Battery fault diagnosis and automatic control system based on RUL prediction and hierarchical self-healing

Abstract

The invention belongs to the technical field of operation, maintenance and intelligent control of Internet of Things equipment, and discloses a battery fault diagnosis and automatic control system based on RUL prediction and hierarchical self-healing, which comprises the following functional modules and a coordination mechanism thereof: a full-dimension monitoring module, which is used for collecting battery operation parameters of not less than 20 dimensions in real time; based on RUL prediction of a multi-factor mixed attenuation model and isolated forest multi-dimensional anomaly diagnosis, early warning and accurate positioning of faults are realized. Meanwhile, the hierarchical self-healing mechanism can automatically execute an optimal control strategy according to the fault level, changes passive alarm into active protection, effectively restrains fault spreading, prevents serious accidents such as thermal runaway, prolongs the service life of the battery pack, reduces non-planned shutdown and manual intervention, and reduces the operation and maintenance cost of the whole life cycle. In combination with closed-loop feedback, system parameters and strategies are continuously optimized, and an active battery health management closed loop integrating monitoring, prediction, diagnosis, self-healing and feedback is integrally constructed.

Description

Technical Field

[0001] This invention belongs to the field of IoT device operation and maintenance and intelligent control technology, specifically a battery fault diagnosis and automatic control system based on RUL prediction and hierarchical self-healing. Background Technology

[0002] Battery failure refers to an abnormal situation in which, during the entire life cycle of a battery, such as charging and discharging cycles and static storage, the core operating parameters of the battery, such as voltage, temperature, internal resistance, and decay rate, deviate from the normal baseline range due to factors such as inconsistent performance degradation of individual battery cells, internal short circuits or open circuits, poor contact of external circuits, precursors to thermal runaway, and drastic fluctuations in environmental conditions. This affects the battery's energy output efficiency, cycle life, and may even lead to safety risks.

[0003] Current energy storage battery management systems generally suffer from limited monitoring dimensions, delayed fault response, and a lack of proactive self-healing capabilities. Traditional BMS typically only monitors basic parameters such as voltage, current, and temperature, with fault detection delays lasting up to several hours. Moreover, they mostly employ passive alarm methods, failing to achieve early warning and automatic repair of faults. While high-end BMS possess some health status prediction capabilities, they still lack multi-dimensional anomaly diagnosis and hierarchical self-healing mechanisms. Therefore, there is an urgent need for a closed-loop control system capable of early warning, intelligent diagnosis, and automatic self-healing to improve the reliability and economy of energy storage systems. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing, comprising the following functional modules and their coordination mechanisms: Full-dimensional monitoring module: used to collect battery operating parameters in no less than 20 dimensions in real time; RUL prediction algorithm module: Based on the hybrid degradation model, it uses the SOH data and operating parameters of the past 3 months to fit the degradation coefficient, calculate the remaining service life RUL, and trigger green, yellow and red three-level warnings based on the RUL value. At the same time, it calculates the battery degradation rate based on the SOH data, outputs the real-time value of the degradation rate and compares it with the preset degradation rate threshold. Battery module anomaly diagnosis module: Based on the isolated forest algorithm, it performs anomaly detection on 20+ dimension real-time running data, outputs anomaly scores and makes anomaly judgments, and classifies faults into first-level, second-level, and third-level categories according to the anomaly duration, RUL value, and number of anomaly dimensions. Battery module self-healing control module: Automatically executes graded self-healing strategies according to fault level. For level 1 abnormalities, active balancing control is executed; for level 2 faults, power reduction operation control is executed; and for level 3 faults, topology reconstruction isolation control is executed. Closed-loop feedback mechanism: The results of each diagnosis and self-healing are fed back to the RUL prediction and anomaly diagnosis module to achieve adaptive optimization.

[0006] Preferably, the 20-dimensional battery operating parameters in the full-dimensional monitoring module include: total voltage, total current, SOC, SOH, single cell voltage, maximum voltage, minimum voltage, voltage difference, battery pack temperature, average temperature, temperature difference, internal resistance, internal resistance rate, charge / discharge cycle count, and SOH decay rate.

[0007] Preferably, the formula for calculating the SOH data in the past 3 months and the operating parameters is as follows: In the formula, The battery health status at time t. Let be the actual discharge capacity of the battery at time t. This refers to the initial rated capacity of the battery.

[0008] Preferably, the hybrid decay model integrates three major decay factors: cycle, calendar, and temperature, which improves prediction accuracy compared to a single linear model. The hybrid decay model is as follows: In the formula, This is the initial health constant; , The attenuation coefficient; This is the temperature coupling coefficient; The cumulative total number of charge-discharge cycles at time t; For calendar lifespan; This is a temperature compensation function, when Temp ≤ 25℃. =0, when Temp>25℃ =(Temp-25).

[0009] Preferably, the attenuation coefficient , The value is automatically matched based on the battery type and real-time ambient temperature, and the specific value is: When the battery type is LFP and the ambient temperature is 15-25℃ =0.08% / 100cycles =1.5% / year; When the battery type is LFP and the ambient temperature is ≥25℃ =0.10% / 100cycle, =3.5% / year; When the battery type is ternary lithium battery =0.12% / 100cycle, =2.5% / year.

[0010] Preferably, the RUL prediction formula in the RUL prediction algorithm module is: In the formula, Let t be the remaining service life at time t. The linear decay rate of SOH over the past 90 days.

[0011] Preferably, the grading criteria for the green, yellow, and red three-level early warnings are as follows: Green alert: RUL > 60 days, decay rate < 0.3% / month; Yellow alert: RUL∈[30,60] days, decay rate∈[0.3,0.5]% / month; Red alert: RUL < 30 days, decay rate > 0.5% / month.

[0012] Preferably, the degradation rate threshold preset by the RUL prediction algorithm module is dynamically adjusted according to the temperature fluctuation conditions of the battery's operating environment. The temperature fluctuation conditions and the corresponding threshold values ​​are as follows: Constant temperature operation (temperature fluctuation <5℃): the attenuation rate threshold is 0.5% / month; Fluctuating operating conditions (temperature fluctuation of 5-10℃): the attenuation rate threshold is 0.3% / month; Extreme operating conditions (temperature fluctuations >10℃ or sustained high temperatures): the attenuation rate threshold is 0.2% / month.

[0013] Preferably, the formula for calculating the anomaly score is: In the formula, For anomaly scoring, the value range is [0-1], where 0 is completely normal and 1 is completely abnormal; X is an input feature vector composed of 20 or more dimensions of battery operating parameters, and the parameters of each dimension are normalized to the interval [0,1]. Let X be the path length of X in a single decision tree within an isolated forest; Let X be the mean path length of X across multiple decision trees in an isolated forest; This is the average path length constant of the decision tree when the sample size is n; The anomaly determination rule is: when When the value is >0.6, the battery is considered to be in an abnormal operating state; when it is <0.4... When the value is ≤0.6, the battery operating state is determined to be abnormal; when... When the value is ≤0.4, the battery is considered to be in normal operating condition.

[0014] Preferably, the specific rules for anomaly detection in the battery module anomaly diagnosis module include setting quantitative judgment thresholds for the following four core parameter dimensions based on 20 or more operating parameters collected by the full-dimensional monitoring module: Individual cell voltage difference: If the voltage of any individual cell deviates from the average voltage of the cells in the battery pack by more than 0.3V, it is considered an abnormal voltage, which may indicate inconsistent cell attenuation or circuit disconnection. Temperature difference: If the difference between the highest and lowest temperatures inside the battery pack is greater than 10°C, it is considered an abnormal temperature and may indicate an impending thermal runaway. Sudden increase in internal resistance: If the increase in the internal resistance value of the current monitoring cycle is greater than 5% compared with the average internal resistance value of the previous monitoring cycle, it is judged as an abnormal internal resistance, which may indicate poor contact or internal short circuit fault. Degradation rate: If the battery degradation rate exceeds the preset degradation rate threshold of 0.5% / month under the corresponding operating conditions, it is judged as abnormal degradation and there may be an accelerated degradation fault.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: Based on the multi-factor hybrid decay model for RUL prediction and the isolated forest multi-dimensional anomaly diagnosis, early warning and precise location of faults are achieved. At the same time, the hierarchical self-healing mechanism can automatically execute the optimal control strategy according to the fault level, transforming passive alarms into active protection, effectively curbing fault propagation, preventing serious accidents such as thermal runaway, thereby extending battery pack life, reducing unplanned downtime and manual intervention, and lowering the total life cycle maintenance cost. Combined with closed-loop feedback, the system parameters and strategies are continuously optimized to adapt to complex operating conditions, and an active battery health management closed loop integrating "monitoring-prediction-diagnosis-self-healing-feedback" is constructed. Detailed Implementation

[0016] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] This invention provides a battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing, including the following functional modules and their coordination mechanisms: Full-dimensional monitoring module: used to collect battery operating parameters in no less than 20 dimensions in real time; RUL prediction algorithm module: Based on the hybrid degradation model, it uses the SOH data and operating parameters of the past 3 months to fit the degradation coefficient, calculate the remaining service life RUL, and trigger green, yellow and red three-level warnings based on the RUL value. At the same time, it calculates the battery degradation rate based on the SOH data, outputs the real-time value of the degradation rate and compares it with the preset degradation rate threshold. Battery module anomaly diagnosis module: Based on the isolated forest algorithm, it performs anomaly detection on 20+ dimension real-time running data, outputs anomaly scores and makes anomaly judgments, and classifies faults into first-level, second-level, and third-level categories according to the anomaly duration, RUL value, and number of anomaly dimensions. Battery module self-healing control module: Automatically executes graded self-healing strategies according to fault level. For level 1 abnormalities, active balancing control is executed; for level 2 faults, power reduction operation control is executed; and for level 3 faults, topology reconstruction isolation control is executed. Closed-loop feedback mechanism: The results of each diagnosis and self-healing are fed back to the RUL prediction and anomaly diagnosis module to achieve adaptive optimization.

[0018] A closed-loop technology system integrating monitoring, prediction, diagnosis, control, and feedback is constructed to achieve precise prevention and control of battery faults and extend battery life. Comprehensive parameter acquisition provides ample data support for fault diagnosis, while the hybrid degradation model RUL predicts and provides early warnings of performance degradation risks. The isolated forest algorithm enables accurate identification of multi-dimensional anomalies, and a hierarchical self-healing strategy addresses faults of varying degrees, avoiding excessive intervention or insufficient control. The closed-loop feedback mechanism continuously optimizes model parameters, enhancing the adaptive capabilities of diagnosis and control, and ensuring the safe and stable operation of the battery system.

[0019] The 20-dimensional battery operating parameters in the full-dimensional monitoring module include: total voltage, total current, SOC, SOH, single cell voltage, maximum voltage, minimum voltage, voltage difference, battery pack temperature, average temperature, temperature difference, internal resistance, internal resistance rate, charge / discharge cycle count, and SOH decay rate.

[0020] The advantage of this design is that it covers all dimensions of battery electrical, performance, and degradation status, providing comprehensive data support for fault diagnosis and RUL prediction. Electrical parameters such as total voltage and current reflect real-time operating conditions, while individual cell voltage difference and temperature difference capture local anomalies. SOC, SOH, and degradation rate quantify performance degradation trends. The complementary verification of multi-dimensional parameters avoids the limitations of single-dimensional monitoring and significantly improves the accuracy of anomaly detection and the reliability of RUL prediction.

[0021] The formula for calculating SOH data in the past 3 months and operating parameters is as follows: In the formula, The battery health status at time t. Let be the actual discharge capacity of the battery at time t. This refers to the initial rated capacity of the battery.

[0022] By unifying the SOH calculation standard through a quantitative formula, a precise and traceable performance degradation benchmark is provided for RUL prediction. At the same time, by selecting data from the past 3 months, the phased degradation trend can be accurately captured, providing reliable data support for the fitting coefficient of the hybrid degradation model, thereby improving the accuracy of RUL prediction and the timeliness of fault warning.

[0023] The hybrid decay model integrates three major decay factors: cycle, calendar, and temperature. Compared with the single linear model, it improves prediction accuracy. The hybrid decay model is as follows: In the formula, This is the initial health constant; , The attenuation coefficient; This is the temperature coupling coefficient; The cumulative total number of charge-discharge cycles at time t; For calendar lifespan; This is a temperature compensation function, when Temp ≤ 25℃. =0, when Temp>25℃ =(Temp-25).

[0024] By integrating multi-dimensional decay factors, an accurate prediction model is constructed, overcoming the limitations of a single linear model.

[0025] Among them, attenuation coefficient , The value is automatically matched based on the battery type and real-time ambient temperature, and the specific value is: When the battery type is LFP and the ambient temperature is 15-25℃ =0.08% / 100cycles =1.5% / year; When the battery type is LFP and the ambient temperature is ≥25℃ =0.10% / 100cycle, =3.5% / year; When the battery type is ternary lithium battery =0.12% / 100cycle, =2.5% / year.

[0026] The coefficient values ​​are set according to battery type and ambient temperature, which fits the degradation characteristics and temperature sensitivity of different batteries and avoids the prediction bias of fixed coefficients. The system can automatically match parameters according to the real-time monitored battery type and ambient temperature without manual intervention and adapt to complex working conditions. The differentiated coefficients make the output of the hybrid degradation model more in line with the actual aging law of the battery, providing more reliable data support for fault classification early warning and self-healing control.

[0027] The RUL prediction formula in the RUL prediction algorithm module is as follows: In the formula, Let t be the remaining service life at time t. The linear decay rate of SOH over the past 90 days.

[0028] The remaining service life is calculated based on the linear decay pattern of the past 90 days. The data is timely and can accurately capture the phased decay trend.

[0029] The grading standards for green, yellow, and red alerts are as follows: Green alert: RUL > 60 days, decay rate < 0.3% / month; Yellow alert: RUL∈[30,60] days, decay rate∈[0.3,0.5]% / month; Red alert: RUL < 30 days, decay rate > 0.5% / month.

[0030] Using RUL (Remaining Lifetime) and degradation rate as criteria, the risk of misjudgment based on a single parameter threshold is avoided; the three-level standards of green, yellow, and red are clearly defined and can be matched with graded self-healing strategies to achieve precise linkage between early warning and control; the quantified thresholds are adapted to the battery aging patterns under different operating conditions, providing maintenance personnel with clear decision-making basis and ensuring the safe and stable operation of batteries.

[0031] The RUL prediction algorithm module dynamically adjusts the preset degradation rate threshold based on the temperature fluctuations in the battery's operating environment. The temperature fluctuation conditions and corresponding threshold values ​​are as follows: Constant temperature operation (temperature fluctuation <5℃): the attenuation rate threshold is 0.5% / month; Fluctuating operating conditions (temperature fluctuation of 5-10℃): the attenuation rate threshold is 0.3% / month; Extreme operating conditions (temperature fluctuations >10℃ or sustained high temperatures): the attenuation rate threshold is 0.2% / month.

[0032] The threshold is set according to the temperature fluctuation conditions, which conforms to the aging pattern of batteries under different environments and avoids the risk of misjudgment under complex conditions with a fixed threshold; the dynamic matching mechanism can automatically switch the threshold according to real-time environmental parameters without manual intervention.

[0033] The formula for calculating the anomaly score is as follows: In the formula, For anomaly scoring, the value range is [0-1], where 0 is completely normal and 1 is completely abnormal; X is an input feature vector composed of 20 or more dimensions of battery operating parameters, and the parameters of each dimension are normalized to the interval [0,1]. Let X be the path length of X in a single decision tree within an isolated forest; Let X be the mean path length of X across multiple decision trees in an isolated forest; This is the average path length constant of the decision tree when the sample size is n; The exception determination rule is: when When the value is >0.6, the battery is considered to be in an abnormal operating state; when it is <0.4... When the value is ≤0.6, the battery operating state is determined to be abnormal; when... When the value is ≤0.4, the battery is considered to be in normal operating condition.

[0034] The advantage of this design is that it builds a quantitative and standardized anomaly judgment system, improving the accuracy and consistency of fault identification; at the same time, the three-level judgment rules have clear boundaries, which can distinguish between normal, boundary anomaly and abnormal state, avoiding the risk of misjudgment based on a single threshold.

[0035] The specific rules for anomaly detection in the battery module anomaly diagnosis module include setting quantitative judgment thresholds for the following four core parameter dimensions based on 20 or more operating parameters collected by the full-dimensional monitoring module: Individual cell voltage difference: If the voltage of any individual cell deviates from the average voltage of the cells in the battery pack by more than 0.3V, it is considered an abnormal voltage, which may indicate inconsistent cell attenuation or circuit disconnection. Temperature difference: If the difference between the highest and lowest temperatures inside the battery pack is greater than 10°C, it is considered an abnormal temperature and may indicate an impending thermal runaway. Sudden increase in internal resistance: If the increase in the internal resistance value of the current monitoring cycle is greater than 5% compared with the average internal resistance value of the previous monitoring cycle, it is judged as an abnormal internal resistance, which may indicate poor contact or internal short circuit fault. Degradation rate: If the battery degradation rate exceeds the preset degradation rate threshold of 0.5% / month under the corresponding operating conditions, it is judged as abnormal degradation and there may be an accelerated degradation fault.

[0036] Focusing on four core parameters—voltage, temperature, internal resistance, and degradation rate—and setting clear quantitative thresholds, this approach directly addresses typical fault risks such as inconsistent battery degradation and precursors to thermal runaway. The rules are deeply integrated with all-dimensional monitoring parameters, complementing and validating the isolated forest algorithm to avoid misjudgments by a single algorithm. Each rule is clearly targeted, enabling rapid identification of fault types and causes, providing precise decision-making support for graded self-healing strategies, and ensuring the safe operation of the battery system.

[0037] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0038] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing, characterized in that, It includes the following functional modules and their collaborative mechanisms: Full-dimensional monitoring module: used to collect battery operating parameters in no less than 20 dimensions in real time; RUL prediction algorithm module: Based on the hybrid degradation model, it uses the SOH data and operating parameters of the past 3 months to fit the degradation coefficient, calculate the remaining service life RUL, and trigger green, yellow and red three-level warnings based on the RUL value. At the same time, it calculates the battery degradation rate based on the SOH data, outputs the real-time value of the degradation rate and compares it with the preset degradation rate threshold. Battery module anomaly diagnosis module: Based on the isolated forest algorithm, it performs anomaly detection on 20+ dimension real-time running data, outputs anomaly scores and makes anomaly judgments, and classifies faults into first-level, second-level, and third-level categories according to the anomaly duration, RUL value, and number of anomaly dimensions. Battery module self-healing control module: Automatically executes graded self-healing strategies according to fault level. For level 1 abnormalities, active balancing control is executed; for level 2 faults, power reduction operation control is executed; and for level 3 faults, topology reconstruction isolation control is executed. Closed-loop feedback mechanism: The results of each diagnosis and self-healing are fed back to the RUL prediction and anomaly diagnosis module to achieve adaptive optimization.

2. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The 20-dimensional battery operating parameters in the full-dimensional monitoring module include: total voltage, total current, SOC, SOH, single cell voltage, maximum voltage, minimum voltage, voltage difference, battery pack temperature, average temperature, temperature difference, internal resistance, internal resistance rate, charge / discharge cycle count, and SOH decay rate.

3. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The formula for calculating the SOH data in the past 3 months and the operating parameters is as follows: In the formula, The battery health status at time t. Let be the actual discharge capacity of the battery at time t. This refers to the initial rated capacity of the battery.

4. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The hybrid decay model integrates three major decay factors: cycle, calendar, and temperature, improving prediction accuracy compared to a single linear model. The hybrid decay model is as follows: In the formula, This is the initial health constant; , The attenuation coefficient; This is the temperature coupling coefficient; The cumulative total number of charge-discharge cycles at time t; For calendar lifespan; This is a temperature compensation function, when Temp ≤ 25℃. =0, when Temp>25℃ =(Temp-25).

5. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 4, characterized in that: The attenuation coefficient , The value is automatically matched based on the battery type and real-time ambient temperature, and the specific value is: When the battery type is LFP and the ambient temperature is 15-25℃ =0.08% / 100cycles =1.5% / year; When the battery type is LFP and the ambient temperature is ≥25℃ =0.10% / 100cycle, =3.5% / year; When the battery type is ternary lithium battery =0.12% / 100cycle, =2.5% / year.

6. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The RUL prediction formula in the RUL prediction algorithm module is: In the formula, Let t be the remaining service life at time t. The linear decay rate of SOH over the past 90 days.

7. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The grading standards for the green, yellow, and red alerts are as follows: Green alert: RUL > 60 days, decay rate < 0.3% / month; Yellow alert: RUL∈[30,60] days, decay rate∈[0.3,0.5]% / month; Red alert: RUL < 30 days, decay rate > 0.5% / month.

8. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The RUL prediction algorithm module dynamically adjusts the preset degradation rate threshold according to the temperature fluctuation conditions of the battery's operating environment. The temperature fluctuation conditions and the corresponding threshold values ​​are as follows: Constant temperature operation (temperature fluctuation <5℃): the attenuation rate threshold is 0.5% / month; Fluctuating operating conditions (temperature fluctuation of 5-10℃): the attenuation rate threshold is 0.3% / month; Extreme operating conditions (temperature fluctuations >10℃ or sustained high temperatures): the attenuation rate threshold is 0.2% / month.

9. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing as described in claim 1, characterized in that: The formula for calculating the anomaly score is as follows: In the formula, For anomaly scoring, the value range is [0-1], where 0 is completely normal and 1 is completely abnormal; X is an input feature vector composed of 20 or more dimensions of battery operating parameters, and the parameters of each dimension are normalized to the interval [0,1]. Let X be the path length of X in a single decision tree within an isolated forest; Let X be the mean path length of X across multiple decision trees in an isolated forest; This is the average path length constant of the decision tree when the sample size is n; The anomaly determination rule is: when When the value is >0.6, the battery is considered to be in an abnormal operating state; when it is <0.4... When the value is ≤0.6, the battery operating state is determined to be abnormal; when... When the value is ≤0.4, the battery is considered to be in normal operating condition.

10. The battery fault diagnosis and automatic control system based on RUL prediction and graded self-healing according to claim 1, characterized in that: The specific rules for anomaly detection in the battery module anomaly diagnosis module include setting quantitative judgment thresholds for the following four core parameter dimensions based on 20 or more operating parameters collected by the full-dimensional monitoring module: Individual cell voltage difference: If the voltage of any individual cell deviates from the average voltage of the cells in the battery pack by more than 0.3V, it is considered an abnormal voltage, which may indicate inconsistent cell attenuation or circuit disconnection. Temperature difference: If the difference between the highest and lowest temperatures inside the battery pack is greater than 10°C, it is considered an abnormal temperature and may indicate an impending thermal runaway. Sudden increase in internal resistance: If the increase in the internal resistance value of the current monitoring cycle is greater than 5% compared with the average internal resistance value of the previous monitoring cycle, it is judged as an abnormal internal resistance, which may indicate poor contact or internal short circuit fault. Degradation rate: If the battery degradation rate exceeds the preset degradation rate threshold of 0.5% / month under the corresponding operating conditions, it is judged as abnormal degradation and there may be an accelerated degradation fault.

Images