Temperature control early warning method and system for energy storage battery

Abstract

The invention relates to the technical field of battery temperature control early warning, in particular to an energy storage battery temperature control early warning method and system. The method comprises the following steps: acquiring an internal structure diagram of the energy storage battery; performing internal space layout analysis and inter-unit physical connection analysis on the internal structure diagram of the energy storage battery to construct a three-dimensional battery topological structure model; acquiring real-time temperature monitoring parameters of the energy storage battery; dynamic temperature distribution visual rendering is carried out on the three-dimensional battery topological structure model according to real-time temperature monitoring parameters of the energy storage battery, so that thermal state portraits of a plurality of unit areas are constructed; performing multi-period battery operation simulation on the thermal state portraits of the plurality of unit areas, and performing sliding time window temperature change situation prediction to obtain a temperature change prediction situation of each area; and carrying out continuous abnormal temperature rise trend detection on the temperature change prediction situation of each area so as to generate an abnormal temperature rise trend characteristic. According to the invention, safe and efficient pool temperature control early warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of battery temperature control and early warning, and in particular to a temperature control and early warning method and system for an energy storage battery. Background Art

[0002] As a key component of modern energy systems, energy storage batteries are widely used in power storage, smart grids, mobile power supplies and other fields, carrying important tasks of power conversion and storage. With the rapid development of new energy and energy storage technologies, energy storage batteries are playing an increasingly significant role in power regulation, load balancing and emergency power supply. However, during the long-term and high-frequency charging and discharging process, energy storage batteries often generate a lot of heat due to internal chemical reactions, external load changes and environmental factors. If the battery temperature is not effectively controlled, it will lead to a decline in battery performance, reduced charging and discharging efficiency, and even safety accidents such as battery overheating, fire or explosion.

[0003] Traditional battery temperature control technology mainly relies on simple temperature monitoring and heat dissipation design, and uses physical heat dissipation and temperature control systems to prevent battery overheating. However, these methods usually fail to take into account the temperature variation of batteries under different working conditions and their long-term impact on battery life and safety, and lack sufficient intelligent early warning capabilities, and cannot effectively predict the timing and cause of temperature anomalies, making it difficult to take effective measures in time before abnormal temperature rise occurs.

[0004] With the increasing complexity of energy storage battery application scenarios and the continuous improvement of battery safety requirements, traditional temperature control technology can no longer meet the accuracy and real-time requirements of modern energy storage systems for temperature control warning. Therefore, it is urgent to develop an energy storage battery temperature control warning method that integrates advanced monitoring technology, predictive analysis technology and intelligent warning mechanism. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a temperature control early warning method and system for energy storage batteries to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a temperature control early warning method for an energy storage battery, comprising the following steps:

[0007] Step S1: obtaining an internal structure diagram of an energy storage battery; performing internal space layout analysis and inter-unit physical connection analysis on the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model;

[0008] Step S2: obtaining real-time temperature monitoring parameters of the energy storage battery; performing dynamic temperature distribution visualization rendering on the three-dimensional battery topology structure model according to the real-time temperature monitoring parameters of the energy storage battery, thereby constructing thermal state portraits of multiple unit areas;

[0009] Step S3: performing multi-period battery operation simulation on the thermal state portraits of multiple unit areas, and performing sliding window temperature change trend prediction to obtain the temperature change prediction trend of each area;

[0010] Step S4: Detecting the continuous abnormal temperature rise trend of the temperature change prediction situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics;

[0011] Step S5: performing dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then making a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result;

[0012] Step S6: Obtain the historical work log of the energy storage battery; conduct in-depth mining of the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

[0013] The present invention accurately constructs a three-dimensional battery topology model by analyzing the internal structure of the energy storage battery and the connection relationship between the cells, which provides a basis for subsequent temperature distribution and thermal state monitoring, ensures that the temperature monitoring system accurately maps various areas inside the battery, obtains real-time temperature data, and combines it with the three-dimensional battery topology model for visual rendering to fully display the temperature distribution of each unit area of ​​the battery, so that temperature anomalies can be discovered in time to avoid overheating or uneven cooling of the battery. Through multi-period battery operation simulation, the temperature change trend of the battery under different working conditions is predicted, and combined with the sliding window temperature change trend prediction, the early signs of temperature anomalies are effectively captured, which helps to take preventive measures in advance to prevent battery overheating or performance degradation. Through continuous monitoring of the temperature change trend, the abnormal temperature rise trend of the battery unit is detected in time to avoid battery temperature overheating. The safety hazards caused by high temperature are eliminated. Based on the early warning signals generated by abnormal temperature rise trends, the operators are reminded in real time to take necessary measures, such as adjusting the battery load, starting the cooling system, etc., to ensure that the battery is in a safe operating temperature range. Through the dynamic evolution analysis of the battery temperature rise diffusion, the temperature anomaly source can be more accurately located to diagnose battery faults, which helps to identify battery faults such as short circuit, overheating, unit damage, etc., and reduce the probability of failure. Through in-depth mining of the historical work logs of energy storage batteries, the performance of batteries in different working environments and load conditions can be understood, which helps to identify the historical factors that lead to temperature anomalies. According to the abnormal fault diagnosis results of the battery and the analysis of the historical work logs, the battery's temperature control strategy is adjusted, the temperature control parameters are optimized, and the battery's operating temperature is dynamically adjusted to ensure that the battery works in the safest and most efficient state, further extending the battery life.

[0014] Preferably, step S1 comprises the following steps:

[0015] Step S11: Perform CT scanning on the target energy storage battery to obtain an internal structure diagram of the energy storage battery;

[0016] Step S12: Perform deep visual recognition on the internal structure diagram of the energy storage battery and mark each battery cell;

[0017] Step S13: performing precise positioning calculation of the internal space of each battery cell, and extracting the internal space coordinates of each battery cell;

[0018] Step S14: performing internal space layout analysis according to the internal space coordinates of each battery cell to generate battery space layout features;

[0019] Step S15: performing inter-cell physical connection analysis on each battery cell to obtain a topological connection relationship of the battery cells;

[0020] Step S16: reconstructing the three-dimensional structure of the battery space layout features based on the topological connection relationship of the battery cells to construct a three-dimensional battery topological structure model.

[0021] The present invention uses CT scanning imaging technology to non-destructively obtain detailed structural diagrams of the internal structure of energy storage batteries, including the arrangement, size and shape of battery cells. Compared with traditional disassembly or external scanning methods, CT scanning can provide more accurate and comprehensive internal information of batteries. Through deep visual recognition technology, each battery cell inside the energy storage battery can be automatically and accurately identified. This technology can identify the shape, size and relative position of the battery cell, reduce manual intervention, improve accuracy and efficiency, accurately locate each battery cell, and obtain the spatial coordinates of the battery cell. This is the core step in constructing a three-dimensional topological structure model. Through the precise calculation of the spatial coordinates, errors are eliminated to ensure the spatial accuracy of the model. Accuracy, by analyzing the layout of the spatial coordinates of the battery cells, the relative positions, arrangements and heat flow channels between the battery cells are identified. The generated battery spatial layout characteristics help to understand the working environment and performance of the battery. By analyzing the physical connections between the battery cells, it is clear how each battery cell is connected to each other, including electrical connections and thermal connections. This is crucial for understanding the electrical performance, heat conduction and load balancing inside the battery. According to the spatial layout characteristics and topological connection relationships of the battery cells, the three-dimensional structure is reconstructed to generate a complete and accurate three-dimensional battery topological structure model. This three-dimensional model provides an intuitive and accurate foundation for the subsequent temperature control warning system.

[0022] Preferably, step S2 comprises the following steps:

[0023] Step S21: Acquire real-time temperature monitoring parameters of the energy storage battery based on multiple sensors;

[0024] Step S22: Calculate the internal position of each sensor node one by one according to the multi-sensor to obtain the internal space coordinates of the battery of each sensor node;

[0025] Step S23: performing spatial temperature distribution mining on the real-time temperature monitoring parameters of the energy storage battery according to the internal spatial coordinates of the battery of each sensor node, so as to obtain the temperature distribution characteristics of each battery cell;

[0026] Step S24: dividing the three-dimensional battery topology model into multiple battery unit regions, thereby generating multiple unit region models;

[0027] Step S25: Based on the temperature distribution characteristics of each battery cell, the multiple unit area models are dynamically temperature-distributed and visualized, thereby constructing thermal state portraits of the multiple unit areas.

[0028] The present invention uses multi-sensor technology to synchronously monitor the temperature at various key positions of the battery system. Multiple sensors cover different areas of the battery to ensure that complete temperature data is obtained and avoid inaccurate temperature monitoring caused by single sensor errors or failures. By calculating the internal position of each sensor node, the spatial coordinates of each sensor inside the battery can be accurately determined. This ensures that the temperature data can be correctly mapped to the actual position of the battery and avoids temperature data deviation caused by inaccurate positioning. By mining the spatial temperature distribution, abnormal temperature areas are discovered in advance, which provides an important basis for the early warning system. At the same time, analyzing the temperature characteristics of different units helps to optimize the battery unit layout, cooling system and thermal management design. By dividing the three-dimensional battery topology model into multiple battery unit areas, the battery is divided into several different areas, which is convenient for further analysis of the temperature conditions of each area. This division is based on factors such as the physical structure of the battery, temperature distribution characteristics and load requirements, making the analysis more targeted. By combining the temperature distribution characteristics of each battery unit with the three-dimensional model for dynamic visualization rendering, the thermal state of different areas of the battery is displayed in real time. This thermal state portrait provides an intuitive temperature monitoring view for operators, making it easy to determine which areas have problems such as overheating or uneven temperature. Dynamic thermal status visualization can reveal abnormal temperature areas in advance, helping to identify potential safety risks early, such as overheating of battery cells, large temperature differences, etc. This helps to take timely measures to prevent failures or accidents caused by temperature runaway.

[0029] Preferably, step S25 specifically comprises the following steps:

[0030] Performing real-time temperature distribution change analysis on the temperature distribution characteristics of each battery cell to generate real-time temperature distribution change data;

[0031] Perform discrete fitting of time series temperature values ​​on real-time temperature distribution change data to construct the temperature distribution change curve of each battery cell;

[0032] Perform temperature positioning matching on the temperature distribution change curve of each battery cell and multiple unit area models to obtain the matching temperature curve relationship of each unit area;

[0033] The temperature distribution of multiple unit area models is visualized according to the matching temperature curve relationship of each unit area, so as to obtain the temperature distribution thermodynamic map of multiple unit areas;

[0034] The temperature distribution thermodynamic map of multiple unit areas is dynamically rendered to construct thermal state portraits of multiple unit areas.

[0035] The present invention analyzes the real-time changes of the temperature distribution characteristics of each battery cell and timely captures the fluctuations of the battery temperature under different working conditions. This helps to track the temperature change trend in real time and ensure that the temperature remains within a safe range during the operation of the battery. The real-time temperature data is converted into a smooth temperature change curve through discrete fitting of the time series temperature values. This curve can intuitively display the temperature changes of each battery cell at different time points, which is convenient for analysis and early warning. By performing temperature positioning matching on the temperature change curve of each battery cell and the multi-cell regional model, it is ensured that the temperature change of each region is consistent with the temperature data of the actual environment. This step provides accurate results for subsequent thermodynamic analysis. The positioning and data of the battery system are visualized to generate a temperature distribution thermogram of each unit area, which makes the temperature distribution and change more intuitive. The operator can quickly identify which areas have higher temperatures or larger temperature differences, so as to take timely treatment measures. By dynamically rendering the temperature change of the thermograms of multiple unit areas, the process of temperature distribution change over time is displayed to form a dynamic thermal state portrait. This dynamic image helps the operator to understand the temperature state change trend of the battery system in real time. The dynamic thermal state portrait can display sudden changes or anomalies in the temperature change process, and help to timely discover potential safety hazards, such as local overheating, uneven cooling, etc., and give early warnings and take preventive measures.

[0036] Preferably, step S3 specifically comprises the following steps:

[0037] Step S31: defining multiple operating conditions of the battery pack;

[0038] Step S32: performing multi-period battery operation simulation on thermal state portraits of multiple unit areas according to multiple operating conditions of the battery pack to collect multi-period operation simulation data;

[0039] Step S33: performing temperature change trend analysis on each region of the multi-period operation simulation data to obtain temperature change trend characteristics of each region;

[0040] Step S34: predicting the temperature change situation of the sliding window according to the temperature change trend characteristics of each area to obtain the predicted temperature change situation of each area.

[0041] The present invention defines multiple operating conditions to simulate the performance of the battery pack under different loads, temperatures, working pressures and other conditions. This helps to comprehensively evaluate the thermal management capabilities of the battery pack under a variety of working scenarios and ensure that the temperature control warning system can cope with various complex working conditions in practical applications. Through multi-period simulation, the temperature changes of the battery pack in different time periods and different working conditions can be obtained. These data can reflect the temperature fluctuations of the battery during long-term operation and help to gain a deeper understanding of the thermal management performance of the battery. The multi-period operation simulation can simulate the temperature response of the battery under different working conditions to ensure that the system can provide effective temperature control warnings even under extreme conditions to avoid excessive battery temperature. The risk of high or too low temperature is eliminated by analyzing the temperature change trend of each area of ​​the simulation data one by one, and the temperature fluctuation characteristics of each unit area in different time periods are accurately obtained. This helps to identify which areas have large temperature changes and which areas have temperature hotspots. The sliding window temperature change trend prediction method uses the existing temperature change trend to predict the temperature change in the future. In this way, the impending temperature anomaly can be discovered in advance, providing sufficient response time for the temperature control warning system. Through temperature change prediction, the abnormal temperature trend of battery cells or areas in the future, such as overheating, uneven heat dissipation, etc., can be identified, thereby providing sufficient time for the battery management system to intervene.

[0042] Preferably, the specific steps of step S32 are:

[0043] Based on the real-time temperature monitoring parameters of the energy storage battery, the temperature difference change correlation mining between adjacent cells is carried out for each battery cell to obtain the temperature change relationship between adjacent cells;

[0044] Based on the temperature change relationship between adjacent cells, the heat transfer between battery cells is evolved to obtain the heat transfer law of battery cells;

[0045] Calculate the cooling efficiency of energy storage batteries;

[0046] According to the multiple operating conditions of the battery pack, a multi-operating condition operating characteristic quantitative analysis is performed on the heat transfer law of the battery unit and the cooling efficiency of the energy storage battery, so as to generate battery operating characteristics of different operating conditions;

[0047] Based on the battery operation characteristics of different working conditions, multi-period battery operation simulation is performed on the thermal state portraits of multiple unit areas to collect multi-period operation simulation data.

[0048] The present invention reveals the heat transfer effect between different battery cells by exploring the correlation between the temperature difference changes between adjacent cells, which helps to deeply understand how heat is transferred between cells in the battery pack, especially which cells have a greater impact on temperature changes. Through the temperature change relationship between battery cells, a heat transfer evolution model between battery cells is constructed, which helps to understand and predict the propagation mode and rate of heat between cells in the battery pack, provides a basis for optimizing cooling design, calculates the cooling efficiency, clarifies the performance of the cooling system in the battery pack, and determines whether it can effectively take away the heat of the battery cells and maintain the system temperature within a safe range. By analyzing the heat transfer laws and cooling methods of battery cells under various working conditions, the present invention also provides a basis for optimizing cooling design, and calculates the cooling efficiency. The quantitative analysis of cooling efficiency can evaluate the thermal performance of the battery under different operating environments. This analysis helps to optimize the thermal management strategy of the battery under different working conditions. The analysis under different working conditions helps to understand the thermal response of the battery under extreme conditions (such as high load or low temperature), ensuring that the battery can maintain stable operation under various working conditions and will not cause performance degradation or safety hazards due to temperature problems. Through multi-period simulation, the temperature data of the battery pack under various working conditions in different time periods are collected. Such data provides sufficient information for subsequent temperature control predictions. Simulation under different working conditions can more accurately simulate the various situations encountered by the battery pack in actual operation, thereby improving the adaptability and accuracy of the simulation model.

[0049] Preferably, the specific steps of step S4 are:

[0050] Step S41: Calculating the temperature change rate of each region's predicted temperature change situation to generate a temperature change situation rate for each region;

[0051] Step S42: analyzing the temperature change fluctuation range of each region on the temperature change prediction trend to obtain the temperature change fluctuation range of each region;

[0052] Step S43: identifying the temperature rise deviation based on the temperature change trend rate of each area and the temperature change fluctuation range of each area, and marking the temperature change deviation area;

[0053] Step S44: performing continuous abnormal temperature rise trend detection on the temperature change deviation area based on a preset temperature rise deviation threshold, thereby generating abnormal temperature rise trend features;

[0054] Step S45: generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics.

[0055] By calculating the temperature change rate, the present invention can timely understand the temperature change rate of each area, which helps to quickly identify those areas with rapid temperature changes, facilitate quick response and take measures, and the calculation of the temperature change rate enables the system to capture areas where the temperature changes too fast and timely discover existing overheating or uneven heat dissipation problems, which provides a data basis for taking cooling, temperature reduction or load adjustment measures in advance. By analyzing the temperature change fluctuation range of each area, the amplitude and change trend of temperature fluctuation can be clearly identified, so as to identify areas with large temperature fluctuations or too high temperatures, thereby avoiding the battery from being affected by unstable temperature. The analysis of the temperature change fluctuation range helps to optimize the battery's heat dissipation system. For areas with large fluctuations, cooling is strengthened, radiators are added, etc. to maintain the battery temperature stability. By combining the temperature change rate and the temperature change fluctuation range, areas where the temperature changes deviate from the normal trend can be accurately identified, which helps to discover potential temperature as early as possible. Abnormalities can be detected to avoid failures caused by local overheating. By marking the temperature change deviation area, the problem can be concentrated in a specific area, which is convenient for targeted adjustments and optimizations, such as enhancing cooling or adjusting workloads. By continuously monitoring the temperature change deviation area, the trend of abnormal temperature rise can be tracked in real time to ensure a response before the temperature exceeds the standard. Early detection of the temperature rise trend can help reduce system overheating and safety accidents. The preset temperature rise deviation threshold helps set a warning line. Once the temperature changes beyond the set range, the system can immediately generate an alarm signal and automatically trigger temperature control measures to avoid temperature out of control. By generating temperature abnormality warning signals, the system can automatically detect and prompt operators to take measures, reduce reliance on human intervention, and improve the system's response speed. The temperature abnormality warning signal helps operators deal with problems such as excessive temperature rise or uneven temperature in a timely manner when the battery pack has problems, preventing the problem from deteriorating to the point of causing equipment damage or danger.

[0056] Preferably, the specific steps of step S5 are:

[0057] Step S51: locating the abnormal temperature starting point based on the abnormal temperature warning signal, thereby marking the abnormal temperature rise starting unit;

[0058] Step S52: analyzing the abnormal temperature rise evolution path of the abnormal temperature rise starting unit based on the three-dimensional battery topology model to generate multiple abnormal temperature rise evolution paths;

[0059] Step S53: performing dynamic temperature rise diffusion evolution based on multiple abnormal temperature rise evolution paths, thereby generating a diffusion prediction trend of the abnormal starting unit;

[0060] Step S54: making a battery abnormal fault diagnosis decision based on the diffusion prediction trend of the abnormal starting cell to obtain a battery abnormal fault diagnosis result.

[0061] The present invention can accurately identify the starting unit of abnormal temperature rise through the temperature abnormality warning signal, ensuring that the problem can be discovered and handled in time from the source. This step helps to avoid the spread of temperature rise due to failure to identify the abnormality in time, affecting the entire battery pack. Marking the starting unit of abnormal temperature rise provides a clear target for subsequent thermal management and fault diagnosis, which can quickly track the problem area, reduce unnecessary intervention, and optimize processing efficiency. By analyzing the evolution path of the starting unit of temperature rise, we can deeply understand how the temperature abnormality propagates inside the battery pack, which units are affected, and which areas become hot spots of temperature rise, which helps to predict the thermal response of the entire system. Diffusion prediction based on multiple abnormal temperature rise evolution paths can predict the further expansion trend of abnormal temperature rise, helping The management system accurately grasps the dynamic changes of temperature evolution, which provides a scientific basis for subsequent temperature control decisions. Through the diffusion prediction trend, it can determine in advance which areas will have temperature abnormalities in the future, and adjust the cooling system or battery load in time to avoid further deterioration of the problem and enhance the safety and reliability of the system. By combining the diffusion prediction trend with the battery fault diagnosis model, automated fault identification and processing decisions are achieved. The system can automatically determine whether abnormal temperature rise will cause a fault and generate an alarm in advance to avoid human delays. According to the fault diagnosis results, it can accurately determine whether the battery is facing a failure risk and help operation and maintenance personnel take timely preventive measures, such as shutting down some modules, adjusting the cooling system, reducing the battery load, etc., thereby avoiding more serious failures.

[0062] Preferably, the specific steps of step S6 are:

[0063] Step S61: Obtaining the historical work log of the energy storage battery;

[0064] Step S62: performing multiple operation scenario identification on the energy storage battery historical work log to obtain multiple battery operation scenarios;

[0065] Step S63: performing in-depth mining of the load state of each of the multiple battery operation scenarios, thereby generating a load state feature of each operation scenario;

[0066] Step S64: dynamically transfer learning the load state characteristics of each operation scenario to generate a deep representation of the battery load state;

[0067] Step S65: Optimize dynamic temperature control parameters according to the deep characterization of battery load status and the abnormal battery fault diagnosis results to build a dynamic battery temperature control engine to execute battery temperature control warning and temperature control parameter decision.

[0068] The present invention identifies different battery operation scenarios and understands the performance of the battery under different environmental and load conditions. This helps to build a comprehensive model covering various operating conditions experienced by the battery (such as high load, low temperature, high temperature, etc.). This scenario recognition helps to evaluate the performance of the battery under diverse operating conditions and ensures that the battery can maintain an efficient and safe operating state under various actual operating conditions. By deeply mining the load state of each scenario, the load characteristics and temperature control requirements of each scenario can be accurately identified. Under different operating loads, the temperature distribution and heat dissipation requirements of the battery will be different. This step helps to accurately locate the thermal management requirements under different operating conditions. Dynamic transfer learning allows the model to extract experience from past scenarios and migrate it to new operating scenarios, which helps the battery in the face of Under different loads and working conditions, it can make adaptive adjustments to improve the intelligence level of the system. Based on the deep characterization of load status and fault diagnosis results, the system can dynamically adjust the temperature control parameters and optimize the battery's heat dissipation, cooling and load management strategies, thereby minimizing the occurrence of temperature anomalies and ensuring the temperature stability of the battery under different load conditions. The dynamic temperature control engine can monitor the battery temperature in real time, and adjust the temperature control strategy in time according to load changes, providing automated temperature control warnings and decisions, which helps to improve the safety of the energy storage system and reduce battery failures caused by temperature control problems. The dynamic optimization of temperature control parameters ensures that the battery can always maintain the best working state. According to changes in load status, the battery's temperature control system responds at any time to extend the battery life and improve the efficiency of the system.

[0069] In this specification, a temperature control warning system for an energy storage battery is provided, which is used to execute the temperature control warning method for an energy storage battery as described above, including:

[0070] The topology module is used to obtain the internal structure diagram of the energy storage battery; analyze the internal space layout and the physical connection between units of the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model;

[0071] Visualization rendering module, used to obtain real-time temperature monitoring parameters of energy storage batteries; perform dynamic temperature distribution visualization rendering of the three-dimensional battery topology model according to the real-time temperature monitoring parameters of the energy storage batteries, thereby constructing thermal state portraits of multiple unit areas;

[0072] The multi-period simulation module is used to perform multi-period battery operation simulation on the thermal state portraits of multiple unit areas and to predict the temperature change trend of the sliding window to obtain the predicted temperature change trend of each area;

[0073] The abnormal warning module is used to continuously detect the abnormal temperature rise trend of the temperature change forecast situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating temperature abnormality warning signals based on the abnormal temperature rise trend characteristics;

[0074] A diagnosis decision module is used to perform dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then make a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result;

[0075] The dynamic temperature control optimization module is used to obtain the historical work log of the energy storage battery; deeply mine the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

[0076] The present invention can accurately construct a three-dimensional topological structure model of the battery by obtaining the internal structure diagram of the energy storage battery and performing a detailed analysis. In this way, the layout and physical connection relationship of each battery cell can be accurately understood, which is helpful for subsequent thermal state analysis and fault location. The physical connection relationship between the battery cells is clarified, and a clear structural view is provided for the thermal management system of the battery, ensuring that the specific layout of the battery module is taken into account during design and optimization, improving the heat dissipation efficiency and performance of the battery, and real-time monitoring and rendering of the temperature distribution of the battery to generate a thermal state portrait, so that the temperature state of the battery is clear at a glance. Through dynamic rendering, the thermal changes of the battery can be observed in real time, and temperature anomalies can be quickly discovered. Through the thermal state portrait, the area and trend of temperature anomalies can be accurately identified, providing an intuitive basis for subsequent temperature control optimization and fault detection, ensuring that temperature changes are responded to in a timely manner. Through multi-period simulation, the battery temperature changes in different operating time periods are simulated, and the temperature fluctuations of the battery in future time periods are predicted. Combined with the temperature change trend prediction of the sliding window, potential temperature control problems can be identified in advance. The temperature prediction information provided by this module is helpful for real-time adjustment of the battery operation strategy (such as load distribution, cooling adjustment, etc.), ensuring that the battery is always maintained within a safe temperature range, reducing the risk of failure, and Real-time monitoring can detect abnormal temperature rise trends in time to ensure that early warnings can be issued before problems occur. This early warning mechanism can effectively avoid battery damage or safety accidents caused by excessive temperature. The identification of abnormal temperature rise trend characteristics enables the early warning system to not only rely on simple temperature thresholds, but also to identify potential fault sources from more complex temperature change patterns, thereby improving the accuracy of early warnings. Based on the early warning signal, through diffusion evolution analysis, it can track the process of temperature anomalies spreading from the source to other areas, accurately determine the root cause of the temperature rise anomaly, and perform fault diagnosis. Through accurate diagnostic decisions, it can quickly identify whether the battery has Faults can be detected and countermeasures can be formulated (such as shutting down certain modules, adjusting loads, increasing cooling, etc.) to prevent further expansion of the faults. Through in-depth mining of historical work logs, the load status characteristics in different scenarios can be identified to provide the necessary adjustment basis for the temperature control system, which helps to dynamically optimize the temperature control parameters and achieve the best operation of the temperature control system under various working conditions. The constructed dynamic battery temperature control engine can automatically adjust the battery temperature control strategy (such as cooling intensity, load distribution, etc.) according to the load, working conditions and fault diagnosis results in real time, thereby effectively avoiding over-high or over-low temperatures, extending battery life and improving system efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 A schematic diagram of the steps of a temperature control and early warning method for an energy storage battery according to the present invention;

[0078] Figure 2 Detailed implementation flow chart of step S1;

[0079] Figure 3 Detailed implementation flow chart of step S2;

[0080] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0081] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0082] The present application example provides a method and system for temperature control and early warning of energy storage batteries. The execution subjects of the method and system for temperature control and early warning of energy storage batteries include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of audio and image management system, information management system, and cloud data management system.

[0083] See also Figures 1 to 4 The present invention provides a temperature control and early warning method for an energy storage battery, and the temperature control and early warning method for an energy storage battery comprises the following steps:

[0084] Step S1: obtaining an internal structure diagram of an energy storage battery; performing internal space layout analysis and inter-unit physical connection analysis on the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model;

[0085] Step S2: obtaining real-time temperature monitoring parameters of the energy storage battery; performing dynamic temperature distribution visualization rendering on the three-dimensional battery topology structure model according to the real-time temperature monitoring parameters of the energy storage battery, thereby constructing thermal state portraits of multiple unit areas;

[0086] Step S3: performing multi-period battery operation simulation on the thermal state portraits of multiple unit areas, and performing sliding window temperature change trend prediction to obtain the temperature change prediction trend of each area;

[0087] Step S4: Detecting the continuous abnormal temperature rise trend of the temperature change prediction situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics;

[0088] Step S5: performing dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then making a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result;

[0089] Step S6: Obtain the historical work log of the energy storage battery; conduct in-depth mining of the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

[0090] The present invention accurately constructs a three-dimensional battery topology model by analyzing the internal structure of the energy storage battery and the connection relationship between the cells, which provides a basis for subsequent temperature distribution and thermal state monitoring, ensures that the temperature monitoring system accurately maps various areas inside the battery, obtains real-time temperature data, and combines it with the three-dimensional battery topology model for visual rendering to fully display the temperature distribution of each unit area of ​​the battery, so that temperature anomalies can be discovered in time to avoid overheating or uneven cooling of the battery. Through multi-period battery operation simulation, the temperature change trend of the battery under different working conditions is predicted, and combined with the sliding window temperature change trend prediction, the early signs of temperature anomalies are effectively captured, which helps to take preventive measures in advance to prevent battery overheating or performance degradation. Through continuous monitoring of the temperature change trend, the abnormal temperature rise trend of the battery unit is detected in time to avoid battery temperature overheating. The safety hazards caused by high temperature are eliminated. Based on the early warning signals generated by abnormal temperature rise trends, the operators are reminded in real time to take necessary measures, such as adjusting the battery load, starting the cooling system, etc., to ensure that the battery is in a safe operating temperature range. Through the dynamic evolution analysis of the battery temperature rise diffusion, the temperature anomaly source can be more accurately located to diagnose battery faults, which helps to identify battery faults such as short circuit, overheating, unit damage, etc., and reduce the probability of failure. Through in-depth mining of the historical work logs of energy storage batteries, the performance of batteries in different working environments and load conditions can be understood, which helps to identify the historical factors that lead to temperature anomalies. According to the abnormal fault diagnosis results of the battery and the analysis of the historical work logs, the battery's temperature control strategy is adjusted, the temperature control parameters are optimized, and the battery's operating temperature is dynamically adjusted to ensure that the battery works in the safest and most efficient state, further extending the battery life.

[0091] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a temperature control and early warning method for an energy storage battery of the present invention. In this example, the steps of the temperature control and early warning method for an energy storage battery include:

[0092] Step S1: obtaining an internal structure diagram of an energy storage battery; performing internal space layout analysis and inter-unit physical connection analysis on the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model;

[0093] In this embodiment, the source of the internal structure diagram is determined. The source includes design drawings, technical manuals, CAD files or existing engineering models provided by the manufacturer. Identify the specific model and specifications of the battery to ensure that the obtained drawings are consistent with the actual battery. If the required internal structure diagram exists in a digital format (such as PDF, CAD file), use professional software (such as AutoCAD, SolidWorks) to open and extract the relevant content. If necessary, contact the manufacturer to obtain a high-definition internal structure diagram to ensure that the details of the drawing are clearly visible. Identify the layout of the battery cells in the internal structure diagram, including the relative positions of the battery cells, conductive connections, cooling pipes and other auxiliary components. Record the size and position relationship of each component for use in three-dimensional modeling. Use computer-aided design (CAD) software to measure and analyze the extracted structure diagram to ensure that the spatial layout of each component meets the design standards. Use the measurement tools in the application software to record the spacing between battery cells, connection methods and layout of the cooling system. Organize the results of the internal space layout analysis to form a layout analysis report, including key dimensions, spacing and layout features, to provide a reference for subsequent modeling. Identify the physical connection between each battery cell in the internal structure diagram, including the electrical connection, mechanical fixation and thermal conduction path of the battery cell. Record the material properties (such as electrical conductivity, thermal conductivity) and design details of the connection parts for each connection method. Use finite element analysis (FEA) software (such as ANSYS, COMSOL) to simulate the connection between battery cells and analyze the strength and thermal conductivity of the connection. Set simulation parameters such as material properties, load conditions and boundary conditions to evaluate the reliability of the connection. Record the performance analysis results of each connection, including important parameters such as electrical impedance, thermal impedance and structural strength. Organize the results of the physical connection analysis and form a report to provide detailed information on the connection characteristics. Select a suitable 3D modeling tool (such as SolidWorks, CATIA, Autodesk Inventor) to build a 3D topological structure model. Build the 3D model step by step according to the internal space layout and the results of the physical connection analysis. First create the geometry of the battery cell, then add connectors and cooling systems. Ensure that each component of the model conforms to the actual size and connection method, laying the foundation for subsequent thermal and electrical analysis. After the model is built, use the software's verification tools to check the model's integrity and accuracy to ensure that there are no missing or incorrect connections. Perform preliminary functional tests to ensure that the model is used for subsequent analysis and optimization. Organize the construction process of the 3D battery topology model, including the internal space layout, physical connection analysis, and various steps of model construction. Generate the final 3D model file and store the analysis results and model file together for subsequent use and modification.

[0094] Step S2: obtaining real-time temperature monitoring parameters of the energy storage battery; performing dynamic temperature distribution visualization rendering on the three-dimensional battery topology structure model according to the real-time temperature monitoring parameters of the energy storage battery, thereby constructing thermal state portraits of multiple unit areas;

[0095] In this embodiment, the real-time temperature monitoring device and its data source are determined, which usually include a sensor network (such as a thermocouple, an infrared sensor) and a data acquisition system (such as a data logger, a PLC system), and the type of monitoring parameters is determined, including the temperature of each battery cell, the ambient temperature, the charge and discharge status, etc. The temperature sensor is connected using data acquisition software (such as LabVIEW, MATLAB), and the temperature data is collected in real time to ensure that the data collection frequency can meet the needs of dynamic analysis, for example, once per second, and the data storage format of the collection system is set (such as CSV, JSON), and the accuracy and completeness of the data are ensured, and the collected real-time temperature data is analyzed. Clean the data, remove outliers and noise, ensure the quality of the data set, use statistical methods (such as median filtering, Z-score outlier detection) for processing, standardize the processed data, ensure the comparability of different sensor data, ensure the accuracy of the three-dimensional battery topology model, the model should include all battery cells and their physical connections, use CAD software (such as SolidWorks, CATIA) to open the model for confirmation, check the scale and unit of the model to ensure that it matches the real-time temperature data, map the acquired real-time temperature monitoring parameters to the three-dimensional model according to the number and position of the battery cell, use Python's NumPy and P The andas library processes the data, generates the temperature distribution matrix for each unit, uses interpolation methods (such as linear interpolation and cubic interpolation) to smooth the temperature data on the surface of the 3D model to obtain a more detailed temperature distribution, uses visualization tools (such as Paraview, MATLAB's surf function, Blender) to dynamically render the 3D model, sets the temperature color mapping (such as heat map) to ensure that different temperature ranges are represented by different colors, sets visualization parameters such as viewing angle, lighting and shadow effects to enhance the visualization effect, makes the temperature distribution clear at a glance, and sets a dynamic update mechanism so that the visualization results can reflect the temperature data in real time. Changes, by writing scripts to regularly read temperature data and update the model, ensure that the visualization interface is user-friendly and can easily switch between different viewing angles and display modes, record the dynamic rendering results as video or animation files for subsequent analysis and display, use tools such as FFmpeg for video encoding to ensure video quality and smoothness, generate static images (such as PNG, JPEG format), which are used to display thermal status portraits in reports and documents, analyze the generated thermal status portraits, identify temperature abnormalities and hotspots, evaluate the thermal management performance of the battery, record the temperature change trend of each unit, and analyze the temperature distribution characteristics under different working conditions to provide a basis for subsequent optimization.

[0096] Step S3: performing multi-period battery operation simulation on the thermal state portraits of multiple unit areas, and performing sliding window temperature change trend prediction to obtain the temperature change prediction trend of each area;

[0097] In this embodiment, based on the constructed three-dimensional battery topology model and real-time thermal state portrait, a suitable thermal simulation software (such as ANSYS Fluent, COMSOL Multiphysics) is selected for multi-period thermal simulation. Determine the simulation parameters, including the thermophysical properties of the battery (such as thermal conductivity, specific heat capacity), environmental conditions (such as external temperature, wind speed) and working conditions (such as charge and discharge rate, load change). According to the working cycle of the battery, set the simulation period (such as 1 hour, 2 hours), and subdivide each period (such as performing a simulation calculation every 10 minutes). Record the battery state changes in each period, such as power output and ambient temperature changes during charging and discharging. Start the simulation software, enter the set initial conditions and boundary conditions, and run multi-period thermal simulation. Ensure that the software can handle complex heat conduction and convection problems and accurately simulate the temperature distribution inside the battery. Monitor the calculation progress during the simulation process to ensure that the temperature changes in each unit area can be updated and recorded in time. Select a suitable time series prediction model (such as ARIMA, LSTM, SVR) to predict the temperature change trend of the sliding window. Determine the input characteristics and output targets of the model according to the data characteristics and prediction requirements. Determine the settings of the sliding window, such as the window size (e.g., 10 minutes, 30 minutes) and the sliding step size (e.g., 1 minute, 5 minutes), to achieve high-frequency temperature prediction. Preprocess the recorded temperature change data, including normalization, detrending, and seasonal decomposition, to improve the accuracy of the prediction model. Generate training and test sets to ensure the representativeness and diversity of the data for model training and evaluation. Use the training set to train the prediction model and adjust the model hyperparameters (e.g., learning rate, regularization term) to optimize performance. Use the cross-validation method to evaluate the performance of the model on the test set to ensure the generalization ability of the model on different data sets. Apply the sliding window prediction model to real-time temperature data to generate temperature change prediction trends for future periods. Update the model input every time new data arrives and calculate future temperature changes in real time. Record the temperature change prediction results for each region and compare them with the actual monitoring data to analyze the accuracy and reliability of the prediction. Use visualization tools (e.g., Matplotlib, Tableau) to generate visualization charts of the temperature change prediction trend to show the temperature change trend and warning threshold of each region. Organize temperature change prediction results and analysis reports, record the temperature prediction situation and risk points in each area, and provide a basis for subsequent fault warning and thermal management.

[0098] Step S4: Detecting the continuous abnormal temperature rise trend of the temperature change prediction situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics;

[0099] In this embodiment, the judgment standard of temperature anomaly is set. Generally, the temperature is defined as exceeding the normal operating range (such as being higher than a certain threshold, such as 45°C) or the temperature change rate exceeds a preset value (such as exceeding 2°C per minute) as abnormal temperature rise. A time window (such as 5 minutes, 10 minutes) is determined, and the temperature change is monitored within the window to capture the abnormal trend in a short time. The temperature change prediction situation data of each area is collected, including the real-time temperature value, the predicted value and its change trend, to ensure that the data can reflect the real-time dynamics of the temperature, and the temperature data from different areas are integrated to form a comprehensive data set for subsequent analysis. Suitable anomaly detection algorithms are selected, such as control charts, Z-score methods or machine learning models (such as Isolation Forest, One-Class SVM) is used to detect the temperature rise trend. The control chart method is used to draw a control chart of temperature changes to monitor whether the temperature exceeds the control range in real time. A script (such as Python) is written to implement the anomaly detection algorithm. The temperature data of each area is monitored in real time and a threshold is set. If the data exceeds the threshold, it is marked as abnormal. At the end of each time window, the statistical features such as the temperature average and standard deviation in the window are calculated to facilitate the subsequent abnormal trend analysis. The abnormal temperature rise events detected in each area are recorded, including the occurrence time, duration, and excess amplitude. An abnormal log is generated for subsequent analysis and tracking. According to the monitored abnormal temperature rise events, relevant features are extracted, such as abnormal duration, temperature change rate, abnormal amplitude, etc. Feature vectors are designed, including the combination of these features. Statistical analysis methods are used to analyze the relationship between the abnormal temperature rise trend characteristics and the battery operating status, and their impact on battery safety is evaluated. If there is enough historical data, machine learning methods (such as decision trees and random forests) are used to analyze the abnormal trend features. The model is trained to identify potential abnormal patterns, training sets and test sets are set, and the accuracy of the model is evaluated through cross-validation. Based on the abnormal temperature rise trend characteristics, the trigger threshold of the temperature abnormality warning signal is set. For example, a warning signal is issued when the temperature in a certain area exceeds the set threshold for 5 consecutive minutes. A multi-level warning mechanism is set, such as green (normal), yellow (caution), red (critical), etc., to reflect different risk levels. An early warning system is developed to monitor the temperature changes in each area in real time and generate corresponding early warning signals according to the set thresholds. Alarm systems, SMS notifications, or visual dashboards are used to ensure that the early warning system can respond in a timely manner, and detailed information of each warning event, including time, area, temperature changes, etc., is recorded. The performance of the early warning system is regularly checked, the accuracy and timeliness of the warning signal are analyzed, feedback data is collected, the early warning mechanism is optimized, the probability of false alarms and missed alarms is reduced, a summary report is formed, the entire process of abnormal temperature rise trend detection, feature generation, and early warning signal generation is recorded, and improvement suggestions are put forward.

[0100] Step S5: performing dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then making a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result;

[0101] In this embodiment, a suitable heat diffusion model is selected, such as a one-dimensional heat conduction equation or a more complex two-dimensional or three-dimensional heat diffusion model, to simulate the diffusion process of temperature anomalies, and the parameters of the model are determined, including thermal conductivity, specific heat capacity, ambient temperature, initial temperature distribution, etc. These parameters are set according to the physical properties of the battery material and experimental data, and the boundary conditions (such as adiabatic and constant temperature) are set according to the actual working conditions of the battery, and the initial conditions are set to ensure that the model can accurately reflect the working state of the battery. The initial temperature of a certain battery cell in the model is set to the identified abnormal temperature, and the temperature of other cells is set to the normal working temperature. Numerical simulation software (such as COMSOL Multiphysics, ANSYS) is used to simulate the dynamic temperature rise diffusion evolution, and a suitable solver is selected to ensure that it can handle multi-dimensional heat conduction problems. The simulation is started, and the temperature change over time is calculated step by step to observe how the temperature anomaly diffuses between the battery cells. The temperature distribution of each time step is recorded to ensure that the dynamic process of temperature change can be captured, and the simulation time step (such as per second, per minute) is set. In order to capture the evolution of temperature more accurately, record the simulation results, including the temperature changes of each unit at different time points, generate a temperature distribution map, intuitively display the temperature rise diffusion trend, and output the results in a visual format (such as animation, heat map) for easy analysis and display. Select a suitable fault diagnosis algorithm, such as a model-based method (such as Kalman filtering, state observer) or a data-based method (such as machine learning classifiers, such as support vector machines, decision trees). According to the specific operating status and historical data of the battery, design a suitable fault diagnosis model, extract fault features from the temperature rise diffusion simulation results, such as temperature change rate, abnormal area, duration and temperature peak, etc. These features will be used as input data for fault diagnosis, and standardize the extracted features to ensure that the contribution of different features in the model is balanced. If a machine learning model is used, the model must be trained using historical fault data, and the performance of the model must be evaluated through cross-validation to ensure that the model has good generalization ability. Record the model's accuracy, recall rate, F1 score and other evaluation indicators to evaluate the effectiveness of the model in fault diagnosis.

[0102] Step S6: Obtain the historical work log of the energy storage battery; conduct in-depth mining of the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

[0103] In this embodiment, the storage location of the historical work log of the energy storage battery is determined, including a database (such as SQL Server, MySQL), a file system (such as CSV, JSON format) or cloud storage, and the key data fields in the log are identified, such as charge and discharge status, temperature data, current, voltage, operating time and fault records, etc. The historical work log is exported using a data extraction tool (such as Python script, SQL query), ensuring that the extracted data contains a complete time range, usually covering at least the complete working cycle of the battery, and formulating a standard data extraction format to ensure that the field name and data type are consistent for subsequent analysis. The extracted data is cleaned to remove duplicate records and missing values ​​to ensure the accuracy and completeness of the data. The Pandas library is used for data processing. Key indicators such as temperature, voltage, and current are standardized for subsequent analysis and comparison. Different battery operation scenarios are identified based on the data in the historical work log, such as normal charging and discharging, overload, and low-temperature operation. The characteristic indicators of each scenario, such as current, temperature, and charging and discharging rate, are clarified. For each scenario, a feature extraction method is set. For example, statistical methods such as mean, variance, maximum, and minimum are used to describe the load state. Data mining tools (such as Python's Scikit-learn library) are used to analyze historical data, extract the load state characteristics of each scenario, and use clustering analysis methods (such as K-Means) to group similar scenarios together. For each scenario, a load status characteristic report is generated, including temperature change trends, charge and discharge efficiency, and risk points. The load status characteristics of each scenario are sorted out to form a detailed analysis report, recording the scenario characteristics and the corresponding load status to facilitate subsequent dynamic temperature control parameter optimization. The previous battery abnormal fault diagnosis results are combined with the scenario load status to identify the temperature control requirements in different operating scenarios. Under high load conditions, more stringent temperature control strategies are required. The relationship between battery temperature changes and fault occurrence in each scenario is recorded to help optimize temperature control parameters. According to historical data and fault diagnosis results, optimization goals are set, such as maximizing charge and discharge efficiency, minimizing temperature fluctuations, or reducing the occurrence of faults, and determining dynamic temperature control parameters. range, such as setting temperature control thresholds, cooling rates, and heating rates, selecting appropriate optimization algorithms (such as genetic algorithms and particle swarm optimization) for dynamic temperature control parameter optimization, setting the input (such as current temperature and load status) and output (such as temperature control strategy) of the optimization model, running the optimization algorithm, generating the best dynamic temperature control strategy, ensuring that the battery maintains the optimal operating temperature under different load conditions, and designing the structure and function of the dynamic battery temperature control engine according to the optimization results, including real-time temperature monitoring, control algorithms, and feedback mechanisms, to ensure that the engine can respond to the battery's operating status and environmental changes in real time, and maintain the battery's safety and efficiency by adjusting the cooling and heating strategies, and developing temperature control engine software, integrating the temperature monitoring system, control algorithm, and user interface.Ensure that the software can process real-time data and respond quickly, test the performance of the dynamic temperature control engine in an experimental environment, and record its response time, temperature control accuracy, and performance under different load conditions.

[0104] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0105] Step S11: Perform CT scanning on the target energy storage battery to obtain an internal structure diagram of the energy storage battery;

[0106] Step S12: Perform deep visual recognition on the internal structure diagram of the energy storage battery and mark each battery cell;

[0107] Step S13: performing precise positioning calculation of the internal space of each battery cell, and extracting the internal space coordinates of each battery cell;

[0108] Step S14: performing internal space layout analysis according to the internal space coordinates of each battery cell to generate battery space layout features;

[0109] Step S15: performing inter-cell physical connection analysis on each battery cell to obtain a topological connection relationship of the battery cells;

[0110] Step S16: reconstructing the three-dimensional structure of the battery space layout features based on the topological connection relationship of the battery cells to construct a three-dimensional battery topological structure model.

[0111] In this embodiment, a high-resolution X-ray computed tomography (CT) device is selected to ensure that the highly detailed internal structure can be captured. The device should have a slice thickness of at least 1 mm and a spatial resolution of up to 0.5 mm. The outside of the energy storage battery is cleaned to ensure that there is no dust or other substances that affect the imaging quality. The battery is fixed on the sample stage of the CT scanner to avoid movement during the scanning process. An appropriate scanning protocol is set, including voltage (such as 120 kV), current (such as 200 μA) and scanning time (such as 10 minutes) to ensure image clarity and detailed visualization. The scanning program is executed and multiple slice images obtained are recorded. According to the scanning results, it is usually necessary to obtain 360-degree slices to ensure full coverage of the energy storage battery. The internal structure of the battery, apply filtered back projection (FBP) and iterative reconstruction algorithms to process the raw data obtained from the scan to generate a complete three-dimensional image, use software (such as CT scan-specific analysis software) to process the data to generate a high-quality internal structure map, use image processing software (such as OpenCV) to remove noise and enhance the image of the internal structure map to improve recognition accuracy, apply Gaussian blur and edge detection algorithms, and use threshold segmentation (such as Otsu algorithm) to separate the battery cell from the background to ensure the accuracy of the subsequent recognition process, use convolutional neural network (CNN) for image recognition, select a pre-trained model (such as ResNet or U-Net) for fine-tuning to adapt to the characteristics of the battery cell, and use the model to automatically Automatically mark each battery cell and output the marking result map to ensure that each cell has a unique ID for subsequent analysis. Determine the origin and axis of the three-dimensional coordinate system. Usually, the geometric center of the battery is selected as the origin. The X, Y, and Z axes point to the length, width, and height of the battery respectively. By traversing the pixel points of each battery cell, calculate its geometric center coordinates. For each battery cell, use the formula to calculate its center of mass position to obtain the three-dimensional coordinates. Record the internal space coordinates of each battery cell in the database to ensure the accuracy of the data and the convenience of subsequent analysis. Determine the layout features that need to be analyzed, such as the distance between battery cells, the arrangement method (such as straight line, matrix), space utilization, etc. Use the Euclidean distance formula to calculate the distance between each battery cell. The distance between battery cells is used to evaluate the effectiveness of the layout. The ratio of the total occupied volume to the available volume of the battery's internal space is calculated to evaluate the rationality of the spatial layout. The connection relationship between battery cells is defined, including electrical connection and mechanical connection. Electrical connection is usually achieved through wires or welds, while mechanical connection involves the physical position and support of the cell. A topological model of the battery cell is constructed using graph theory methods. Each cell is regarded as a node in the graph and the connection as an edge. By analyzing the actual layout of the battery cells, the connection relationship is extracted, a topological connection matrix is ​​generated, and the connection status of each cell is recorded. A suitable 3D reconstruction algorithm is selected, such as voxel reconstruction or patch-based modeling technology, to ensure that the spatial layout and connection relationship of the battery cells can be accurately reflected.The extracted layout features and topological connection relationships are used as input to generate a 3D model of the battery using 3D modeling software (such as Blender or MeshLab). The generated 3D model is visualized to ensure that it is consistent with the actual physical structure. The accuracy of the 3D model is verified by comparing it with the CT scan image.

[0112] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0113] Step S21: Acquire real-time temperature monitoring parameters of the energy storage battery based on multiple sensors;

[0114] Step S22: Calculate the internal position of each sensor node one by one according to the multi-sensor to obtain the internal space coordinates of the battery of each sensor node;

[0115] Step S23: performing spatial temperature distribution mining on the real-time temperature monitoring parameters of the energy storage battery according to the internal spatial coordinates of the battery of each sensor node, so as to obtain the temperature distribution characteristics of each battery cell;

[0116] Step S24: dividing the three-dimensional battery topology model into multiple battery unit regions, thereby generating multiple unit region models;

[0117] Step S25: Based on the temperature distribution characteristics of each battery cell, the multiple unit area models are dynamically temperature-distributed and visualized, thereby constructing thermal state portraits of the multiple unit areas.

[0118] In this embodiment, a suitable temperature sensor is selected, such as a thermocouple, RTD (resistance temperature detector) or a digital temperature sensor (such as DS18B20), and is evaluated based on its accuracy, response time and working range. The number and arrangement of sensors are determined to cover all key areas of the energy storage battery, ensuring that each area is monitored by at least one sensor. Sensors are reasonably arranged inside and outside the battery module to ensure that the temperature changes of the battery are monitored. Sensors are placed at the top, bottom and middle of the battery, and the position and number of each sensor are recorded for subsequent data analysis and processing. A data acquisition system is developed, and a microcontroller (such as Arduino or Raspberry Pi) is used to connect each sensor, and temperature data is periodically read. The frequency of data acquisition is set, for example, temperature data is collected once every second to ensure that temperature changes can be monitored in real time. The collected temperature data is sent to a central server or cloud platform via wireless transmission (such as Wi-Fi or Bluetooth) for storage and processing, ensuring the stability and security of data transmission, using an encryption protocol to protect the integrity of the data, and designing a position calibration scheme to determine how to calculate the coordinates of the sensor inside the battery. A three-dimensional coordinate system (X, Y,Z) to represent the position of each sensor, set the reference point of each sensor, such as the center point of the battery module, in order to perform relative position calculation, collect the actual installation position data of each sensor, including the distance and direction between the sensor and the reference point, and record them as relative coordinates. Use three-dimensional measurement tools (such as laser rangefinders) to ensure the accuracy of the position data, use geometric calculation methods to determine the spatial coordinates of each sensor, use basic three-dimensional coordinate formulas, combine the relative position data of the sensor, calculate its absolute coordinates, organize the calculation results into a data table, record the unique identification of each sensor and its corresponding spatial coordinates, integrate the real-time temperature monitoring parameters with the spatial coordinates of the sensor, and form a table containing sensor ID, temperature value and spatial coordinates, clean the data, remove invalid or abnormal temperature readings, ensure data accuracy, format the integrated data into a form suitable for analysis, such as using a data frame (DataFrame) structure to facilitate subsequent processing, select a suitable temperature distribution analysis algorithm, such as interpolation, heat map generation, or Kriging interpolation (Kriging) to visualize the temperature distribution, set the spatial range of the analysis to ensure that the entire battery module is covered, analyze the data using the selected algorithm, generate a temperature distribution model inside the battery, use Python's SciPy or Matplotlib library for data visualization, and record the temperature distribution characteristics of each battery cell, including the mean and maximum temperature , minimum value and standard deviation, use computer-aided design (CAD) software (such as SolidWorks or AutoCAD) to build a three-dimensional battery topology model to ensure that the model accurately reflects the actual shape of the battery, determine the boundaries and connection methods of the battery cells in the model, record the characteristics of each battery cell, set regional division standards according to the function, thermal state or physical location of the battery cell, divide the battery cell according to the thermal performance or cooling efficiency, determine the number and shape of the divisions, ensure that each area can be analyzed independently and is representative, use regional division algorithms (such as Voronoi diagrams or K-means clustering) to divide the three-dimensional model, generate multiple battery cell regional models, and record the boundaries of each area and features for subsequent analysis and processing, select a suitable visualization rendering tool (such as Blender, Unity or VTK), build a visualization environment, ensure that the rendering tool can support dynamic data visualization and has good graphics rendering performance, set rendering parameters according to the temperature distribution characteristics of each battery cell, including color mapping, temperature range and dynamic change rate, design visual effects, such as high temperature areas are displayed in red and low temperature areas are displayed in blue to facilitate identification of temperature anomalies, input temperature data into the visualization tool, perform dynamic rendering, use real-time data streams to update rendering parameters, display temperature changes over time, and record rendering results, including thermal state portraits at different time points, for subsequent analysis and reporting. ,

[0119] In this embodiment, the specific steps of step S25 are:

[0120] Performing real-time temperature distribution change analysis on the temperature distribution characteristics of each battery cell to generate real-time temperature distribution change data;

[0121] Perform discrete fitting of time series temperature values ​​on real-time temperature distribution change data to construct the temperature distribution change curve of each battery cell;

[0122] Perform temperature positioning matching on the temperature distribution change curve of each battery cell and multiple unit area models to obtain the matching temperature curve relationship of each unit area;

[0123] The temperature distribution of multiple unit area models is visualized according to the matching temperature curve relationship of each unit area, so as to obtain the temperature distribution thermodynamic map of multiple unit areas;

[0124] The temperature distribution thermodynamic map of multiple unit areas is dynamically rendered to construct thermal state portraits of multiple unit areas.

[0125] In this embodiment, a multi-sensor architecture is used to ensure that the temperature of each battery cell can be recorded regularly. The sensors should be installed at key locations of the battery and ensure that they work properly. The frequency of data collection is set, such as once per second, to ensure that subtle changes in temperature can be captured. A microcontroller (such as Arduino or Raspberry Pi) is used. Pi) for data collection, transmit data to the server in real time through wireless network, use MQTT protocol to effectively process real-time data stream, store the collected temperature data in time series database (such as InfluxDB) for subsequent analysis and query, use Python's Pandas library to analyze temperature data in real time, calculate the mean, maximum and minimum temperature of each battery cell, and monitor the temperature change trend, set monitoring indicators, such as when the temperature change rate exceeds a certain threshold (such as more than 5°C per minute), record it as abnormal, use Matplotlib or Plotly to generate real-time temperature distribution map, show the temperature change of each battery cell, help real-time monitoring and decision-making, form a visual report of real-time temperature change data, which is convenient for engineers to analyze and troubleshoot, clean real-time temperature data, remove outliers and noise, ensure data accuracy, use Z-score or IQR method to detect and remove abnormal data points, sort temperature data by timestamp, create a time series data set for subsequent analysis, and select appropriate fitting algorithm, such as polynomial fitting or spline interpolation (spline The temperature variation curve of each battery cell is constructed by using the scipy.interpolate module in Python's SciPy library for discretization. The selected fitting algorithm is applied to fit the temperature time series data of each battery cell to generate a temperature variation curve. The fitting parameters, including the coefficients of the fitting function and the goodness of fit (R 2 value) to evaluate the fitting effect.

[0126] Use Matplotlib to generate a temperature change curve for each battery cell, showing the change of temperature over time, which is convenient for subsequent analysis and comparison, and form a graphical report of the temperature distribution change, which is provided to relevant personnel for evaluation, determine the standard for temperature matching, set a similarity measurement method, such as mean square error (MSE) or dynamic time warping (DTW) to evaluate the similarity between curves, set a matching threshold, and if the MSE is less than a certain value (such as 0.5), the two temperature curves are considered to match. Use Python to implement the curve matching algorithm, match the temperature change curve of the battery cell with each unit area model, calculate the similarity, record the matching results, generate the matching temperature curve relationship for each unit area, including the matching battery cell ID and area ID, analyze the matching results, identify the best matching unit area, evaluate the accuracy and stability of the matching, adjust the parameters of the matching algorithm according to the analysis results to improve the matching effect and accuracy, select a suitable visualization tool (such as Matplotlib, Seaborn or Plotly) to generate a temperature distribution heat map, and ensure that the selected tool supports multidimensional data Visualize and process temperature data from different battery cells. Set the color mapping of the heat map according to the matching temperature curve relationship of each area, ensure that the high temperature area is displayed in red and the low temperature area is displayed in blue. Set the temperature range and resolution to clearly display the temperature distribution of each unit area. Use the selected visualization tool to generate temperature distribution heat maps of multiple unit areas to display the temperature characteristics of each area. Record the generated heat map data for subsequent analysis and monitoring. Select a dynamic rendering tool (such as Unity, Blender, or OpenGL) to ensure support for real-time data streaming and dynamic effects. Build a rendering environment, configure necessary libraries and dependencies, and ensure that the tool can run efficiently. Set dynamic rendering parameters according to the heat map data, including rendering speed, temperature change range, and visual effects (such as animation transition effects). Design visual effects to make the temperature change process intuitive and easy to understand, such as displaying temperature changes through gradient colors. Input the heat map data into the dynamic rendering tool to generate dynamic pictures of the thermal status of multiple unit areas to display the temperature change process. Record key data in the rendering process for analysis and reporting.

[0127] In this embodiment, refer to Figure 4 , is a flowchart of detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0128] Step S31: defining multiple operating conditions of the battery pack;

[0129] Step S32: performing multi-period battery operation simulation on thermal state portraits of multiple unit areas according to multiple operating conditions of the battery pack to collect multi-period operation simulation data;

[0130] Step S33: performing temperature change trend analysis on each region of the multi-period operation simulation data to obtain temperature change trend characteristics of each region;

[0131] Step S34: predicting the temperature change situation of the sliding window according to the temperature change trend characteristics of each area to obtain the predicted temperature change situation of each area.

[0132] In this embodiment, according to the application scenarios of the battery pack (such as electric vehicles, energy storage systems, etc.), multiple key operating environment parameters are identified, including temperature, humidity, load type, charge and discharge rate, etc. Set different operating conditions, such as high temperature, high load, low temperature, etc., to ensure that various situations encountered by the battery pack in actual applications are covered. The defined operating conditions are classified into normal operating conditions, extreme operating conditions, and critical operating conditions, and each operating condition is described in detail, including its impact on battery performance. Develop specific parameter ranges for each operating condition. Under high temperature conditions, set the temperature range to 40°C to 60°C and the load to 1C rate. Organize all defined operating conditions into documents, including the name, parameter range, expected impact and other information of each operating condition. Ensure that the document is clear and easy to understand for subsequent simulation and analysis personnel. Invite experts in related fields to review the defined operating conditions to ensure that they meet the actual application requirements and correct the deficiencies. Adjustments are made based on feedback, and the definitions of multiple operating conditions are finally confirmed. Use battery simulation software (such as MATLAB / Simulink, COMSOL Multiphysics, or ANSYS) to build a thermal state model of the battery pack to ensure that the model can reflect the thermal behavior and electrochemical characteristics of the battery. Set model parameters, including the thermal conductivity, specific heat capacity, exothermic reaction, etc. of the battery to ensure the accuracy of the simulation results. According to the defined operating conditions, set the input parameters of the simulation model, including the initial temperature of the battery pack, the charging and discharging strategy, and the external environmental conditions (such as heat dissipation conditions). Ensure that each operating condition can be reflected in the simulation to compare the battery performance under different operating conditions. Run the simulation model for multiple time periods (such as 0-1 hour, 1-2 hours, 2-3 hours), and record the temperature changes of each battery cell in each time period. After each time period, save the simulation results, including the temperature distribution, heat flux density, and other data of each area. Export the simulation results to a database or file system to ensure the traceability and integrity of the data. Use CSV or JSON format to record data for subsequent analysis. Integrate the simulation results from different time periods into a data set to ensure that the temperature changes in each area are fully recorded. Clean the data to remove invalid or abnormal temperature records to ensure the accuracy of data analysis. Select appropriate trend analysis methods, such as linear regression, moving average, or exponential smoothing, to identify trends in temperature changes. Determine the time window for analysis, such as using data from the past three time periods for trend analysis to obtain a smooth change curve. Use Python's Pandas and NumPy libraries to perform trend analysis on the integrated data and calculate the average temperature change, maximum, and minimum values ​​for each region. Record the temperature change trend characteristics of each region, such as temperature increases, decreases, or remains stable. Use Matplotlib or Seaborn to generate a temperature change trend chart to show the temperature changes in each region for subsequent decision-making and analysis.Organize the results into a report to facilitate sharing and discussion with the team. Choose an appropriate time series forecasting model, such as ARIMA (autoregressive integrated moving average model), LSTM (long short-term memory network), or SARIMA (seasonal ARIMA). Train the model based on historical temperature change data to ensure that the model can capture the trend and periodicity of temperature changes. Determine the parameters of the model, such as p, d, q values ​​in the ARIMA model, or the number of network layers and nodes in the LSTM. Use methods such as grid search to optimize parameter selection. Apply the sliding window method on the trained model to make temperature forecasts for each region. Use data from the past 5 time periods to predict temperature changes in the next time period. Record the predicted values ​​for each region, including the predicted temperature and confidence interval, to assess the reliability of the forecast. Verify the accuracy of the forecast model by comparing the predicted temperature with the actual temperature (if any), and calculate the forecast error (such as MSE or MAE). Use visualization tools to display the forecast results and compare the predicted curve with the actual temperature curve to facilitate analysis and adjustment of the model.

[0133] In this embodiment, the specific steps of step S32 are:

[0134] Based on the real-time temperature monitoring parameters of the energy storage battery, the temperature difference change correlation mining between adjacent cells is carried out for each battery cell to obtain the temperature change relationship between adjacent cells;

[0135] Based on the temperature change relationship between adjacent cells, the heat transfer between battery cells is evolved to obtain the heat transfer law of battery cells;

[0136] Calculate the cooling efficiency of energy storage batteries;

[0137] According to the multiple operating conditions of the battery pack, a multi-operating condition operating characteristic quantitative analysis is performed on the heat transfer law of the battery unit and the cooling efficiency of the energy storage battery, so as to generate battery operating characteristics of different operating conditions;

[0138] Based on the battery operation characteristics of different working conditions, multi-period battery operation simulation is performed on the thermal state portraits of multiple unit areas to collect multi-period operation simulation data.

[0139] In this embodiment, the established multi-sensor temperature monitoring system is used to collect real-time temperature data of each battery cell, and the data collection frequency is ensured, for example, once per second, to capture subtle changes in temperature. The data is stored in a database (such as InfluxDB) for subsequent analysis. The temperature data of each battery cell is integrated with the data of its neighboring cells to form a data frame containing temperature, timestamp and location. The data is cleaned to remove missing values ​​and outliers to ensure the accuracy of the analysis results. Statistical methods such as Pearson Correlation Coefficient or Spearman Rank Correlation are used to analyze the temperature difference changes between neighboring cells, and a correlation threshold is set, for example, the correlation coefficient is greater than 0.7 is considered as a strong correlation to screen out significant temperature change relationships, calculate the correlation of the temperature data of each pair of adjacent cells, record the correlation value and change trend of each cell pair, generate a report to show the correlation of the temperature difference changes between adjacent cells and its statistical analysis results, which is convenient for subsequent research and decision-making, select a suitable heat transfer model, such as a one-dimensional heat conduction equation or a two-dimensional heat transfer model, to reflect the heat transfer characteristics between battery cells, build a model in MATLAB or Python, set initial conditions and boundary conditions, such as the initial temperature of the battery cell and the external environment temperature, run the heat transfer model, simulate the heat transfer process in different time periods, and record the temperature of each battery cell at different time points Changes, set the simulation time step, for example, update the temperature distribution every 1 minute, analyze the temperature change trend during the heat transfer process, record the temperature response and heat transfer rate of each unit, generate a heat transfer law report, and display the heat transfer relationship and evolution characteristics between different battery cells. The cooling efficiency is defined by comparing the actual temperature of the battery cell with the theoretical temperature (such as ambient temperature), determine the source of the ambient temperature and theoretical temperature, ensure the accuracy of the data, integrate the actual temperature data, ambient temperature data and theoretical temperature data of the battery cell to form the data set required for calculation, analyze the cooling efficiency changes under different working conditions, identify the main factors affecting the cooling efficiency (such as load, environmental conditions, etc.), and generate cooling Cooling efficiency analysis report is provided to facilitate subsequent optimization of the cooling system. Different test scenarios are set according to the previously defined multiple operating conditions, such as high temperature and high load, low temperature and low load, etc., to ensure that all application scenarios are covered. Specific parameters are defined for each operating condition, such as temperature range, charge and discharge rate and cooling method. The battery pack is operated under different operating conditions, and the temperature data, cooling efficiency data and heat transfer law data of the battery cells are collected. Statistical analysis methods (such as variance analysis ANOVA) are used to compare the data under different operating conditions, quantify the impact of each operating condition on battery performance, and organize the quantitative analysis results of different operating conditions, such as forming a summary of the temperature distribution, cooling efficiency and heat transfer characteristics under each operating condition. Report, visualize the results to facilitate understanding and communication. Update the input parameters of the simulation model based on the battery operation characteristics obtained in the previous steps, such as adjusting the thermal conductivity, heat capacity, etc. to reflect the actual conditions under different working conditions. Set new simulation initial conditions to ensure consistency with the actual operating conditions. Perform multi-period simulations in the updated model, record the temperature changes of the battery cells in each time period, set the simulation time range, such as from startup to full charge and discharge, record the temperature data every minute, record the simulation results in a database or file to ensure the data integrity of subsequent analysis, and use visualization tools to generate thermal state portraits to display the temperature distribution of battery cells in different time periods. .

[0140] In this embodiment, step S4 includes the following steps:

[0141] Step S41: Calculating the temperature change rate of each region's predicted temperature change situation to generate a temperature change situation rate for each region;

[0142] Step S42: analyzing the temperature change fluctuation range of each region on the temperature change prediction trend to obtain the temperature change fluctuation range of each region;

[0143] Step S43: identifying the temperature rise deviation based on the temperature change trend rate of each area and the temperature change fluctuation range of each area, and marking the temperature change deviation area;

[0144] Step S44: performing continuous abnormal temperature rise trend detection on the temperature change deviation area based on a preset temperature rise deviation threshold, thereby generating abnormal temperature rise trend features;

[0145] Step S45: generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics.

[0146] In this embodiment, the temperature change data of each area in the specified time period is extracted from the previous temperature prediction results, ensuring that the timestamps of the data are continuous and the units are consistent (minutes), the temperature change data of each area is iteratively calculated, the temperature change rate of each time period is calculated, the temperature change rate data of each area is recorded, and a rate analysis report is generated, including statistical information such as the average value, maximum value and minimum value of the rate. The temperature change rate formula is: Among them, R tis the temperature change rate, ΔT is the temperature change amount, Δt is the time interval, and Matplotlib or Seaborn is used to generate a temperature change rate graph to show the temperature change rate changes in different regions for easy analysis and comparison. The temperature change fluctuation range is usually defined as the difference between the maximum and minimum temperature changes in a period of time, W = T max-T min, W is the fluctuation range, T max is the highest temperature in the time period, T min is the lowest temperature. The maximum, minimum and change of temperature in each area within the specified time are extracted from the temperature prediction data and organized into a data frame for statistical analysis. The data of each area are traversed, and the above fluctuation range formula is applied to calculate the temperature fluctuation range of each area. The fluctuation range of each area is recorded, and an analysis report is generated, including the average value, maximum value and minimum value of the fluctuation range. The temperature fluctuation range of each area is displayed using a bar chart or a box plot to make the analysis result more intuitive. The deviation identification standard is set according to the temperature change rate and the fluctuation range. When the temperature change rate exceeds a certain threshold (such as 2℃ / min) and the temperature change fluctuation range exceeds a certain threshold (such as 5℃), it is marked as a temperature change deviation area. The data of the temperature change rate and the fluctuation range are integrated to form a comprehensive data frame. The data of each area are traversed to check whether the deviation identification standard is met. If the conditions are met, the area is marked as a temperature change deviation area, the marking results are recorded, and a deviation identification report is generated, including a list of deviation areas and their related features. According to the historical temperature data and the current temperature change situation The temperature rise deviation threshold (such as 3°C exceeding the range) is set to detect abnormal trends, and the marked deviation areas are continuously monitored to check their temperature changes in real time. The temperature data of each time period is recorded, and the sliding window method is applied to calculate the temperature changes in each time period to check whether it exceeds the set threshold. If the temperature is detected to exceed the threshold, the event is recorded, and an abnormal temperature rise trend feature report is generated, including information such as the time, area and temperature change amount of the abnormality. The temperature trend is displayed using a time series graph to highlight the abnormal temperature rise events to help analysts quickly understand the temperature changes. The warning signal standard is set according to the abnormal temperature rise detection results. For example, if the temperature exceeds the threshold in three consecutive tests, the warning signal is triggered. The recorded abnormal temperature rise trend is monitored. If the set warning conditions are met, a temperature abnormality warning signal is generated, the warning event is recorded, the time, area and temperature change information are recorded, the response mechanism of the warning signal is determined, and an alarm email, text message or automatic start of the cooling system is sent, etc. The generated warning signal is visualized to form a warning report, which is convenient for relevant personnel to respond and handle in time.

[0147] In this embodiment, step S5 includes the following steps:

[0148] Step S51: locating the abnormal temperature starting point based on the abnormal temperature warning signal, thereby marking the abnormal temperature rise starting unit;

[0149] Step S52: analyzing the abnormal temperature rise evolution path of the abnormal temperature rise starting unit based on the three-dimensional battery topology model to generate multiple abnormal temperature rise evolution paths;

[0150] Step S53: performing dynamic temperature rise diffusion evolution based on multiple abnormal temperature rise evolution paths, thereby generating a diffusion prediction trend of the abnormal starting unit;

[0151] Step S54: making a battery abnormal fault diagnosis decision based on the diffusion prediction trend of the abnormal starting cell to obtain a battery abnormal fault diagnosis result.

[0152] In this embodiment, the temperature abnormality warning signal data is collected to ensure that all abnormal temperature rise records, timestamps and corresponding battery cell positions are included, and data from different monitoring systems (such as temperature sensors, data acquisition systems, etc.) are integrated to form a complete data set. According to the characteristics of the abnormal temperature rise, the starting point recognition standard is set, and the time point when the temperature first exceeds the set threshold is selected as the abnormal starting point. The threshold is set, such as 3°C, to facilitate accurate identification. The collected abnormal temperature rise data is traversed to find the battery cells whose temperature exceeds the threshold for the first time, and the specific location and time of these cells are recorded. Each abnormal temperature rise starting cell is marked, and a report is generated, including the ID, location and abnormal temperature of the abnormal starting cell. The temperature rise information such as temperature value is displayed by using a thermal map or 3D model to show the marked abnormal temperature rise starting unit, helping engineers and analysts to quickly identify the problem area. Computer-aided design (CAD) software or simulation tools (such as ANSYS and COMSOL Multiphysics) are used to build a three-dimensional topological structure model of the battery pack to ensure that the model can accurately reflect the layout and connection relationship of the battery cells. According to the position of the abnormal temperature rise starting unit and the heat transfer characteristics between adjacent units, the standard of evolution path analysis is set, and the adjacent heat transfer coupling units are selected as the temperature rise propagation path. The three-dimensional model is thermally simulated by the simulation tool to simulate the diffusion process of the abnormal temperature rise and record the diffusion process from the abnormal starting unit to the adjacent units. The temperature change path of the adjacent unit is used to generate multiple abnormal temperature rise evolution paths, record the temperature change information, propagation rate and time of each path, and use 3D visualization tools to display the evolution path of abnormal temperature rise, so as to help analysts deeply understand the temperature rise propagation mechanism, establish a dynamic temperature rise diffusion model, select a suitable numerical model (such as a finite element model) to describe the temperature change over time, and consider the physical properties of the material such as thermal conductivity and specific heat capacity, define the standard of the diffusion path, based on the identified abnormal temperature rise evolution path, combined with the thermal conduction characteristics of the adjacent units, predict the diffusion trend of the temperature rise, apply the constructed diffusion model, perform temperature diffusion simulation at different time steps (such as every 1 minute), and record the temperature change. Calculate the temperature distribution at each time point and generate diffusion prediction trend data. Use heat maps or dynamic charts to display the dynamic temperature rise diffusion evolution, and intuitively display the temperature changes at different time points. Set the fault diagnosis standard according to the temperature rise diffusion prediction trend. If the temperature in a certain area continues to exceed the threshold (such as 5°C) within the specified time, it can be determined as a potential fault. Analyze the diffusion prediction trend data, identify the area where the temperature continues to rise within the set time period, and determine whether it meets the fault diagnosis standard. Record the diagnosis results, including the ID of the fault area, temperature changes and time information, organize the fault diagnosis results, and form a comprehensive report with detailed information on the abnormal fault area to facilitate subsequent processing and decision support.

[0153] In this embodiment, step S6 includes the following steps:

[0154] Step S61: Obtaining the historical work log of the energy storage battery;

[0155] Step S62: performing multiple operation scenario identification on the energy storage battery historical work log to obtain multiple battery operation scenarios;

[0156] Step S63: performing in-depth mining of the load state of each of the multiple battery operation scenarios, thereby generating a load state feature of each operation scenario;

[0157] Step S64: dynamically transfer learning the load state characteristics of each operation scenario to generate a deep representation of the battery load state;

[0158] Step S65: Optimize dynamic temperature control parameters according to the deep characterization of battery load status and the abnormal battery fault diagnosis results to build a dynamic battery temperature control engine to execute battery temperature control warning and temperature control parameter decision.

[0159] In this embodiment, the content of the historical work log is determined, which usually includes the battery's charge and discharge records, temperature monitoring data, operating hours, load status and fault records, etc. The data storage location is identified, such as a database, file system or cloud storage, and a data extraction tool (such as SQL query, Python script) is used to extract the required work log from the relevant data source to ensure that the extracted time range includes the complete working cycle of the battery. The collected data should include timestamps, charge and discharge status, temperature changes and any related event records. The extracted data is cleaned to remove duplicate records and missing values ​​to ensure data consistency and integrity. Key indicators such as temperature and load are standardized for subsequent analysis. According to the battery usage, Define operating scenarios, such as normal charging and discharging, overload, low-temperature operation, high-temperature operation, etc., clarify the characteristic indicators of each scenario, such as the battery charging rate, discharge rate, temperature range and load type, use data mining and machine learning techniques (such as clustering analysis, classification algorithm) to analyze historical work logs, identify different operating scenarios, use K-Means clustering algorithm, cluster according to battery temperature, load and charging and discharging status and other characteristics, identify various operating scenarios, assign labels to each identified operating scenario for subsequent analysis, record the characteristics and number of data samples for each scenario, determine the key indicators of load status, such as instantaneous load, battery charging and discharging efficiency, temperature change rate, etc., and select appropriate statistical methods (such as mean, variance, maximum, minimum) to describe the load characteristics, analyze each scene one by one, extract its load status indicators, use Python's Pandas library to group the data, calculate the key indicators, record the load characteristics of each scene, including the load change trend and stability, etc., organize the load status characteristics of each operating scene, and form a data report containing the key indicators, statistical information and charts of each scene to prepare for subsequent dynamic transfer learning, ensure that the data is clear and available, design a transfer learning framework, select a suitable model (such as convolutional neural network CNN, long short-term memory network LSTM) to process the battery load status characteristics, set the source task (historical data) and the target task (real-time data) The relationship between them is used to better transfer knowledge. The load state feature data in the historical work log is used to train the model. Appropriate training parameters (such as learning rate and batch size) are set. The performance of the model is evaluated by cross-validation method to ensure that the model can effectively learn and identify the load state features. The deep representation is extracted from the trained model to form a feature vector of the battery load state for subsequent analysis and application. The deep representation of each scene is recorded to form a representation result report. According to the deep representation of the battery load state and the fault diagnosis results, the standard for temperature control parameter optimization is set, the temperature range (such as 15℃-35℃) and the load change rate are set, and the optimization algorithm (such as genetic algorithm and particle swarm optimization algorithm) is used to optimize the temperature control parameters.Ensure the best temperature control effect of the battery under various load conditions, set optimization goals, such as maximizing charging and discharging efficiency or minimizing temperature fluctuations, build a dynamic temperature control engine based on the optimization results, integrate automatic temperature control system and early warning mechanism to monitor and adjust battery temperature in real time, and develop control algorithms to ensure that the temperature control system can quickly respond to load changes and environmental changes.

[0160] In this embodiment, a temperature control warning system for an energy storage battery is provided, which is used to execute the temperature control warning method for an energy storage battery as described above, including:

[0161] The topology module is used to obtain the internal structure diagram of the energy storage battery; analyze the internal space layout and the physical connection between units of the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model;

[0162] Visualization rendering module, used to obtain real-time temperature monitoring parameters of energy storage batteries; perform dynamic temperature distribution visualization rendering of the three-dimensional battery topology model according to the real-time temperature monitoring parameters of the energy storage batteries, thereby constructing thermal state portraits of multiple unit areas;

[0163] The multi-period simulation module is used to perform multi-period battery operation simulation on the thermal state portraits of multiple unit areas and to predict the temperature change trend of the sliding window to obtain the predicted temperature change trend of each area;

[0164] The abnormal warning module is used to continuously detect the abnormal temperature rise trend of the temperature change forecast situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating temperature abnormality warning signals based on the abnormal temperature rise trend characteristics;

[0165] A diagnosis decision module is used to perform dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then make a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result;

[0166] The dynamic temperature control optimization module is used to obtain the historical work log of the energy storage battery; deeply mine the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

[0167] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0168] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A temperature control early warning method for an energy storage battery, characterized in that: The following steps are involved: Step S1: obtaining an internal structure diagram of an energy storage battery; performing internal space layout analysis and inter-unit physical connection analysis on the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model; Step S2: obtaining real-time temperature monitoring parameters of the energy storage battery; performing dynamic temperature distribution visualization rendering on the three-dimensional battery topology structure model according to the real-time temperature monitoring parameters of the energy storage battery, thereby constructing thermal state portraits of multiple unit areas; Step S3: performing multi-period battery operation simulation on the thermal state portraits of multiple unit areas, and performing sliding window temperature change trend prediction to obtain the temperature change prediction trend of each area; Step S4: Detecting the continuous abnormal temperature rise trend of the temperature change prediction situation in each area, thereby generating abnormal temperature rise trend characteristics; Generate temperature abnormality warning signals based on abnormal temperature rise trend characteristics; Step S5: performing dynamic temperature rise diffusion evolution based on the abnormal temperature warning signal, and then making a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result; Step S6: Obtain the historical work log of the energy storage battery; conduct in-depth mining of the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

2. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: Perform CT scanning on the target energy storage battery to obtain an internal structure diagram of the energy storage battery; Step S12: Perform deep visual recognition on the internal structure diagram of the energy storage battery and mark each battery cell; Step S13: performing precise positioning calculation of the internal space of each battery cell, and extracting the internal space coordinates of each battery cell; Step S14: performing internal space layout analysis according to the internal space coordinates of each battery cell to generate battery space layout features; Step S15: performing inter-cell physical connection analysis on each battery cell to obtain a topological connection relationship of the battery cells; Step S16: reconstructing the three-dimensional structure of the battery space layout features based on the topological connection relationship of the battery cells to construct a three-dimensional battery topological structure model.

3. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: Acquire real-time temperature monitoring parameters of the energy storage battery based on multiple sensors; Step S22: Calculate the internal position of each sensor node one by one according to the multi-sensor to obtain the internal space coordinates of the battery of each sensor node; Step S23: performing spatial temperature distribution mining on the real-time temperature monitoring parameters of the energy storage battery according to the internal spatial coordinates of the battery of each sensor node, so as to obtain the temperature distribution characteristics of each battery cell; Step S24: dividing the three-dimensional battery topology model into multiple battery unit regions, thereby generating multiple unit region models; Step S25: Based on the temperature distribution characteristics of each battery cell, the multiple unit area models are dynamically temperature-distributed and visualized, thereby constructing thermal state portraits of the multiple unit areas.

4. The energy storage battery temperature control early warning method according to claim 3 is characterized in that: The specific steps of step S25 are: Performing real-time temperature distribution change analysis on the temperature distribution characteristics of each battery cell to generate real-time temperature distribution change data; Perform discrete fitting of time series temperature values ​​on real-time temperature distribution change data to construct the temperature distribution change curve of each battery cell; Perform temperature positioning matching on the temperature distribution change curve of each battery cell and multiple unit area models to obtain the matching temperature curve relationship of each unit area; The temperature distribution of multiple unit area models is visualized according to the matching temperature curve relationship of each unit area, so as to obtain the temperature distribution thermodynamic map of multiple unit areas; The temperature distribution thermodynamic map of multiple unit areas is dynamically rendered to construct thermal state portraits of multiple unit areas.

5. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: defining multiple operating conditions of the battery pack; Step S32: performing multi-period battery operation simulation on thermal state portraits of multiple unit areas according to multiple operating conditions of the battery pack to collect multi-period operation simulation data; Step S33: performing temperature change trend analysis on each region of the multi-period operation simulation data to obtain temperature change trend characteristics of each region; Step S34: predicting the temperature change situation of the sliding window according to the temperature change trend characteristics of each area to obtain the predicted temperature change situation of each area.

6. The energy storage battery temperature control early warning method according to claim 5, characterized in that: The specific steps of step S32 are: Based on the real-time temperature monitoring parameters of the energy storage battery, the temperature difference change correlation mining between adjacent cells is carried out for each battery cell to obtain the temperature change relationship between adjacent cells; Based on the temperature change relationship between adjacent cells, the heat transfer between battery cells is evolved to obtain the heat transfer law of battery cells; Calculate the cooling efficiency of energy storage batteries; According to the multiple operating conditions of the battery pack, a multi-operating condition operating characteristic quantitative analysis is performed on the heat transfer law of the battery unit and the cooling efficiency of the energy storage battery, so as to generate battery operating characteristics of different operating conditions; Based on the battery operation characteristics of different working conditions, multi-period battery operation simulation is performed on the thermal state portraits of multiple unit areas to collect multi-period operation simulation data.

7. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: Calculating the temperature change rate for the predicted temperature change trend of each region to generate the temperature change trend rate of each region; Step S42: analyzing the temperature change fluctuation range of each region on the temperature change prediction trend to obtain the temperature change fluctuation range of each region; Step S43: identifying the temperature rise deviation based on the temperature change trend rate of each area and the temperature change fluctuation range of each area, and marking the temperature change deviation area; Step S44: performing continuous abnormal temperature rise trend detection on the temperature change deviation area based on a preset temperature rise deviation threshold, thereby generating abnormal temperature rise trend features; Step S45: generating a temperature abnormality warning signal based on the abnormal temperature rise trend characteristics.

8. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: locating the abnormal temperature starting point based on the abnormal temperature warning signal, thereby marking the abnormal temperature rise starting unit; Step S52: analyzing the abnormal temperature rise evolution path of the abnormal temperature rise starting unit based on the three-dimensional battery topology model to generate multiple abnormal temperature rise evolution paths; Step S53: performing dynamic temperature rise diffusion evolution based on multiple abnormal temperature rise evolution paths, thereby generating a diffusion prediction trend of the abnormal starting unit; Step S54: making a battery abnormal fault diagnosis decision based on the diffusion prediction trend of the abnormal starting cell to obtain a battery abnormal fault diagnosis result.

9. The energy storage battery temperature control early warning method according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: Obtaining the historical work log of the energy storage battery; Step S62: performing multiple operation scenario identification on the energy storage battery historical work log to obtain multiple battery operation scenarios; Step S63: performing in-depth mining of the load status of each of the multiple battery operation scenarios, thereby generating a load status feature of each operation scenario; Step S64: dynamically transfer learning the load state characteristics of each operation scenario to generate a deep representation of the battery load state; Step S65: Optimize dynamic temperature control parameters according to the deep characterization of battery load status and the abnormal battery fault diagnosis results to build a dynamic battery temperature control engine to execute battery temperature control warning and temperature control parameter decision.

10. A temperature control warning system for energy storage batteries, characterized in that: The method for performing temperature control and early warning of an energy storage battery according to claim 1 comprises: The topology module is used to obtain the internal structure diagram of the energy storage battery; analyze the internal space layout and the physical connection between units of the internal structure diagram of the energy storage battery to construct a three-dimensional battery topology model; Visualization rendering module, used to obtain real-time temperature monitoring parameters of energy storage batteries; perform dynamic temperature distribution visualization rendering of the three-dimensional battery topology model according to the real-time temperature monitoring parameters of the energy storage batteries, thereby constructing thermal state portraits of multiple unit areas; The multi-period simulation module is used to perform multi-period battery operation simulation on the thermal state portraits of multiple unit areas and to predict the temperature change trend of the sliding window to obtain the predicted temperature change trend of each area; The abnormal warning module is used to continuously detect the abnormal temperature rise trend of the temperature change forecast situation in each area, thereby generating abnormal temperature rise trend characteristics; and generating temperature abnormality warning signals based on the abnormal temperature rise trend characteristics; A diagnosis decision module is used to perform dynamic temperature rise diffusion evolution based on the temperature abnormality warning signal, and then make a battery abnormal fault diagnosis decision to obtain a battery abnormal fault diagnosis result; The dynamic temperature control optimization module is used to obtain the historical work log of the energy storage battery; deeply mine the load status of each scene in the historical work log of the energy storage battery, and optimize the dynamic temperature control parameters according to the abnormal fault diagnosis results of the battery to build a dynamic battery temperature control engine.

Images