New energy storage cabinet thermal runaway early warning method and system

Abstract

The invention relates to the technical field of thermal runaway early warning, and particularly discloses a new energy energy storage cabinet thermal runaway early warning method and system, and the system comprises a grouping module which is used for calculating the combination value of each energy storage box according to environment and position data, grouping cabinet bodies meeting conditions into battery clusters, and connecting each cluster to an edge composite module; the edge composite module is composed of an acquisition unit, an analysis module and an interconnection unit; the acquisition unit collects real-time environment data of the battery cluster; the analysis module evaluates the thermal runaway risk and generates an analysis result; the interconnection module exchanges data with the adjacent composite module, generates a real-time analysis report and uploads the real-time analysis report to the central processing module; the central processing module is used for checking the rationality of the analysis result and making a thermal runaway early warning decision based on a real-time analysis report; and a data transmission and interaction mechanism among the modules can timely adopt prevention and control measures such as power-off cooling and the like aiming at the thermal runaway risk, so that the fire occurrence probability is effectively reduced, and safe and stable operation of the new energy storage system is protected.

Description

Technical Field

[0001] This invention relates to the field of thermal runaway early warning technology, specifically to a method and system for early warning of thermal runaway in a new energy storage cabinet. Background Technology

[0002] In existing technologies, to ensure the accuracy of data transmission, signal transmission is often performed using null values. This causes a certain delay in both transmission efficiency and content. For new energy storage devices, battery thermal runaway is the key control path from smoke to fire. Therefore, research on battery thermal runaway and timely measures to cut off power and cool down after determining that the battery has thermal runaway will play a more decisive role in suppressing fire and explosion. Furthermore, most existing sensors use fixed values—that is, pre-set relatively fixed values ​​and combinations of judgments—to trigger alarms. This method is relatively delayed in the field of new energy storage, especially in the early thermal runaway of batteries. In real-world scenarios, the following issues arise: lack of sensitivity: differences in aging between new and old batteries lead to different thermal runaway thresholds, causing older batteries to run away prematurely without triggering an alarm in time; poor environmental adaptability, as fixed thresholds cannot adapt to changes in ambient temperature and humidity, and temperature differences between winter and summer, as well as between different regions, increase the false alarm rate; lack of operational condition dependence, as the temperature rise rate during rapid charging and discharging is higher than normal, leading to false alarms of thermal runaway; limitations of single parameters, ignoring the coupling of multiple factors increases the probability of system-level false alarms; delayed response, missing the optimal period for thermal runaway intervention; and fixed thresholds only trigger in the middle and late stages of thermal runaway, missing the golden period for early intervention. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for early warning of thermal runaway in a new energy storage cabinet, and to solve the following technical problems.

[0004] The objective of this invention can be achieved through the following technical solutions: A thermal runaway early warning system for a new energy storage cabinet includes: Grouping module: Acquires environmental and location data of each energy storage box, obtains environmental and location characteristics of each energy storage box based on the environmental and location data, and obtains the combination value of each energy storage box. Several energy storage boxes whose combination values ​​satisfy the constraints are divided into a battery cluster, and energy storage boxes belonging to the same battery cluster are connected to the same edge composite module. Edge composite module: includes a data acquisition unit, an analysis module, and an interconnection unit; the data acquisition unit acquires real-time environmental data of the battery cluster; the analysis module analyzes the real-time environmental data to determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtains the analysis result data of the battery cluster; The interconnection unit identifies adjacent composite modules, transmits the analysis result data to the adjacent composite modules, and receives adjacent data from the adjacent composite modules; based on the analysis result data and adjacent data, it generates a real-time analysis report and transmits the real-time analysis report to the central processing module. Central processing module: Verifies the rationality of the analysis results based on adjacent data in the real-time analysis report. If the data is rational, it implements a thermal runaway early warning based on the real-time analysis report.

[0005] As a further aspect of the present invention: the process of obtaining the environmental and location characteristics of the energy storage box includes: The indicator data in the environmental data are standardized to eliminate the influence of dimensions. The indicator data of the energy storage box at each time point within a preset time period are converted into feature vectors, which are denoted as indicator vectors. The multidimensional environmental feature vector of the energy storage box is obtained from the indicator vectors of each indicator data. The multidimensional environmental feature vector is reduced in dimensionality based on principal component analysis to obtain the environmental features of the energy storage box. The location data of the energy storage box is standardized to eliminate the influence of dimensions, thereby obtaining the location characteristics of the energy storage box.

[0006] As a further aspect of the present invention: the process of obtaining the combined value of the energy storage box includes: For any two energy storage boxes, obtain their environmental and location characteristics to get the combined value of the two energy storage boxes. E1 and E2 are the environmental characteristics of the two energy storage boxes, P1 and P2 are the location characteristics of the two energy storage boxes, K1 and K2 are weighting coefficients, and K1 = 1 - K2.

[0007] As a further aspect of the present invention: in the process of dividing a number of energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster, the constraints are that the combination value between any two energy storage boxes in each energy storage box is less than a preset combination threshold.

[0008] As a further aspect of the present invention: the process by which the analysis module analyzes the real-time environmental data includes: Obtain sample charge and discharge data of the energy storage box, wherein the sample charge and discharge data is the environmental data of the energy storage box at each power value during the charge and discharge cycle, and generate the temperature rise curve of the energy storage box based on the sample charge and discharge data. A regional heat diffusion model is established based on a thermodynamic model. The real-time temperature data is input into the regional heat diffusion model, and the configuration parameters of the regional heat diffusion model are adjusted to obtain a heat diffusion simulation model. The analysis module obtains the energy value of the energy storage box at the next time stamp, and obtains the predicted temperature of the energy storage box at the next time stamp based on the temperature rise curve; the analysis module also obtains the weather data of the location coordinates of the energy storage box at the next time stamp to obtain the predicted environmental data; the predicted environmental data and the predicted temperature are input into the heat diffusion simulation model to obtain the predicted temperature of the energy storage box at the next time stamp.

[0009] As a further aspect of the present invention: the process of determining whether there is a risk of thermal runaway in each energy storage tank within the battery cluster includes: Several energy storage boxes with different battery health levels are selected and denoted as energy storage boxes under test. The critical thermal runaway temperature of each energy storage box under test is obtained. The critical thermal runaway temperature is the lowest self-temperature value of the energy storage box that has the risk of thermal runaway. The energy storage boxes with different battery health levels and their corresponding critical thermal runaway temperatures are fitted to obtain the critical temperature benchmark model. Obtain the current battery health status of the energy storage box, and record it as the current battery health status. Input the current battery health status into the heat diffusion simulation model to simulate the temperature that the energy storage box needs to reach when the surface temperature of the energy storage box reaches the current battery health status. Record it as the warning temperature threshold. If the predicted temperature of the energy storage box at the next time stamp is greater than or equal to the warning temperature threshold, then the energy storage box is at risk of thermal runaway, and the warning level of the energy storage box is determined; if the predicted temperature is less than the warning temperature threshold, then the energy storage box is not at risk of thermal runaway.

[0010] As a further aspect of the present invention, the process of analyzing whether the result data is reasonable includes: Acquire adjacent data, including real-time environmental data from the adjacent data, and record it as adjacent environmental data; acquire the location coordinates of adjacent composite modules, and record them as adjacent points; simulate the environmental data at the adjacent points based on the heat diffusion simulation model, and record it as simulated environmental data; if the simulated environmental data of the adjacent points corresponding to all adjacent data are consistent with the real-time environmental data, then the analysis results are reasonable.

[0011] A method for early warning of thermal runaway in a new energy storage cabinet includes the following steps: Step S1: Obtain environmental and location data for each energy storage box. Based on the environmental and location data, obtain the environmental and location characteristics of each energy storage box and obtain the combination value of each energy storage box. Divide several energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster. Step S2: Collect real-time environmental data of the battery cluster, analyze the real-time environmental data, determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtain the analysis result data of the battery cluster; identify adjacent battery clusters, transmit the analysis result data to the adjacent battery clusters, and receive the adjacent data of the battery clusters; generate a real-time analysis report based on the analysis result data and the adjacent data. Step S3: Verify the rationality of the analysis results based on the adjacent data in the real-time analysis report. If the data is rational, implement thermal runaway early warning based on the real-time analysis report.

[0012] The beneficial effects of this invention are: This invention offers several significant advantages. Regarding early warning timeliness, by leveraging the edge composite module for rapid local acquisition and analysis of real-time environmental data, the risk of thermal runaway in the energy storage tank can be determined in an extremely short time. This completely eliminates the signal lag problem caused by traditional polling mechanisms. Where polling thousands of instructions for a single cabinet previously took several seconds, this solution can complete signal result transmission in 5 nanoseconds, greatly shortening the time from data acquisition to risk assessment and gaining a valuable advantage in preventing thermal runaway.

[0013] Regarding the accuracy of early warnings, the central processing module verifies the analysis results based on adjacent data in the real-time analysis report, avoiding misjudgments caused by individual edge modules due to sensor failures, local environmental interference, or other factors. For example, if an energy storage box is misjudged as having a risk of thermal runaway, but the data from adjacent cabinets is normal, the central module can make a comprehensive judgment to eliminate the false alarm, ensuring the accuracy and reliability of the early warning. If adjacent cabinets also show similar anomalies, the risk assessment can be strengthened, making the early warning more consistent with the actual situation.

[0014] From the perspective of system management efficiency, the grouping module rationally divides battery clusters based on environmental and location data of the energy storage boxes, grouping energy storage boxes with strong risk correlations together and managing them under the same edge composite module. This achieves refined and efficient management of large-scale energy storage equipment, reduces management complexity, and improves overall management efficiency. Simultaneously, the data transmission and interaction mechanisms between modules ensure smooth information flow, enabling timely implementation of preventative measures such as power outages and cooling to address thermal runaway risks, effectively reducing the probability of fires and safeguarding the safe and stable operation of the new energy storage system. Attached Figure Description

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart illustrating a thermal runaway early warning system for a new energy storage cabinet according to the present invention. Figure 2 This is a schematic diagram of the structure of a fire control and explosion suppression integrated detector for a new energy storage box in a thermal runaway early warning system for a new energy storage cabinet according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0018] Currently, existing technologies in the field of new energy storage do not have dedicated integrated fire and explosion suppression detectors. Most of them use general-type single detectors suitable for the site. Their main function is to transmit the detection results, such as VOC detection, carbon monoxide detection, and smoke detection, to the central alarm host through I / O signals. The central alarm host then issues unified alarms based on the detection results of the equipment area.

[0019] Traditional detection devices mostly use a single sensor or a combination of multiple sensors into one sensor, then transmit independent I / O signals to the alarm host, or use methods such as RS-485 polling mechanisms for signal transmission. However, in the field of energy storage, especially new energy storage facilities, most are composed of multiple levels and a large number of sensors, such as box-level, warehouse-level, and cluster-level sensors. A single energy storage device can often have more than 200 composite sensors, and if it is a multi-in-one device, such as a 5-in-1 device (e.g., smoke sensor, hydrogen sensor, carbon monoxide sensor, temperature and humidity sensor, VOC combustible gas sensor), the number of sensors can reach thousands. In the construction of such systems, due to the large number of sensors, especially at the cluster-level sensing level, the communication mechanism used is very important. Currently, whether it is an I / O mechanism or an RS-485 mechanism, or The MODBUS mechanism employs a similar polling mechanism. Assuming a polling time of approximately 5 milliseconds per instruction, it would take about 5 seconds to process thousands of instructions for a single energy storage unit. This is just for a single energy storage unit. If an energy storage station is in use, monitoring of up to hundreds of energy storage stations would also be accomplished using a MODBUS-like master-slave polling mechanism. More importantly, in existing technologies, to ensure the accuracy of data transmission, null value transmission is often used for signal transmission. This causes a certain delay in both transmission efficiency and content. For new energy storage devices, battery thermal runaway is the key control path from smoke to fire. Therefore, researching battery thermal runaway and taking timely measures to cut off power and cool down after determining that the battery has thermal runaway will play a more decisive role in suppressing fire and explosion. This is quite different from traditional fire protection methods.

[0020] Please see Figure 1As shown, based on this application scenario, the present invention provides a thermal runaway early warning system for a new energy storage cabinet, comprising: Grouping module: Acquires environmental and location data of each energy storage box, obtains environmental and location characteristics of each energy storage box based on the environmental and location data, and obtains the combination value of each energy storage box. Several energy storage boxes whose combination values ​​satisfy the constraints are divided into a battery cluster, and energy storage boxes belonging to the same battery cluster are connected to the same edge composite module. In a preferred embodiment of the present invention, the environmental data includes several indicator data of the location of the energy storage box within a preset time period, the indicator data including temperature data, airflow data, smoke data, humidity data and carbon monoxide concentration, and the location data is the location coordinates of the location of the energy storage box. In a preferred embodiment of the present invention, the process of obtaining the environmental and location characteristics of the energy storage box includes: The indicator data in the environmental data are standardized to eliminate the influence of dimensions. The indicator data of the energy storage box at each time point within a preset time period are converted into feature vectors, which are denoted as indicator vectors. The multidimensional environmental feature vector of the energy storage box is obtained from the indicator vectors of each indicator data. The multidimensional environmental feature vector is reduced in dimensionality based on principal component analysis to obtain the environmental features of the energy storage box. The location data of the energy storage box is standardized to eliminate the influence of dimensions, and the location characteristics of the energy storage box are obtained. In a preferred embodiment of the present invention, the process of obtaining the combined value of the energy storage box includes: For any two energy storage boxes, obtain their environmental and location characteristics to get the combined value of the two energy storage boxes. E1 and E2 are the environmental characteristics of the two energy storage boxes, P1 and P2 are the location characteristics of the two energy storage boxes, K1 and K2 are weighting coefficients, and K1 = 1 - K2. In a preferred embodiment of the present invention, during the process of dividing a number of energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster, the constraints are that the combination value between any two energy storage boxes is less than a preset combination threshold. It should be noted that the energy storage cabinet in this invention is a cabinet, which is divided into several battery clusters. Each battery cluster consists of several energy storage boxes. The number of energy storage boxes in a battery cluster is generally 8, and each energy storage box generally contains 104 batteries. Edge composite module: includes a data acquisition unit, an analysis module, and an interconnection unit; the data acquisition unit acquires real-time environmental data of the battery cluster; the analysis module analyzes the real-time environmental data to determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtains the analysis result data of the battery cluster; The interconnection unit identifies adjacent composite modules, transmits the analysis result data to the adjacent composite modules, and receives adjacent data from the adjacent composite modules; based on the analysis result data and adjacent data, it generates a real-time analysis report and transmits the real-time analysis report to the central processing module. Furthermore, the edge composite module in this invention performs signal preprocessing, that is, data with changes is transmitted, while data without changes is encrypted and restored by default at the terminal. This ensures that only valid and useful data is transmitted with minimal data transmission. These two measures directly improve the efficiency of signal transmission and solve the problem of slow speed in traditional solutions. This is very important in the field of new energy storage fire protection, especially in the field of preventing thermal runaway. In a preferred embodiment of the present invention, the process of the acquisition unit acquiring environmental data is based on several sensing devices, including temperature and humidity sensing devices, airflow sensing devices, smoke sensing devices and carbon monoxide concentration sensing devices. In a preferred embodiment of the present invention, the process by which the analysis module analyzes the real-time environmental data includes: Obtain sample charge and discharge data of the energy storage box, wherein the sample charge and discharge data is the environmental data of the energy storage box at each power value during the charge and discharge cycle, and generate the temperature rise curve of the energy storage box based on the sample charge and discharge data. A regional heat diffusion model is established based on a thermodynamic model. The real-time temperature data is input into the regional heat diffusion model, and the configuration parameters of the regional heat diffusion model are adjusted to obtain a heat diffusion simulation model. The analysis module obtains the energy value of the energy storage box at the next time stamp, and obtains the predicted temperature of the energy storage box at the next time stamp based on the temperature rise curve; the analysis module also obtains the weather data of the location coordinates of the energy storage box at the next time stamp to obtain the predicted environmental data; the predicted environmental data and the predicted temperature are input into the heat diffusion simulation model to obtain the predicted temperature of the energy storage box at the next time stamp. The process of obtaining the sample charge-discharge data includes: The energy level Q of the battery in the energy storage box is obtained. If the battery is in a charging state, the energy level of the energy storage box is recorded as Q. If the battery is in a discharging state, the energy level of the energy storage box is recorded as -Q. The charging and discharging cycle is the time period consumed for the energy level of the energy storage box to change from Q to -Q. Several energy storage boxes are selected as sample energy storage boxes, and several power value gradient values ​​are set. The self-temperature of the sample energy storage boxes is obtained when the power value is at each power value gradient value during the charge and discharge cycle. The average self-temperature of all sample energy storage boxes at the power value gradient value is obtained and recorded as the average self-temperature of the power value gradient value. The temperature rise curve of the energy storage box is obtained by fitting each power value gradient value and its corresponding average self-temperature. In a preferred embodiment of the present invention, the process of determining whether there is a risk of thermal runaway in each energy storage box within the battery cluster includes: Several energy storage boxes with different battery health levels are selected and denoted as energy storage boxes under test. The critical thermal runaway temperature of each energy storage box under test is obtained. The critical thermal runaway temperature is the lowest self-temperature value of the energy storage box that has the risk of thermal runaway. The energy storage boxes with different battery health levels and their corresponding critical thermal runaway temperatures are fitted to obtain the critical temperature benchmark model. Obtain the current battery health status of the energy storage box, and record it as the current battery health status. Input the current battery health status into the heat diffusion simulation model to simulate the temperature that the energy storage box needs to reach when the surface temperature of the energy storage box reaches the current battery health status. Record it as the warning temperature threshold. If the predicted temperature of the energy storage box at the next time stamp is greater than or equal to the warning temperature threshold, then the energy storage box is at risk of thermal runaway, and the warning level of the energy storage box is determined; if the predicted temperature is less than the warning temperature threshold, then the energy storage box is not at risk of thermal runaway. The process of classifying the warning levels includes: Set a grading threshold t, and set multiple thresholds z1 and z2, where 0 < z1 < z2. Denote the warning temperature threshold as T, and then obtain three warning temperature range thresholds [T, T+z1×t], [T+z1×t, T+z2×t] and [T+z2×t, +∞). If the predicted temperature is within [T, T+z1×t], the energy storage tank is under a Level 3 warning; if the predicted temperature is within [T+z1×t, T+z2×t], the energy storage tank is under a Level 2 warning; if the predicted temperature is within [T+z2×t, +∞), the energy storage tank is under a Level 1 warning. It should be noted that most existing sensors use fixed values—that is, pre-set relatively fixed values ​​and combinations of mutual judgments—to trigger alarms. This method is relatively delayed in detecting early thermal runaway in the field of new energy storage, especially in real-world scenarios. 1. Lack of sensitivity: Differences in aging between new and old batteries lead to different thresholds for thermal runaway, which may cause older batteries to run away prematurely without triggering an alarm in time; 2. Poor environmental adaptability: The fixed threshold cannot adapt to changes in environmental temperature and humidity. Temperature differences in winter, summer, and different regions will increase the false alarm rate. 3. Lack of operational condition dependence: During rapid charging and discharging, the temperature rise rate will be higher than normal, leading to misjudged thermal runaway problems; 4. The limitations of a single parameter and the neglect of multi-factor coupling increase the probability of false alarms at the system level. 5. The response lag will miss the best time for thermal runaway intervention; the fixed threshold is only triggered in the middle and late stages of thermal runaway (CO burst / high temperature), missing the golden period for early intervention (H2 release / accelerated temperature rise). Therefore, the warning temperature threshold in this invention is a dynamic threshold that changes according to the actual situation. In a preferred embodiment of the present invention, the analysis result data includes whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and the warning level of the energy storage box with a risk of thermal runaway; the analysis result data also includes real-time environmental data; In a preferred embodiment of the present invention, the process of determining the adjacent composite modules includes: The current edge composite module is denoted as the current edge composite module, and all edge composite modules other than the current edge composite module are denoted as the remaining edge composite modules. A radius threshold is set, and a circular region is obtained with the current edge composite module as the center and the radius threshold as the radius. The circular region is divided into several sector regions, and the remaining edge composite modules with the closest Euclidean distance to the current edge composite module are obtained in each sector region and denoted as the adjacent composite modules of the current edge composite module. In a preferred embodiment of the present invention, the adjacent data are the analysis result data of adjacent composite modules; Central processing module: Verifies the rationality of the analysis results based on adjacent data in the real-time analysis report. If the data is rational, it implements a thermal runaway early warning based on the real-time analysis report. In a preferred embodiment of the present invention, the process of analyzing whether the result data is reasonable includes: Acquire adjacent data, and acquire real-time environmental data from the adjacent data, denoted as adjacent environmental data; acquire the location coordinates of adjacent composite modules, denoted as adjacent points; simulate the environmental data at the adjacent points based on the heat diffusion simulation model, denoted as simulated environmental data; if the simulated environmental data of the adjacent points corresponding to all adjacent data are consistent with the real-time environmental data, then the analysis results are reasonable. The process of determining whether the simulated environment data is consistent with the real-time environment data includes: Both the simulated environment data and the real-time environment data are standardized, and the difference between the simulated environment data and the real-time environment data is obtained. If the difference is less than or equal to a preset difference range, the simulated environment data is consistent with the real-time environment data; otherwise, the simulated environment data is inconsistent with the real-time environment data. In a preferred embodiment of the present invention, the central processing module further includes a weighted comprehensive risk scoring model, specifically comprising: Define a risk sub-score for each sensor device, calculate the weighted sum of all sensors as the comprehensive risk score, and take into account the parameter change rate to enhance early warning capabilities; A method for early warning of thermal runaway in a new energy storage cabinet includes the following steps: Step S1: Obtain environmental and location data for each energy storage box. Based on the environmental and location data, obtain the environmental and location characteristics of each energy storage box and obtain the combination value of each energy storage box. Divide several energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster. Step S2: Collect real-time environmental data of the battery cluster, analyze the real-time environmental data, determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtain the analysis result data of the battery cluster; identify adjacent battery clusters, transmit the analysis result data to the adjacent battery clusters, and receive the adjacent data of the battery clusters; generate a real-time analysis report based on the analysis result data and the adjacent data. Step S3: Verify the rationality of the analysis results based on the adjacent data in the real-time analysis report. If the data is rational, implement thermal runaway early warning based on the real-time analysis report.

[0021] This invention discloses a method and system for early warning of thermal runaway in new energy storage cabinets, aiming to solve the problems of single monitoring and delayed response of thermal runaway risk in existing technologies.

[0022] This invention deploys a multi-level composite sensor network at the cabinet, box, and cluster levels in an energy storage box to monitor multiple parameters such as carbon monoxide, temperature and humidity, smoke, hydrogen, and VOC combustible gases in real time. Based on digital twin technology, a dynamic parameter model is constructed to achieve comprehensive weight determination of the data from each sensor.

[0023] The core of this invention lies in the use of multi-parameter and multi-model fusion to comprehensively evaluate alarm levels, solving the problem of false alarms or missed alarms that are easy to occur with traditional single sensors. After integrating the information data from all sensors, the most effective alarm data is given through a self-developed algorithm model and edge computing and transmitted to the integrated alarm host. Through a dynamic threshold adjustment mechanism, the gradual process of thermal runaway is accurately identified.

[0024] When the risk index exceeds the preset threshold, the system triggers a multi-level early warning mechanism, including primary alarm (level 3 alarm), intermediate alarm (level 2 alarm) and advanced alarm (level 1 alarm), and automatically executes graded control strategies, such as strengthening ventilation, cutting off charging, activating fire extinguishing, or notifying maintenance personnel.

[0025] Compared to traditional fixed threshold combination judgment methods, this invention dynamically optimizes weights and thresholds through a parameter model, significantly improving the accuracy and response speed of risk assessment, and providing intelligent and proactive protection for the safe operation of new energy storage systems.

[0026] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the invention.

Claims

1. A thermal runaway early warning system for a new energy storage cabinet, characterized in that, include: Grouping module: Acquires environmental and location data of each energy storage box, obtains environmental and location characteristics of each energy storage box based on the environmental and location data, and obtains the combination value of each energy storage box. Several energy storage boxes whose combination values ​​satisfy the constraints are divided into a battery cluster, and energy storage boxes belonging to the same battery cluster are connected to the same edge composite module. Edge composite module: includes a data acquisition unit, an analysis module, and an interconnection unit; the data acquisition unit acquires real-time environmental data of the battery cluster; the analysis module analyzes the real-time environmental data to determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtains the analysis result data of the battery cluster; The interconnection unit identifies adjacent composite modules, transmits the analysis result data to the adjacent composite modules, and receives adjacent data from the adjacent composite modules. Based on the analysis results and adjacent data, a real-time analysis report is generated and transmitted to the central processing module. Central processing module: Verifies the rationality of the analysis results based on adjacent data in the real-time analysis report. If the data is rational, it implements a thermal runaway early warning based on the real-time analysis report.

2. The thermal runaway early warning system for a new energy storage cabinet according to claim 1, characterized in that, The process of obtaining the environmental and location characteristics of the energy storage box includes: The indicator data in the environmental data are standardized to eliminate the influence of dimensions. The indicator data of the energy storage box at each time point within a preset time period are converted into feature vectors, which are denoted as indicator vectors. The multidimensional environmental feature vector of the energy storage box is obtained from the indicator vectors of each indicator data. The multidimensional environmental feature vector is reduced in dimensionality based on principal component analysis to obtain the environmental features of the energy storage box. The location data of the energy storage box is standardized to eliminate the influence of dimensions, thereby obtaining the location characteristics of the energy storage box.

3. The thermal runaway early warning system for a new energy storage cabinet according to claim 1, characterized in that, The process of obtaining the combined value of the energy storage box includes: For any two energy storage boxes, obtain their environmental and location characteristics to get the combined value of the two energy storage boxes. Where E1 and E2 are the environmental characteristics of the two energy storage boxes, P1 and P2 are the location characteristics of the two energy storage boxes, and K1 and K2 are weighting coefficients, with K1 = 1 - K 2。 4. The thermal runaway early warning system for a new energy storage cabinet according to claim 1, characterized in that, In the process of grouping several energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster, the constraints are that the combination value between any two energy storage boxes is less than a preset combination threshold.

5. The thermal runaway early warning system for a new energy storage cabinet according to claim 1, characterized in that, The process by which the analysis module analyzes the real-time environmental data includes: Obtain sample charge and discharge data of the energy storage box, wherein the sample charge and discharge data is the environmental data of the energy storage box at each power value during the charge and discharge cycle, and generate the temperature rise curve of the energy storage box based on the sample charge and discharge data. A regional heat diffusion model is established based on a thermodynamic model. The real-time temperature data is input into the regional heat diffusion model, and the configuration parameters of the regional heat diffusion model are adjusted to obtain a heat diffusion simulation model. The analysis module obtains the energy value of the energy storage box at the next time stamp, and obtains the predicted temperature of the energy storage box at the next time stamp based on the temperature rise curve; the analysis module also obtains the weather data of the location coordinates of the energy storage box at the next time stamp to obtain the predicted environmental data; the predicted environmental data and the predicted temperature are input into the heat diffusion simulation model to obtain the predicted temperature of the energy storage box at the next time stamp.

6. The thermal runaway early warning system for a new energy storage cabinet according to claim 5, characterized in that, The process of determining whether there is a risk of thermal runaway in each energy storage tank within the battery cluster includes: Several energy storage boxes with different battery health levels are selected and denoted as energy storage boxes under test. The critical thermal runaway temperature of each energy storage box under test is obtained. The critical thermal runaway temperature is the lowest self-temperature value of the energy storage box that has the risk of thermal runaway. The energy storage boxes with different battery health levels and their corresponding critical thermal runaway temperatures are fitted to obtain the critical temperature benchmark model. Obtain the current battery health status of the energy storage box, and record it as the current battery health status. Input the current battery health status into the heat diffusion simulation model to simulate the temperature that the energy storage box needs to reach when the surface temperature of the energy storage box reaches the current battery health status. Record it as the warning temperature threshold. If the predicted temperature of the energy storage box at the next time stamp is greater than or equal to the warning temperature threshold, then the energy storage box is at risk of thermal runaway, and the warning level of the energy storage box is determined; if the predicted temperature is less than the warning temperature threshold, then the energy storage box is not at risk of thermal runaway.

7. A thermal runaway early warning system for a new energy storage cabinet according to claim 5, characterized in that, The process of analyzing whether the data results are reasonable includes: Acquire adjacent data, including real-time environmental data from the adjacent data, and record it as adjacent environmental data; acquire the location coordinates of adjacent composite modules, and record them as adjacent points; simulate the environmental data at the adjacent points based on the heat diffusion simulation model, and record it as simulated environmental data; if the simulated environmental data of the adjacent points corresponding to all adjacent data are consistent with the real-time environmental data, then the analysis results are reasonable.

8. A method for early warning of thermal runaway in a new energy storage cabinet, characterized in that, Includes the following steps: Step S1: Obtain environmental and location data for each energy storage box. Based on the environmental and location data, obtain the environmental and location characteristics of each energy storage box and obtain the combination value of each energy storage box. Divide several energy storage boxes whose combination values ​​satisfy the constraints into a battery cluster. Step S2: Collect real-time environmental data of the battery cluster, analyze the real-time environmental data, determine whether there is a risk of thermal runaway in each energy storage box within the battery cluster, and obtain the analysis result data of the battery cluster; identify adjacent battery clusters, transmit the analysis result data to the adjacent battery clusters, and receive the adjacent data of the battery clusters; Based on the analysis results and adjacent data, a real-time analysis report is generated; Step S3: Verify the rationality of the analysis results based on the adjacent data in the real-time analysis report. If the data is rational, implement thermal runaway early warning based on the real-time analysis report.

Images