Safety Protection Methods for Lithium-ion Battery Energy Storage Systems Based on Multi-dimensional State Monitoring

Abstract

This invention relates to the field of safety protection technology, and in particular to a safety protection method for lithium battery energy storage systems based on multi-dimensional state monitoring. The method includes: real-time parameter acquisition; identification of pending modules; identification of risk modules; identification of abnormal modules; adjustment of abnormal thresholds; and issuance of safety alarms. This invention establishes a causal and synchronicity judgment mechanism between parameters by jointly monitoring and quantitatively analyzing multi-dimensional parameters within a time window. First, abnormal candidates are screened using the state of charge value and duration. Then, a temperature threshold is derived from a current-temperature correspondence table triggered by parallel current, and the temperature deviation and duration are calculated. Simultaneously, the correlation between the abnormality percentage and the sliding fluctuation of the vibration frequency is used as an adjustment criterion. When the correlation significantly increases, the initial duration threshold is relaxed according to the deviation. This effectively solves the problems of inaccurate battery anomaly identification and response delay caused by single-parameter monitoring and fixed threshold judgment.

Description

Technical Field

[0001] This invention relates to the field of safety protection technology, and in particular to a safety protection method for lithium battery energy storage systems based on multi-dimensional state monitoring. Background Technology

[0002] With the widespread application of large-scale machinery in industrial environments, lithium-ion battery energy storage systems are increasingly deployed in production lines, energy storage stations, and mobile equipment. Frequent mechanical vibrations, complex temperature variations, and significant current load fluctuations in industrial environments make battery modules prone to abnormal conditions such as rapid temperature rise, bulging, and cracking, increasing safety risks. Traditional static threshold and single-parameter monitoring methods are insufficient to accurately reflect battery state changes under high vibration and high load conditions, potentially delaying fault detection or causing false alarms. In this multi-factor coupled and dynamically changing environment, achieving real-time, multi-dimensional monitoring of battery module status and adaptive threshold adjustment has become a significant challenge in ensuring the safe and stable operation of energy storage systems.

[0003] Chinese Patent Application Publication No. CN120670784A discloses a lithium battery energy storage safety management system and method. The method includes: acquiring real-time heterogeneous invasion characteristics of the environment in which the lithium battery is located; analyzing heterogeneous contaminant invasion characteristics based on the real-time heterogeneous invasion characteristics and integrating them to form a multi-source invasion dataset; performing multi-level safety risk coupling modeling based on the multi-source invasion dataset and outputting a comprehensive risk level; adaptively activating a defense mechanism based on the comprehensive risk level and generating a cross-dimensional defense strategy; evaluating the battery status after defense according to the cross-dimensional defense strategy and generating and outputting a safety defense effectiveness report.

[0004] Therefore, the lithium battery energy storage safety management method has the following problems: the method lacks real-time correlation analysis of multi-dimensional parameters such as mechanical vibration, current fluctuation and temperature change, making it difficult to accurately judge the dynamic safety status of the battery module; the method's anomaly detection and defense strategies rely heavily on fixed thresholds, lacking adaptive adjustment capabilities, and are prone to delayed response or false alarms; the method does not adequately consider the coupling relationship between series and parallel modules in the same electrical circuit, and cannot take differentiated protection measures for local risks in a timely manner; the method lacks continuous monitoring and trend analysis of physical anomalies such as bulges and coating cracks on the module surface, making it difficult to comprehensively assess potential safety hazards. Summary of the Invention

[0005] To address this, the present invention provides a safety protection method for lithium battery energy storage systems based on multi-dimensional state monitoring. This method overcomes the problems of inaccurate battery anomaly identification and response delay caused by single parameter monitoring and fixed threshold judgment in the prior art by real-time monitoring of multi-dimensional parameters and combining dynamic correlation analysis of series and parallel modules with adaptive threshold adjustment.

[0006] To achieve the above objectives, the present invention provides a safety protection method for a lithium battery energy storage system based on multi-dimensional state monitoring, comprising:

[0007] Real-time acquisition of vibration frequency in the operating area of ​​large mechanical equipment, state of charge value of each battery module in the lithium battery energy storage system in the operating area, module temperature and surface bulge area of ​​series modules in the battery module, and module current and surface coating crack area of ​​parallel modules in the battery module;

[0008] Several undetermined modules are determined based on the state of charge value and the preset anomaly detection threshold.

[0009] Several risk modules are determined based on the module temperature of the series modules in each of the undetermined modules and the module current of the parallel modules in the same electrical circuit.

[0010] Based on the variation characteristics of the surface bulge area of ​​the series module and the surface coating crack area of ​​the parallel module in each of the battery modules within the next preset anomaly determination time, and the risk module, a number of abnormal modules are determined;

[0011] The fluctuation degree is determined based on the number of series modules and parallel modules in the same electrical circuit in the abnormal modules within the next preset adjustment judgment period, and the preset abnormal judgment threshold is adjusted based on the fluctuation degree, the total number of all abnormal modules within the same preset adjustment judgment period, the total number of all battery modules, and the vibration frequency.

[0012] A security alarm is issued for the abnormal module that is re-determined after adjusting the preset abnormality determination threshold.

[0013] Furthermore, the process of determining several undetermined modules based on the state of charge value and the preset anomaly detection threshold includes:

[0014] The duration of the period when the state of charge value is greater than a preset state of charge threshold is obtained to determine the abnormal duration;

[0015] When the duration of the abnormality exceeds the preset abnormality determination threshold, the battery module is determined to be the undetermined module.

[0016] Furthermore, the process of determining a number of risk modules based on the module temperature of the series modules and the module current of the parallel modules in the same electrical circuit in each of the undetermined modules includes:

[0017] When the module current exceeds a preset module current threshold, the parallel module is marked as a temporary module.

[0018] The temperature threshold is determined based on the module current of each temporary module and the preset current temperature gauge.

[0019] Based on the module temperature and the corresponding temperature threshold of the series module that is in the same electrical circuit as the temporary module within the next preset risk determination period, a number of risk modules are determined.

[0020] Furthermore, the process of determining several risk modules based on the module temperature and corresponding temperature threshold of the series modules that are in the same electrical circuit as the temporary module within the next preset risk determination period includes:

[0021] Calculate the difference between the module temperature and the temperature threshold within the preset risk determination period to obtain several expected temperature deviations;

[0022] The duration for which the expected temperature deviation is greater than a preset temperature deviation threshold is obtained to determine the risk duration. When the risk duration is greater than the preset risk duration threshold, the series module and the temporary module in the same electrical circuit are identified as the risk modules to determine a number of risk modules.

[0023] Furthermore, the process of determining several abnormal modules based on the variation characteristics of the surface bulge area of ​​the series modules and the surface coating crack area of ​​the parallel modules in each of the battery modules within the next preset anomaly determination time, as well as the risk modules, includes:

[0024] Calculate the average value of all surface bulge areas at the initial time and each time within the preset anomaly determination period to obtain several bulge area averages, and calculate the standard deviation of all bulge area averages to obtain bulge sliding fluctuation values.

[0025] Calculate the average area of ​​all surface coating cracks within the same preset anomaly determination time period, from the initial time to each time, to obtain several crack area averages, and calculate the standard deviation of all crack area averages to obtain crack slip fluctuation values.

[0026] Several abnormal modules are determined based on the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module.

[0027] Furthermore, the process of determining several abnormal modules based on the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module includes:

[0028] When the swelling fluctuation value is greater than the preset swelling fluctuation threshold, the battery module is determined to be the target module;

[0029] When the crack sliding fluctuation value is greater than the preset crack sliding threshold, the battery module is determined to be the target module;

[0030] The intersection of all the target modules and all the abnormal modules is marked as an abnormal module, thereby identifying several abnormal modules.

[0031] Furthermore, the process of determining the stability of change based on the number of series modules and parallel modules in the same electrical circuit within the next preset adjustment determination period includes:

[0032] Calculate the ratio of the number of series modules and parallel modules in the same electrical circuit to the number of abnormal modules at each moment within the preset adjustment determination time to obtain several adjustment ratios;

[0033] The stability of the change is determined based on the proportion of all the adjustments mentioned.

[0034] Furthermore, the process of determining the stability of change based on all the aforementioned adjustment proportions includes:

[0035] Calculate the absolute value of the difference between the adjustment percentages between any two adjacent moments within the preset adjustment determination time to obtain several percentage change ranges;

[0036] Calculate the standard deviation of all the aforementioned percentage changes to obtain the volatility of the change.

[0037] Furthermore, the process of adjusting the preset anomaly determination threshold based on the fluctuation degree, the number of all abnormal modules within the same preset adjustment determination time, the total number of all battery modules, and the vibration frequency includes:

[0038] When the fluctuation degree is greater than the preset fluctuation degree threshold, the ratio of the number of all abnormal modules to the number of all battery modules at each time within the same preset adjustment judgment period is calculated to obtain several abnormal proportions.

[0039] The preset anomaly determination threshold is adjusted based on the anomaly percentage and the vibration frequency.

[0040] Furthermore, the process of adjusting the preset anomaly determination threshold based on the anomaly percentage and the vibration frequency includes:

[0041] Calculate the standard deviation of the vibration frequency from the initial time to each time within the preset adjustment judgment period to obtain several vibration slip fluctuation values;

[0042] Calculate the standard deviation of the abnormal proportion from the initial time to each time within the same preset adjustment judgment period to obtain several abnormal sliding fluctuation values;

[0043] Calculate the Pearson correlation coefficients of all the vibration slip fluctuation values ​​and all the abnormal slip fluctuation values ​​to obtain the adjustment determination coefficient;

[0044] When the adjustment judgment coefficient is greater than the preset adjustment judgment standard value, the preset abnormal judgment threshold is increased according to the relative deviation between the adjustment judgment coefficient and the preset adjustment judgment standard value.

[0045] Compared with existing technologies, the beneficial effects of this invention lie in establishing a causal and synchronicity judgment mechanism among parameters by jointly monitoring and quantifying multi-dimensional parameters such as state of charge, module temperature, bulge area, parallel module current, coating crack area, and vibration frequency in the operating area within a time window. First, anomaly candidates are screened out based on the state of charge value and duration. Then, a temperature threshold is derived from the current-temperature correspondence table triggered by parallel current, and the temperature deviation and duration are calculated. Finally, the moving average and standard deviation of the bulge area and crack area are used to assess macroscopic surface changes. Anomalies are confirmed when surface changes highly overlap with electrical risk modules. Simultaneously, the proportion of anomalies is used to assess the overall surface changes. Using the correlation with the vibration frequency as an adjustment criterion, the initial duration threshold is relaxed according to the deviation when the correlation increases significantly. This suppresses short-term false alarms caused by mechanical vibration while maintaining sensitivity to persistent hazards. It effectively balances robustness to transient disturbances with sensitive response to persistent anomalies, reducing false alarm rates and unnecessary manual interventions. It also identifies hazardous modules that require isolation, load reduction, or cooling earlier, thereby improving the reliability and safety of the energy storage system and facilitating subsequent maintenance and fault tracing. This effectively solves the problems of inaccurate battery anomaly identification and response delay caused by single parameter monitoring and fixed threshold judgment.

[0046] Furthermore, by quantifying the duration of the state of charge exceeding a preset threshold, this process distinguishes between short-term fluctuations and truly continuous high-charge conditions: continuous high charge increases the rate of internal chemical reactions in the cell, accelerates the accumulation of internal resistance and heat, and promotes stress accumulation in the electrodes and diaphragms, thereby increasing the probability of module bulging, temperature rise, or abnormal current. Therefore, using "charge value + duration" as a screening criterion can identify undetermined modules under long-term high-stress conditions earlier, reduce false alarms caused by transient fluctuations, and facilitate the concentration of monitoring and handling resources on truly risky modules, thereby enabling early implementation of load reduction, enhanced cooling, or maintenance measures, reducing the risk of thermal runaway, and improving operation and maintenance efficiency and equipment lifespan.

[0047] Furthermore, by using the sudden current of the parallel modules as a trigger and determining the corresponding temperature threshold by referring to the current-temperature correspondence table, and then comparing the actual temperature of the series modules in the same electrical circuit with the threshold over a preset risk period, this process can distinguish between transient temperature rise caused by instantaneous current pulses and continuous heat load: increased current will accelerate local heating through I²R loss and electrochemical reaction. If the temperature of the connected series modules continues to exceed the threshold derived from the current expectation, it indicates that the heat cannot be dissipated in time or there is an abnormal local impedance, and there is a risk of further temperature rise or even thermal runaway. Therefore, this method can more accurately identify dangerous modules caused by uneven current, sudden load changes or abnormal internal resistance, reduce the false alarm rate, and provide timely basis for precise handling such as load reduction, enhanced cooling or isolation.

[0048] Furthermore, by continuously monitoring and calculating the difference between the module temperature and the corresponding temperature threshold of temporary and series modules within the same electrical circuit within a preset risk determination time, this method can obtain the expected temperature deviation of the module. By judging the duration for which the deviation exceeds the set threshold, the risk duration is determined. When the risk duration exceeds the preset risk duration threshold, the series module and its related temporary module in a high-temperature abnormal state can be accurately identified, thereby achieving early detection and protection against potential overheating risks, effectively improving the overall operational safety of the energy storage system, and ensuring that the dynamic relationship between current load and temperature changes is reasonably considered and used for risk assessment.

[0049] Furthermore, by continuously monitoring the surface bulge area and coating crack area of ​​the battery module, and combining this with the temperature and current status of the at-risk module, the sliding fluctuation values ​​of the bulge area and crack area are calculated, thereby quantifying the structural change trend of the module within a predetermined anomaly timeframe. This method can correlate minute but continuous physical deformations with potential thermal and electrical stresses, enabling early identification and warning of abnormal modules, and improving the safety and reliability of energy storage systems.

[0050] Furthermore, by setting thresholds for the bulge slippage and crack slippage values ​​of battery modules, modules exceeding the thresholds are marked as target modules. This data is then compared with existing abnormal modules for further screening, thus accurately identifying modules with genuine structural anomalies. This method combines the physical deformation amplitude of a module with its electrical and thermal risk characteristics, enabling precise location and timely warning of potentially faulty modules, thereby enhancing the overall safety protection capability of the energy storage system.

[0051] Furthermore, by calculating the ratio of the number of series and parallel modules in the same electrical circuit to the number of abnormal modules, the operating status and mutual influence of battery modules within a preset adjustment judgment period can be dynamically reflected, the stability of changes between modules can be quantified, thereby effectively identifying the fluctuation trend of the number of abnormal modules and adjusting the judgment strategy in a timely manner to ensure the safety and reliability of the overall system operation.

[0052] Furthermore, by calculating the difference in adjustment percentages between any two adjacent time points and determining the standard deviation, the dynamic fluctuations in the abnormal distribution of battery modules within the preset adjustment judgment period can be quantitatively reflected. This reveals the coupling relationship between the changes in the number of series and parallel modules in the same electrical circuit and the overall changes in the number of abnormal modules. This fluctuation reflects the adaptive adjustment characteristics of the system under the influence of multiple dimensions such as vibration, temperature, and current. It allows for the accurate capture of sudden or continuous changes in abnormal modules by varying the fluctuation degree, thus providing a scientific basis for dynamically adjusting the preset abnormal judgment threshold and ensuring more reliable safety protection of the lithium battery energy storage system under different operating conditions.

[0053] Furthermore, by comprehensively analyzing the fluctuation degree in conjunction with the number of abnormal modules, the total number of battery modules, and the vibration frequency, the preset anomaly detection threshold can be dynamically adjusted. The fluctuation degree reflects the strength of the fluctuation in the number of abnormal modules over time, the anomaly ratio quantitatively describes the proportion of abnormal modules in the system under the current operating state, and the vibration frequency provides quantitative information on the impact of the external mechanical environment on the state of the battery modules. By combining these parameters, the overall risk level of the system can be scientifically assessed and the threshold adjusted, allowing the threshold to adaptively change with abnormal fluctuations in battery modules and external vibration conditions, thereby improving the accuracy of abnormal module detection and the safety of system operation.

[0054] Furthermore, by introducing Pearson correlation analysis between vibration slip fluctuation values ​​and abnormal slip fluctuation values, the synchronicity between changes in mechanical vibration and abnormal distribution of battery modules can be dynamically assessed. When the correlation between the two increases significantly, it means that equipment operation disturbances may have a cumulative stress effect on the energy storage system, thereby increasing the probability of anomalies. By comparing this correlation coefficient with a preset standard value and adaptively adjusting the preset anomaly judgment threshold based on its relative deviation, the system can automatically raise the threshold under conditions of severe vibration or frequent load fluctuations to reduce false alarms; while lowering the threshold during stable operation to improve the sensitivity of anomaly detection, thereby achieving dynamic safety balance and accurate early warning of the energy storage system under multiple operating conditions. Attached Figure Description

[0055] Figure 1 This is a flowchart of the safety protection method for lithium battery energy storage systems based on multi-dimensional state monitoring in this embodiment;

[0056] Figure 2 This embodiment defines the decision logic diagram for determining the modules to be determined.

[0057] Figure 3 This embodiment defines the decision logic diagram for the risk module.

[0058] Figure 4 The determination logic diagram for the target module in this embodiment is shown. Detailed Implementation

[0059] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0060] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0061] Please see Figure 1 The flowchart shown is a process for a safety protection method for a lithium battery energy storage system based on multi-dimensional state monitoring, as described in this embodiment. This embodiment provides a safety protection method for a lithium battery energy storage system based on multi-dimensional state monitoring, including:

[0062] Real-time acquisition of vibration frequency in the operating area of ​​large mechanical equipment, state of charge value of each battery module in the lithium battery energy storage system in the operating area, module temperature and surface bulge area of ​​series modules in the battery module, and module current and surface coating crack area of ​​parallel modules in the battery module;

[0063] Several undetermined modules are determined based on the state of charge value and the preset anomaly detection threshold.

[0064] Several risk modules are determined based on the module temperature of the series modules in each of the undetermined modules and the module current of the parallel modules in the same electrical circuit.

[0065] Based on the variation characteristics of the surface bulge area of ​​the series module and the surface coating crack area of ​​the parallel module in each of the battery modules within the next preset anomaly determination time, and the risk module, a number of abnormal modules are determined;

[0066] The fluctuation degree is determined based on the number of series modules and parallel modules in the same electrical circuit in the abnormal modules within the next preset adjustment judgment period, and the preset abnormal judgment threshold is adjusted based on the fluctuation degree, the total number of all abnormal modules within the same preset adjustment judgment period, the total number of all battery modules, and the vibration frequency.

[0067] A security alarm is issued for the abnormal module that is re-determined after adjusting the preset abnormality determination threshold.

[0068] In this embodiment, the operating area of ​​large mechanical equipment refers to spaces such as factory workshops, power plants, or equipment platforms that generate periodic or random vibrations. In this embodiment, the vibration frequency of this operating area is acquired in real time, and key information of each battery module in the lithium battery energy storage system within the operating area is simultaneously collected. The energy storage system is an energy storage cabinet or mobile energy storage unit installed in this area, containing multiple independently monitorable and manageable battery modules. Each battery module consists of several cells connected in series and parallel. The system contains both series and parallel modules in the same electrical circuit, as well as series and parallel modules in different electrical circuits. The electrical connection relationship between modules is determined by the topology information or wiring mapping file of the BMS. Specifically, the acquisition method involves obtaining the State of Charge (SOC) value at the module side through the BMS and module-level sensors. The SOC is estimated using coulomb counting combined with Kalman filtering. The module temperature of the series modules is obtained through embedded thermocouples or NTC sensors and fixed-point infrared temperature measurement devices. The bulge area and coating crack area on the module surface are captured by a high-resolution industrial camera under controlled lighting, and the images are used for data collection. The method of segmentation and scaling calibration is quantified. The module current is sampled in real time through a shunt or Hall current sensor. The vibration frequency is sampled by a triaxial accelerometer installed on the base of the energy storage cabinet or the ground and the main frequency component is obtained by fast Fourier transform. The data sampling rate is distinguished according to the measurement object. Electrical quantities and vibration acceleration are sampled in real time at high frequency. All sensor data are aggregated to the edge gateway through the module-level BMS via CAN, RS485 bus or industrial Ethernet. The edge gateway completes the timing synchronization using PTP or NTP, and performs noise reduction and feature extraction, such as temperature rise rate, mean and standard deviation of bulge pixels, moving average of crack pixel area, and current pulsation amplitude. The vibration signal is subjected to spectrum analysis at the edge and vibration sliding fluctuation value is generated. After timing alignment, all features are initially judged at the edge and reported to the cloud for historical comparison and threshold optimization. The image processing process uses OpenCV or deep learning segmentation model to perform pixel-level segmentation and area conversion. Electrical measurements are performed using differential amplification and ADC sampling and filtering to save the original timing for backtracking, thereby realizing clear, reproducible real-time acquisition and subsequent joint analysis of the parameters.

[0069] In this embodiment, when the adjusted preset anomaly judgment threshold triggers an anomaly condition, the system immediately sends an alarm trigger command to the anomaly module through the signal control module. After receiving the command, the anomaly module simultaneously activates the audible and visual alarm unit to emit high-frequency flashes and buzzer sounds, and uploads the anomaly type, vibration frequency, and adjusted threshold data to the monitoring center through the communication interface, realizing a dual warning mechanism of local and remote, thereby ensuring timely response before potential faults amplify.

[0070] The preset anomaly judgment threshold is an initial threshold used to determine the duration of abnormal state of charge of the battery module. It depends on the design capacity of the battery module and the fluctuation characteristics of the operating environment, and is usually set between 5 and 30 seconds. In this embodiment, it is set to 15 seconds, which can effectively filter out pending modules with continuous anomalies. The preset anomaly determination time is a time window used to observe the change characteristics of the bulge area and coating crack area. It depends on the change rate of the battery surface and the thermal response characteristics of the system, and is usually set between 60 and 300 seconds. In this embodiment, it is set to 120 seconds, which can capture short-term surface anomalies and reduce the impact of transient fluctuations. The preset adjustment judgment time is a time interval used to calculate the relationship between the number of abnormal modules and the vibration frequency fluctuation to adjust the threshold. It depends on the vibration law of the mechanical equipment and the system response cycle, and is usually set between 300 and 900 seconds. In this embodiment, it is set to 600 seconds, which can balance the sensitivity and stability of the threshold adjustment.

[0071] By jointly monitoring and quantifying multi-dimensional parameters such as state of charge (SCC), module temperature, bulge area, parallel module current, coating crack area, and vibration frequency in the operating area within a time window, a causal and synchronicity judgment mechanism among the parameters was established. First, anomaly candidates were screened out based on SCC values ​​and duration. Then, a temperature threshold was derived from a current-temperature correspondence table triggered by parallel current, and temperature deviation and duration were calculated. Finally, the sliding mean and standard deviation of bulge area and crack area were used to assess macroscopic surface changes. Anomalies were confirmed when surface changes highly overlapped with electrical risk modules. Simultaneously, the sliding wave of the anomaly proportion and vibration frequency was used to assess the anomaly. Using dynamic correlation as an adjustment criterion, the initial duration threshold is relaxed according to the deviation when the correlation increases significantly, in order to suppress short-term false alarms caused by mechanical vibration and maintain sensitivity to persistent dangers. This effectively balances robustness to transient disturbances and sensitive response to persistent anomalies, reducing false alarm rates and unnecessary manual intervention, and identifying dangerous modules that need to be isolated, deloaded, or cooled earlier. This improves the reliability and safety of the energy storage system and facilitates subsequent maintenance and fault tracing. It effectively solves the problems of inaccurate battery anomaly identification and response delay caused by single parameter monitoring and fixed threshold judgment.

[0072] Please see Figure 2 As shown, this is the logic diagram for determining the pending modules in this embodiment. The process of determining several pending modules based on the state of charge value and the preset anomaly determination threshold includes:

[0073] The duration of the period when the state of charge value is greater than a preset state of charge threshold is obtained to determine the abnormal duration;

[0074] When the duration of the abnormality exceeds the preset abnormality determination threshold, the battery module is determined to be the undetermined module.

[0075] The preset state of charge threshold is a percentage value of SOC, which depends on the battery chemistry, rated capacity and operating conditions. It is usually set between 80% and 95%. In this embodiment, it is set to 90% to promptly identify long-term high charge conditions so as to trigger subsequent risk assessment and handling.

[0076] By quantifying the duration of a state of charge exceeding a preset threshold, this process distinguishes between short-term fluctuations and truly sustained high-charge conditions. Sustained high charge increases the rate of internal chemical reactions, accelerates internal resistance and heat accumulation, and promotes stress buildup on electrodes and diaphragms, thereby increasing the probability of module bulging, temperature rise, or abnormal current. Therefore, using "charge value + duration" as a screening criterion can identify undetermined modules under long-term high-stress conditions earlier, reducing false alarms caused by transient fluctuations. This allows monitoring and handling resources to be concentrated on truly risky modules, enabling early implementation of load reduction, enhanced cooling, or maintenance measures, reducing the risk of thermal runaway, and improving operation and maintenance efficiency and equipment lifespan.

[0077] Specifically, in this embodiment, the process of determining a number of risk modules based on the module temperature of the series modules and the module current of the parallel modules in the same electrical circuit in each of the undetermined modules includes:

[0078] When the module current exceeds a preset module current threshold, the parallel module is marked as a temporary module.

[0079] The temperature threshold is determined based on the module current of each temporary module and the preset current temperature gauge.

[0080] Based on the module temperature and the corresponding temperature threshold of the series module that is in the same electrical circuit as the temporary module within the next preset risk determination period, a number of risk modules are determined.

[0081] The preset module current threshold is the current threshold used to mark parallel modules as temporary modules. It depends on the module's rated capacity and heat dissipation capability, and is usually set between 50A and 300A. In this embodiment, it is set to 150A, which can promptly identify overload or abnormal current shunting. The preset risk determination time is the time window used to determine whether the current-temperature deviation is a persistent risk. It depends on the battery module's thermal response speed and load fluctuation characteristics, and is usually set between 30 seconds and 180 seconds. In this embodiment, it is set to 60 seconds, which can capture continuous temperature rise and effectively filter transient pulse interference in a short time.

[0082] The preset current temperature table is a lookup table that maps different current levels to the corresponding allowable module temperatures. It depends on the cell chemical characteristics, module internal resistance, and cooling efficiency. It is usually divided into several levels from 0 to the rated maximum current and gives the corresponding temperature threshold. In this embodiment, the preset current temperature table is shown in Table 1 below:

[0083] Table 1 Preset Current Temperature Table

[0084]

[0085] By using the sudden current of parallel modules as a trigger and referring to a current-temperature correspondence table to determine the corresponding temperature threshold, and then comparing the actual temperature of series modules in the same electrical circuit with this threshold within a preset risk period, this process can distinguish between transient temperature rise caused by instantaneous current pulses and continuous heat load: increased current will accelerate local heating through I²R loss and electrochemical reaction. If the temperature of connected series modules continues to exceed the threshold derived from the current expectation, it indicates that heat cannot be dissipated in time or there is an abnormal local impedance, and there is a risk of further temperature rise or even thermal runaway. Therefore, this method can more accurately identify dangerous modules caused by uneven current, sudden load changes or abnormal internal resistance, reduce the false alarm rate, and provide timely basis for precise handling such as load reduction, enhanced cooling or isolation.

[0086] Please see Figure 3 As shown, this is the logic diagram for determining risk modules in this embodiment. In this embodiment, the process of determining several risk modules based on the module temperature and the corresponding temperature threshold of the series modules that are in the same electrical circuit as the temporary module within the next preset risk determination time includes:

[0087] Calculate the difference between the module temperature and the temperature threshold within the preset risk determination period to obtain several expected temperature deviations;

[0088] The duration for which the expected temperature deviation is greater than a preset temperature deviation threshold is obtained to determine the risk duration. When the risk duration is greater than the preset risk duration threshold, the series module and the temporary module in the same electrical circuit are identified as the risk modules to determine a number of risk modules.

[0089] The preset temperature deviation threshold is the boundary value used to judge the abnormal temperature of the module. It depends on the heat capacity, thermal conductivity and normal operating temperature fluctuation range of the battery material. It is usually set between 2°C and 5°C. In this embodiment, it is set to 3°C, which can accurately distinguish between short-term temperature fluctuations and real overheating risks. The preset risk duration threshold is the boundary value used to judge the continuous abnormal time of the module. It depends on the thermal response speed of the battery module and the system safety requirements. It is usually set between 30 seconds and 120 seconds. In this embodiment, it is set to 60 seconds, which can ensure that only modules with continuous overheating are judged as risk modules, thereby avoiding false alarms.

[0090] By continuously monitoring and calculating the difference between the module temperature and the corresponding temperature threshold of temporary and series modules within the same electrical circuit within a preset risk determination time, this method can obtain the expected temperature deviation of the module. By judging the duration for which the deviation exceeds the set threshold, the risk duration is determined. When the risk duration exceeds the preset risk duration threshold, the series module and its related temporary module in a high-temperature abnormal state can be accurately identified, thereby realizing the early detection and protection of potential overheating risks, effectively improving the overall operational safety of the energy storage system, and ensuring that the dynamic relationship between current load and temperature changes is reasonably considered and used for risk assessment.

[0091] Specifically, the process of determining several abnormal modules based on the variation characteristics of the surface bulge area of ​​the series modules and the surface coating crack area of ​​the parallel modules in each of the battery modules within the next preset anomaly determination time, as well as the risk modules, includes:

[0092] Calculate the average value of all surface bulge areas at the initial time and each time within the preset anomaly determination period to obtain several bulge area averages, and calculate the standard deviation of all bulge area averages to obtain bulge sliding fluctuation values.

[0093] Calculate the average area of ​​all surface coating cracks within the same preset anomaly determination time period, from the initial time to each time, to obtain several crack area averages, and calculate the standard deviation of all crack area averages to obtain crack slip fluctuation values.

[0094] Several abnormal modules are determined based on the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module.

[0095] By continuously monitoring the surface bulge area and coating crack area of ​​the battery module, and combining this with the temperature and current status of the at-risk module, the sliding fluctuation values ​​of the bulge area and crack area are calculated, thereby quantifying the structural change trend of the module within a predetermined anomaly period. This method can correlate minute but continuous physical deformations with potential thermal and electrical stresses, enabling early identification and warning of abnormal modules, and improving the safety and reliability of energy storage systems.

[0096] Please see Figure 4 As shown, this is the logic diagram for determining the target module in this embodiment. In this embodiment, the process of determining several abnormal modules based on the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module includes:

[0097] When the swelling fluctuation value is greater than the preset swelling fluctuation threshold, the battery module is determined to be the target module;

[0098] When the crack sliding fluctuation value is greater than the preset crack sliding threshold, the battery module is determined to be the target module;

[0099] The intersection of all the target modules and all the abnormal modules is marked as an abnormal module, thereby identifying several abnormal modules.

[0100] The preset bulge sliding threshold is a threshold used to determine whether the bulge change of the battery module is abnormal. It depends on the expansion characteristics of the battery material and the ambient temperature. It is usually set between 0.1 mm and 0.5 mm. In this embodiment, it is set to 0.3 mm, which can effectively identify modules with rapid bulge growth or abnormal expansion. The preset crack sliding threshold is a threshold used to determine whether the crack expansion of the surface coating of the battery module is abnormal. It depends on the crack resistance of the coating material and long-term stress accumulation. It is usually set between 0.05 square millimeters and 0.3 square millimeters. In this embodiment, it is set to 0.15 square millimeters, which can promptly detect modules with obvious crack expansion trends.

[0101] By setting thresholds for the bulge and crack slip fluctuation values ​​of battery modules, modules exceeding the thresholds are marked as target modules. This data is then compared with existing abnormal modules for further screening, thus accurately identifying modules with genuine structural anomalies. This method combines the physical deformation amplitude of a module with its electrical and thermal risk characteristics, enabling precise location and timely warning of potentially faulty modules, thereby enhancing the overall safety and protection capabilities of the energy storage system.

[0102] Specifically, the process of determining the stability of change based on the number of series modules and parallel modules in the same electrical circuit within the next preset adjustment determination period includes:

[0103] Calculate the ratio of the number of series modules and parallel modules in the same electrical circuit to the number of abnormal modules at each moment within the preset adjustment determination time to obtain several adjustment ratios;

[0104] The stability of the change is determined based on the proportion of all the adjustments mentioned.

[0105] By calculating the ratio of the number of series and parallel modules in the same electrical circuit to the number of abnormal modules, the operating status and mutual influence of battery modules within a preset adjustment judgment period can be dynamically reflected, the stability of changes between modules can be quantified, thereby effectively identifying the fluctuation trend of the number of abnormal modules and adjusting the judgment strategy in a timely manner to ensure the safety and reliability of the overall system operation.

[0106] Specifically, the process of determining the stability of change based on all the aforementioned adjustment proportions includes:

[0107] Calculate the absolute value of the difference between the adjustment percentages between any two adjacent moments within the preset adjustment determination time to obtain several percentage change ranges;

[0108] Calculate the standard deviation of all the aforementioned percentage changes to obtain the volatility of the change.

[0109] By calculating the difference in adjustment percentages between any two adjacent moments and determining the standard deviation, the dynamic fluctuations in the abnormal distribution of battery modules within a preset adjustment judgment period can be quantitatively reflected. This reveals the coupling relationship between changes in the number of series and parallel modules in the same electrical circuit and changes in the overall number of abnormal modules. This fluctuation reflects the adaptive adjustment characteristics of the system under the influence of multiple dimensions such as vibration, temperature, and current. It allows for accurate capture of sudden or continuous changes in abnormal modules by varying the fluctuation degree, thus providing a scientific basis for dynamically adjusting the preset abnormal judgment threshold and ensuring more reliable safety protection of lithium battery energy storage systems under different operating conditions.

[0110] Specifically, the process of adjusting the preset anomaly determination threshold based on the degree of change, the number of all abnormal modules within the same preset adjustment determination time, the total number of all battery modules, and the vibration frequency includes:

[0111] When the fluctuation degree is greater than the preset fluctuation degree threshold, the ratio of the number of all abnormal modules to the number of all battery modules at each time within the same preset adjustment judgment period is calculated to obtain several abnormal proportions.

[0112] The preset anomaly determination threshold is adjusted based on the anomaly percentage and the vibration frequency.

[0113] The preset fluctuation threshold is a critical parameter used to determine whether there are significant fluctuations in the system's operating state. It depends on the historical fluctuation characteristics of the lithium battery energy storage system under different operating loads and the stability of the vibration environment. It is usually set between 0.05 and 0.15. In this embodiment, it is set to 0.1, which can ensure sensitive response to minor anomalies while avoiding misjudgments caused by fluctuations in normal operation, thereby achieving effective identification and dynamic risk control of abnormal change trends in the battery module.

[0114] By comprehensively analyzing the fluctuation degree in conjunction with the number of abnormal modules, the total number of battery modules, and the vibration frequency, the preset anomaly detection threshold can be dynamically adjusted. The fluctuation degree reflects the strength of the fluctuation in the number of abnormal modules over time, the anomaly ratio quantitatively describes the proportion of abnormal modules in the system under the current operating state, and the vibration frequency provides quantitative information on the impact of the external mechanical environment on the state of the battery modules. By combining these parameters, the overall risk level of the system can be scientifically assessed and the threshold adjusted, allowing the threshold to adaptively change with abnormal fluctuations in battery modules and external vibration conditions, thereby improving the accuracy of abnormal module detection and the safety of system operation.

[0115] Specifically, the process of adjusting the preset anomaly determination threshold based on the anomaly percentage and the vibration frequency includes:

[0116] Calculate the standard deviation of the vibration frequency from the initial time to each time within the preset adjustment judgment period to obtain several vibration slip fluctuation values;

[0117] Calculate the standard deviation of the abnormal proportion from the initial time to each time within the same preset adjustment judgment period to obtain several abnormal sliding fluctuation values;

[0118] Calculate the Pearson correlation coefficients of all the vibration slip fluctuation values ​​and all the abnormal slip fluctuation values ​​to obtain the adjustment determination coefficient;

[0119] When the adjustment judgment coefficient is greater than the preset adjustment judgment standard value, the preset abnormal judgment threshold is increased according to the relative deviation between the adjustment judgment coefficient and the preset adjustment judgment standard value, where T'=T×(1+k×︱A-A0︱ / A0), T' is the increased preset abnormal judgment threshold, T is the original preset abnormal judgment threshold, k is the preset threshold adjustment coefficient, A is the adjustment judgment coefficient, and A0 is the preset adjustment judgment standard value.

[0120] The preset adjustment judgment standard value is a judgment benchmark used to measure the strength of the correlation between vibration slip fluctuation value and abnormal slip fluctuation value. It depends on the system noise level and equipment operating stability, and is usually set between 0.6 and 0.8. In this embodiment, it is set to 0.7, which can ensure that the judgment sensitivity is improved in time when the abnormal correlation is significantly enhanced, thus conforming to the nonlinear response law brought about by the enhanced wave coupling in the natural system. The preset threshold adjustment coefficient is a proportional parameter used to control the adjustment range of the preset abnormal judgment threshold. It depends on the equipment type, load inertia and vibration signal response rate, and is usually set between 0.1 and 0.3. In this embodiment, it is set to 0.2, which can achieve dynamic adjustment while maintaining judgment stability, conforming to the energy regulation law of disturbance and feedback balance in the natural system.

[0121] By introducing Pearson correlation analysis between vibration slip fluctuation values ​​and abnormal slip fluctuation values, the synchronicity between changes in mechanical vibration and abnormal distribution of battery modules can be dynamically assessed. When the correlation between the two increases significantly, it means that equipment operation disturbances may have a cumulative stress effect on the energy storage system, thereby increasing the probability of anomalies. By comparing this correlation coefficient with a preset standard value and adaptively adjusting the preset anomaly judgment threshold based on its relative deviation, the system can automatically raise the threshold under conditions of severe vibration or frequent load fluctuations to reduce false alarms; while lowering the threshold during stable operation to improve the sensitivity of anomaly detection, thereby achieving dynamic safety balance and accurate early warning of the energy storage system under multiple operating conditions.

[0122] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for safety protection of a lithium battery energy storage system based on multi-dimensional state monitoring, characterized in that, The method comprises: acquiring, in real time, a vibration frequency of a running area of a large mechanical equipment, a state of charge value of each battery module in a lithium battery energy storage system in the running area, a module temperature of a series module and a surface bulge area of the series module in the battery module, and a module current of a parallel module and a surface coating crack area of the parallel module in the battery module; determining a plurality of pending modules according to the state of charge value and a preset abnormality determination threshold value; determining a plurality of risk modules according to the module temperature of the series module in each of the pending modules and the module current of the parallel module in the same electrical loop; determining a plurality of abnormal modules according to a change feature of the surface bulge area of the series module and the surface coating crack area of the parallel module in each of the battery modules within a next preset abnormality determination time length and the risk modules; determining a change fluctuation degree according to a number of the series module and the parallel module in the same electrical loop in the abnormal modules within a next preset adjustment determination time length, and adjusting the preset abnormality determination threshold value according to the change fluctuation degree, a number of all the abnormal modules within the same preset adjustment determination time length, a number of all the battery modules, and the vibration frequency; issuing a safety alarm for the abnormal modules re-determined after the preset abnormality determination threshold value is adjusted. 2.The lithium battery energy storage system safety protection method based on multi-dimension state monitoring of claim 1, wherein, The process of determining a plurality of pending modules according to the state of charge value and a preset abnormality determination threshold value comprises: acquiring a duration when the state of charge value is greater than a preset state of charge threshold value to obtain an abnormality duration; determining the battery module as the pending module when the abnormality duration is greater than the preset abnormality determination threshold value. 3.The lithium battery energy storage system safety protection method based on multi-dimension state monitoring of claim 2, wherein, The process of determining a plurality of risk modules according to the module temperature of the series module in each of the pending modules and the module current of the parallel module in the same electrical loop comprises: labeling the parallel module as a temporary module when the module current is greater than a preset module current threshold value; determining a temperature threshold value according to the module current of each of the temporary modules and a preset current temperature table; determining a plurality of risk modules according to the module temperature of the series module in the same electrical loop as the temporary module within a next preset risk determination time length and the corresponding temperature threshold value.

4. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 3, wherein, The process of determining a plurality of risk modules according to the module temperature of the series module in the same electrical loop as the temporary module within a next preset risk determination time length and the corresponding temperature threshold value comprises: calculating a difference value of the module temperature and the temperature threshold value within the preset risk determination time length to obtain a plurality of expected temperature deviations; acquiring a duration when the expected temperature deviation is greater than a preset temperature deviation threshold value to obtain a risk duration, and determining the series module and the temporary module in the same electrical loop as the risk module when the risk duration is greater than a preset risk duration threshold value to determine a plurality of risk modules.

5. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 4, wherein, The process of determining a plurality of abnormal modules according to a change feature of the surface bulge area of the series module and the surface coating crack area of the parallel module in each of the battery modules within a next preset abnormality determination time length and the risk modules comprises: Calculate the average of all surface bulge areas at initial time and each time within the preset abnormality determination duration to obtain a plurality of bulge area averages, and calculate the standard deviation of all bulge area averages to obtain a bulge sliding fluctuation value; Calculate the average of all surface coating crack areas between initial time and each time within the same preset abnormality determination duration to obtain a plurality of crack area averages, and calculate the standard deviation of all crack area averages to obtain a crack sliding fluctuation value; Determine a plurality of abnormal modules according to the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module.

6. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 5, wherein, The process of determining a plurality of abnormal modules according to the bulge sliding fluctuation value, the crack sliding fluctuation value, and the risk module includes: When the bulge sliding fluctuation value is greater than a preset bulge sliding threshold, determining that the battery module is a target module; When the crack sliding fluctuation value is greater than a preset crack sliding threshold, determining that the battery module is a target module; The intersection of all target modules and all abnormal modules is marked as an abnormal module to determine a plurality of abnormal modules.

7. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 6, wherein, The process of determining the change stability according to the number of series modules and parallel modules in the same electrical loop in the abnormal module within the next preset adjustment determination duration includes: Calculate the ratio of the number of series modules and parallel modules in the same electrical loop to the number of abnormal modules at each time within the preset adjustment determination duration to obtain a plurality of adjustment proportions; Determine the change stability according to all adjustment proportions.

8. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 7, wherein, The process of determining the change stability according to all adjustment proportions includes: Calculate the absolute value of the difference of the adjustment proportions between any two adjacent times within the preset adjustment determination duration to obtain a plurality of proportion change amplitudes; Calculate the standard deviation of all proportion change amplitudes to obtain the change fluctuation degree. 9.The lithium battery energy storage system safety protection method based on multi-dimension state monitoring of claim 8, wherein, The process of adjusting the preset abnormality determination threshold according to the change fluctuation degree, the number of all abnormal modules within the same preset adjustment determination duration, the number of all battery modules, and the vibration frequency includes: When the change fluctuation degree is greater than a preset change fluctuation degree threshold, calculate the ratio of the number of all abnormal modules to the number of all battery modules at each time within the same preset adjustment determination duration to obtain a plurality of abnormal proportions; Adjust the preset abnormality determination threshold according to the abnormal proportion and the vibration frequency.

10. The multi-dimensional state monitoring based lithium battery energy storage system safety protection method of claim 9, wherein, The process of adjusting the preset abnormality determination threshold according to the abnormal proportion and the vibration frequency includes: Calculate the standard deviation of the vibration frequency from the initial time to each time within the preset adjustment determination duration to obtain a plurality of vibration sliding fluctuation values; Calculate the standard deviation of the abnormal proportion from the initial time to each time within the same preset adjustment determination duration to obtain a plurality of abnormal sliding fluctuation values; Calculate the Pearson correlation coefficient of all vibration sliding fluctuation values and all abnormal sliding fluctuation values to obtain an adjustment determination coefficient; When the adjustment determination coefficient is greater than a preset adjustment determination standard value, increase the preset abnormality determination threshold according to the relative deviation of the adjustment determination coefficient and the preset adjustment determination standard value.

Images