lithium battery protection circuit, the method comprising: terminal voltage of each single battery in lithium battery pack is collected in real time Total current of power loop Array of monomer surface temperatures The relay contact state signal is denoised to construct a standardized time sequence tensor; Based on a one-dimensional electric-thermal-aging coupling model, calculating the core temperature of each single battery by utilizing the normalized time sequence tensor inversion ; Estimating the state of charge (SOC) and the state of health (SOH) of the single battery by adopting a joint extended Kalman filtering algorithm, and combining the core temperature Dynamically calibrating aging rate index ; Constructing a dynamic adjacency matrix based on the voltage correlation and the core temperature gradient among the single batteries, and identifying the topological vulnerability risk score of the battery pack by utilizing a lightweight graph convolutional neural network; Operating a multi-mode collaborative logic state machine according to the core temperature The ageing rate index The topological vulnerability risk score and the state of charge (SOC) difference degree among the single batteries are dynamically switched among a very fast safety mode, a life optimization mode, a cooperative balance mode and a conventional monitoring mode, and a control instruction set is output; And according to the control instruction set, the conduction time sequence of the multiphase parallel power tubes in the protection circuit is regulated through PWM phase compensation, and a distributed model predictive control algorithm is adopted to drive the equalization circuit to execute energy transfer.
2. The method for controlling a multi-mode cooperative lithium battery protection circuit according to claim 1, characterized in that the core temperature The following formula is adopted for calculation: ; Wherein, the For the rate of heat generation of the unit cell, In order to sample the period of time, In order to achieve an equivalent heat capacity, Is the internal thermal resistance of the single battery, In the event of a loop current, Is the ohmic internal resistance, Is an entropy change coefficient.
3. The method for controlling a multi-mode cooperative lithium battery protection circuit according to claim 1, characterized in that the aging rate index The following formula is adopted for calculation: ; Wherein, the As a factor of the reference ageing rate, In order to be a dynamic gain factor, In order to provide a rate of change of the terminal voltage, In order to be able to activate the electrochemical, As a result of the ideal gas constant, Is the core temperature.
4. The method for controlling a multi-mode cooperative lithium battery protection circuit according to claim 3, characterized in that the method further comprises: Using the aging rate index Linear contraction of the continuous current protection threshold of the power loop, if the ageing rate is exponential If the current exceeds the preset aging threshold, the current is protected to be the current protection threshold Multiplying by the attenuation coefficient.
5. The method for controlling a multi-mode coordinated lithium battery protection circuit according to claim 1, wherein the constructing a dynamic adjacency matrix comprises: calculating the edge weights among the nodes; feature aggregation is carried out on the specific neighborhood with the maximum edge weight through a pooling layer of the graph neural network, and a topological weak point risk score of thermal-electric coupling unbalance in the battery pack is output; the edge weights among the nodes are calculated in the following mode: ; Wherein, the Representing the edge weights between the nodes, Is the first Monomers of the first kind The terminal voltage correlation coefficient between the individual cells, The coefficients are adjusted for the weights.
6. The method for controlling a lithium battery protection circuit according to claim 1, wherein the multimode cooperative logic state machine determines to perform mode switching by: If it is Or single body terminal voltage Activating the high-speed safety mode and sending a power loop turn-off instruction; If the aging rate index is The service life optimization mode is activated when the preset aging threshold is exceeded, the charging multiplying power is limited, and a load transfer instruction of the main relay and the standby relay is issued; If between single batteries Activating the cooperative balance mode and starting an equalization circuit; otherwise, the regular monitoring mode is maintained.
7. The method for controlling a multi-mode cooperative lithium battery protection circuit according to claim 1, wherein the adjusting the conduction time sequence of the multiphase parallel power tubes in the protection circuit by PWM phase compensation comprises: aiming at the power switch tubes connected in parallel in multiple ways, the conduction time intervals of the power switch tubes in each way are controlled to be distributed according to an arithmetic progression according to the charging current amplitude command.
8. The control method of the multi-mode collaborative lithium battery protection circuit according to claim 1, wherein the distributed model predictive control algorithm iteratively solves the optimal balance current of each single battery through an alternate direction multiplier method ADMM To drive the equalization circuit to perform energy transfer; wherein the optimal balance current The solution formula of (2) is as follows: ; Wherein, the As the average state of charge of the battery pack, As a value of the average temperature of the material, Is a thermal stress penalty factor for minimizing system thermal losses during energy transfer.
9. The method for controlling a lithium battery protection circuit in a multi-mode coordination manner according to claim 8, wherein the distributed model predictive control algorithm adopts an event trigger mechanism; And when the terminal voltage change rate absolute value or the SOC difference value of the single battery exceeds a preset trigger threshold, activating the ADMM iteration solving process, otherwise, continuing to use the optimal balanced current of the previous period.
10. The method for controlling a multi-mode cooperative lithium battery protection circuit according to claim 1, characterized in that the method further comprises: monitoring the core temperature drop slope after executing the hardware action in real time ; And if the core temperature falling slope is lower than 90% of the model prediction slope, judging that the efficiency of the trigger actuator deviates, generating a risk penalty factor and feeding back the risk penalty factor.

Description

Control method for multi-mode cooperative lithium battery protection circuit Technical Field The application relates to the technical field of power battery management systems of new energy automobiles, in particular to a control method of a multi-mode collaborative lithium battery protection circuit. Background With the rapid development of new energy automobiles, lithium ion power batteries have become the main stream energy storage devices of electric automobiles due to the advantages of high energy density, long cycle life and the like. The Battery Management System (BMS) is used as a core component for guaranteeing the safety of the battery and improving the efficiency of the battery, and the reliability, the instantaneity and the accuracy of a protection circuit of the battery management system are directly related to the safety performance and the service life of the whole vehicle. Currently, the mainstream lithium battery protection circuit mostly adopts a passive protection strategy based on a fixed threshold value, such as hardware triggering protection of overvoltage, undervoltage, overcurrent, over-temperature and the like. In terms of state estimation, most systems rely on Equivalent Circuit Models (ECMs) in combination with Extended Kalman Filtering (EKF) for SOC/SOH estimation, and in terms of thermal management, monitoring is mainly performed by temperature sensors arranged on the battery surface. In addition, some advanced systems attempt to introduce Graph Neural Networks (GNNs) for battery pack inconsistency analysis or to employ Model Predictive Control (MPC) for equalization optimization. However, these schemes have problems such as insufficient sensing precision, poor model adaptability, uneven calculation force distribution, single protection mode, and the like, and cannot meet the actual demands. Therefore, the application provides a control method of a multi-mode collaborative lithium battery protection circuit, so as to solve one of the technical problems. Disclosure of Invention The application aims to provide a control method of a multi-mode collaborative lithium battery protection circuit, which can solve at least one technical problem. The specific scheme is as follows: according to a first aspect of the present application, there is provided a control method for a multi-mode cooperative lithium battery protection circuit, including: terminal voltage of each single battery in lithium battery pack is collected in real time Total current of power loopArray of monomer surface temperaturesThe relay contact state signal is denoised to construct a standardized time sequence tensor, and the core temperature of each single battery is calculated by inversion of the standardized time sequence tensor based on a one-dimensional electric-thermal-aging coupling modelEstimating the state of charge (SOC) and the state of health (SOH) of the single battery by adopting a joint extended Kalman filtering algorithm, and combining the core temperatureDynamically calibrating aging rate indexConstructing a dynamic adjacency matrix based on the voltage correlation among single batteries and the core temperature gradient, identifying the topological vulnerability risk score of the battery pack by utilizing a lightweight graph convolutional neural network, running a multi-mode collaborative logic state machine, and according to the core temperatureThe ageing rate indexAnd according to the control instruction set, the conduction time sequence of the multiphase parallel power tubes in the protection circuit is regulated through PWM phase compensation, and a distributed model predictive control algorithm is adopted to drive the equalization circuit to execute energy transfer. In one embodiment, the core temperatureThe following formula is adopted for calculation: Wherein, the method comprises the steps of, For the rate of heat generation of the unit cell,In order to sample the period of time,In order to achieve an equivalent heat capacity,Is the internal thermal resistance of the single battery,In the event of a loop current,Is the ohmic internal resistance,Is an entropy change coefficient. In one embodiment, the aging rate indexThe following formula is adopted for calculation: Wherein, the method comprises the steps of, As a factor of the reference ageing rate,In order to be a dynamic gain factor,In order to provide a rate of change of the terminal voltage,In order to be able to activate the electrochemical,As a result of the ideal gas constant,Is the core temperature. In one embodiment, the method further comprises utilizing the aging rate indexLinear contraction of the continuous current protection threshold of the power loop, if the ageing rate is exponentialIf the current exceeds the preset aging threshold, the current is protected to be the current protection thresholdMultiplying by the attenuation coefficient. In one embodiment, the construction of the dynamic adjacency matrix comprises the steps of