A distributed safety supervision system based on ordered energy storage big data

Abstract

The present invention discloses a safety supervision system based on orderly energy storage big data distribution, including: a thermal runaway risk assessment module, used to judge the thermal runaway risk of the current energy storage system and determine the thermal runaway position of the energy storage system; a thermal runaway position management module, used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position; a load fluctuation thermal runaway risk assessment module, used to build a load fluctuation thermal runaway risk assessment model, and judge the thermal runaway risk under different load fluctuations; a local overheating area migration analysis module, used to build a local overheating area migration prediction model, and predict the local overheating area position within a preset time period in the future; a local overheating area cooling execution module, used to adjust the cooling parameters of the local cooling device in real time. The present invention realizes the all-round management and control of the thermal runaway risk of the energy storage system by combining a variety of historical data and real-time monitoring, and significantly improves the safety and operational stability of the energy storage system.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a distributed safety supervision system based on ordered energy storage big data. Background Art

[0002] With the continuous growth of global energy demand and the widespread application of renewable energy, the position of energy storage systems in power systems has become increasingly important. Energy storage systems can not only balance supply and demand and improve grid stability, but also provide backup power during peak load periods. However, the operating environment of energy storage systems is complex and changeable, especially under the condition of high-frequency load fluctuations, its internal thermal management faces huge challenges. Local overheating in energy storage systems, namely micro-hot spots, is one of the main factors leading to thermal runaway. These micro-hot spots will continue to migrate with load fluctuations, increasing the risk of thermal runaway of the system. The thermal management of current energy storage systems mainly relies on fixed-position temperature sensors, which works well when dealing with slow temperature changes or small load fluctuations. However, when energy storage systems are subjected to high-frequency load fluctuations, the internal temperature distribution will change rapidly, resulting in the location and intensity of micro-hot spots constantly changing. It is difficult for fixed-position temperature sensors to accurately capture these dynamic changes, resulting in monitoring blind spots and response delays. In addition, load fluctuations will also interfere with the data transmission and processing systems of temperature sensors, further increasing the difficulty of monitoring. Another key issue is the frequency and amplitude of load fluctuations in energy storage systems. High-frequency and large-amplitude load fluctuations will place higher requirements on the thermal management of energy storage systems. Traditional risk assessment models may not be able to provide sufficiently accurate risk predictions when dealing with frequent and drastic load fluctuations, resulting in the system's slow response to sudden thermal runaway situations, increasing the risk of equipment damage and safety accidents. At the same time, the thermal management of energy storage systems also faces challenges in real-time and accuracy. Existing monitoring systems and risk assessment models often cannot reflect the changes in the thermal state inside the energy storage system in real time. Especially in the case of frequent load fluctuations, the lack of real-time and accurate monitoring means makes it impossible for the system to adjust the cooling strategy in time, resulting in local overheating or insufficient cooling, increasing the risk of thermal runaway. The generation and migration of micro-hot spots is another difficult problem in the thermal management of energy storage systems. Micro-hot spots not only cause local overheating, but also migrate within the system due to load fluctuations. Traditional monitoring methods make it difficult to track the dynamic changes of micro-hot spots in real time, resulting in the inability to take targeted cooling measures in time, increasing the risk of thermal runaway of equipment. Especially under high load conditions, the frequent generation and migration of micro-hot spots pose a major threat to the safe operation of the system. Existing risk assessment models and monitoring systems often show great limitations when dealing with complex environmental changes, and it is difficult to provide reliable thermal management solutions. In addition, the temperature change rate is also an important factor. The temperature change rate in the local overheating area is too fast. The existing temperature sensors and monitoring systems cannot capture this rapid change in time, resulting in a lag in temperature monitoring and risk assessment. This lag increases the risk of thermal runaway of the energy storage system under high-frequency load fluctuation conditions.Therefore, energy storage systems face multiple thermal management challenges in actual operation. The existence of these problems seriously affects the safety and reliability of the energy storage systems. There is an urgent need for a more advanced and effective thermal runaway risk assessment and management method to meet these complex thermal management requirements. Summary of the invention

[0003] The present invention solves the problems existing in the above-mentioned prior art and provides a distributed safety supervision system based on ordered energy storage big data, which mainly includes:

[0004] The thermal runaway risk assessment module is used to build a thermal runaway risk assessment model for the energy storage system based on historical thermal runaway event data and historical energy storage system operation data, judge the thermal runaway risk of the current energy storage system, and build a thermal runaway location prediction model for the energy storage system based on historical thermal runaway data to determine the thermal runaway location of the energy storage system;

[0005] A thermal runaway position management module is used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect;

[0006] The load fluctuation thermal runaway risk assessment module is used to build a load fluctuation thermal runaway risk assessment model based on historical load fluctuation data to determine the thermal runaway risk of the energy storage system under different load fluctuations;

[0007] The local overheating area migration analysis module is used to obtain historical load fluctuation data, historical local overheating area data and historical local overheating area migration data according to the operation log of the energy storage system, build a local overheating area migration prediction model, and predict the local overheating area location, local overheating area temperature and migration path within a preset time period in the future;

[0008] The local overheating area cooling execution module is used to adjust the cooling position, cooling power and cooling time of the local cooling device in real time according to the local overheating area position, local overheating area temperature, migration path and migration speed within a preset time period in the future of the energy storage system.

[0009] Furthermore, the thermal runaway risk assessment module is used to construct a thermal runaway risk assessment model for the energy storage system based on historical thermal runaway event data and historical energy storage system operation data, judge the thermal runaway risk of the current energy storage system, and construct a thermal runaway location prediction model for the energy storage system based on historical thermal runaway data to determine the thermal runaway location of the energy storage system, including:

[0010] Obtain historical thermal runaway event data through the operation log of the energy storage system. The historical thermal runaway event data includes the thermal runaway location of the historical thermal runaway event and the operation data of the energy storage system. The operation data of the energy storage system includes current, voltage and power. Obtain the operation data of the energy storage system under normal operation of the energy storage system through the operation log of the energy storage system, and preprocess the acquired historical thermal runaway event data and the operation data of the energy storage system under normal operation, including removing noise and outliers. Use the decision tree algorithm to train the model based on the preprocessed operation data of the energy storage system under normal operation of the energy storage system and the operation data of the energy storage system in the historical thermal runaway event data, and build a thermal runaway risk assessment model for the energy storage system. Obtain real-time energy storage system operation data through data acquisition equipment, and use the thermal runaway risk assessment model of the energy storage system to determine whether the current energy storage system has a thermal runaway risk. If the current energy storage system has a thermal runaway risk, use a recurrent neural network to train the model based on the historical thermal runaway data, build a thermal runaway location prediction model for the energy storage system, and determine the thermal runaway location of the energy storage system. Further, the thermal runaway position management module is used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect, including:

[0011] Through the data acquisition equipment, the real-time energy storage system operation data is obtained, and the energy storage system thermal runaway position prediction model is used to predict the energy storage system thermal runaway position; the infrared thermal imager is used to monitor the temperature of the position in real time to obtain the temperature data of the thermal runaway position, including the current temperature and the temperature rise rate; if the real-time monitored temperature is greater than the preset temperature threshold or the temperature rise rate is greater than the preset rate threshold, the control system sends a command to start the local cooling device for emergency cooling operation; the infrared thermal imager is used to obtain the real-time temperature monitoring data of the thermal runaway position, judge the cooling effect, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect, and the cooling effect includes excellent, good and poor; if the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to the relevant maintenance personnel.

[0012] Furthermore, the load fluctuation thermal runaway risk assessment module is used to construct a load fluctuation thermal runaway risk assessment model based on historical load fluctuation data to determine the thermal runaway risk of the energy storage system under different load fluctuations, including:

[0013] The load fluctuation frequency of the energy storage system is obtained in real time through data acquisition equipment to determine whether the load fluctuation frequency is greater than the preset frequency threshold; if the load fluctuation frequency of the energy storage system is greater than the preset frequency threshold, the historical load fluctuation data is obtained through the operation log of the energy storage system, and the thermal runaway risk status is marked for the historical load fluctuation data. The load fluctuation data includes average load, load fluctuation frequency, load fluctuation amplitude, maximum load, and minimum load. The thermal runaway risk status includes presence and absence; based on the historical load fluctuation data, the random forest algorithm is used for model training to build a load fluctuation thermal runaway risk assessment model to determine the thermal runaway risk of the energy storage system under different load fluctuations; the load fluctuation data of the energy storage system is obtained in real time through data acquisition equipment, and the load fluctuation thermal runaway risk assessment model is used to determine the thermal runaway risk of the energy storage system under the current load fluctuation.

[0014] Furthermore, the local overheating area migration analysis module is used to obtain historical load fluctuation data, historical local overheating area data and historical local overheating area migration data according to the operation log of the energy storage system, build a local overheating area migration prediction model, and predict the local overheating area position, local overheating area temperature and migration path within a preset time period in the future, including:

[0015] If the energy storage system under the current load fluctuation has the risk of thermal runaway, the historical load fluctuation data of the energy storage system in different time periods, as well as the historical local overheating area data and the historical local overheating area migration data in the corresponding time period are obtained through the operation log of the energy storage system. The local overheating area data includes the local overheating area position, the local overheating area temperature, and the timestamp, and the migration data includes the migration path and the migration speed. If the temperature change rate of the local overheating area is greater than the preset change rate threshold, the temperature of the local overheating area is subjected to temperature compensation processing. According to the historical load fluctuation data, the compensated historical local overheating area data and the historical local overheating area migration data, the long short-term memory network is used for model training to construct a local overheating area migration prediction model. The load fluctuation data and the local overheating area data of the energy storage system are obtained in real time through data acquisition equipment and infrared thermal imagers, and the temperature of the local overheating area whose temperature change rate is greater than the preset change rate threshold is subjected to temperature compensation processing. The local overheating area migration prediction model is used to predict the local overheating area position, the local overheating area temperature, the migration path and the migration speed in the future preset time period under the current load fluctuation.

[0016] The system also includes a local overheating area temperature compensation processing unit, which is used to perform temperature compensation processing on the temperature of the local overheating area if the temperature change rate of the local overheating area is greater than a preset change rate threshold, specifically including:

[0017] If the temperature change rate of the local overheating area is greater than the preset change rate threshold, the temperature change rate compensation formula Temperature compensation is performed on the temperature of the local overheated area. c (t) represents the compensated temperature at time t, T(t) represents the actual measured temperature at time t, represents the rate of temperature change at time t, β is a custom temperature compensation coefficient, and Δt is the time step.

[0018] Furthermore, the local overheating area cooling execution module is used to adjust the cooling position, cooling power and cooling time of the local cooling device in real time according to the local overheating area position, local overheating area temperature, migration path and migration speed within a future preset time period of the energy storage system, including:

[0019] According to the location of the local overheating area, the temperature of the local overheating area, the migration path and the migration speed within the future preset time period of the energy storage system, the control system sends instructions to start the local cooling device for cooling operation, and adjusts the cooling position, cooling power and cooling time of the local cooling device in real time; obtains real-time temperature monitoring data of the local overheating area location through an infrared thermal imager to judge the cooling effect, and adjusts the cooling power and cooling time of the local cooling device according to the cooling effect; if the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to relevant maintenance personnel.

[0020] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0021] The present invention provides a safety supervision system based on orderly energy storage big data distribution. The present invention accurately determines the thermal runaway risk of the current energy storage system by combining historical thermal runaway event data and historical energy storage system operation data, and accurately locates the location where thermal runaway may occur. This precise positioning and risk assessment helps to identify potential thermal runaway areas in advance, take targeted preventive measures, avoid accidents, and significantly improve the safety of the system. Based on the temperature data of the thermal runaway position, a local cooling device is used to perform emergency cooling operations, and the cooling power and cooling time are dynamically adjusted according to the cooling effect monitored in real time to ensure the efficiency and timeliness of the cooling operation and effectively control the risk of thermal runaway. By analyzing historical load fluctuation data, the thermal runaway risk of the energy storage system under different load fluctuations can be determined, the load management strategy can be optimized, the impact of load fluctuations on system safety can be reduced, and the stability of the system can be improved. The present invention can also predict the location, temperature and migration path of the local overheating area in the future preset time period by analyzing historical load fluctuation data, historical local overheating area data and historical local overheating area migration data, thereby realizing effective prediction and monitoring of overheating areas. If the temperature change rate of the local overheating area is greater than the preset change rate threshold, temperature compensation processing is performed to ensure the accuracy and real-time performance of temperature monitoring, further improving the accuracy of thermal runaway risk assessment and management. By adjusting the cooling position, cooling power and cooling time of the local cooling device in real time, efficient utilization of cooling resources is ensured, significantly improving the flexibility and effectiveness of cooling operations. By combining a variety of historical data and real-time monitoring, the present invention achieves all-round management and control of the thermal runaway risk of the energy storage system, significantly improving the safety and operational stability of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A flow chart of a distributed safety supervision system based on ordered energy storage big data of the present invention;

[0023] Figure 2 A schematic diagram of a distributed safety supervision system based on ordered energy storage big data according to the present invention;

[0024] Figure 3 It is another schematic diagram of a distributed safety supervision system based on ordered energy storage big data of the present invention; DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] like Figure 1-3 In this embodiment, a distributed safety supervision system based on ordered energy storage big data may specifically include:

[0027] Step S101, a thermal runaway risk assessment module is used to construct a thermal runaway risk assessment model for the energy storage system based on historical thermal runaway event data and historical energy storage system operation data, determine the thermal runaway risk of the current energy storage system, and construct a thermal runaway location prediction model for the energy storage system based on historical thermal runaway data to determine the thermal runaway location of the energy storage system.

[0028] Through the operation log of the energy storage system, historical thermal runaway event data is obtained. The historical thermal runaway event data includes the thermal runaway location of the historical thermal runaway event and the operation data of the energy storage system. The operation data of the energy storage system includes current, voltage, and power. Through the operation log of the energy storage system, the operation data of the energy storage system under normal operation of the energy storage system is obtained, and the historical thermal runaway event data and the energy storage system operation data under normal operation are preprocessed, including removing noise and outliers. According to the preprocessed energy storage system operation data under normal operation of the energy storage system and the energy storage system operation data in the historical thermal runaway event data, the decision tree algorithm is used for model training to build an energy storage system thermal runaway risk assessment model. Through the data acquisition equipment, real-time energy storage system operation data is obtained, and the energy storage system thermal runaway risk assessment model is used to determine whether the current energy storage system has a thermal runaway risk. If the current energy storage system has a thermal runaway risk, the recurrent neural network is used for model training based on the historical thermal runaway data to build an energy storage system thermal runaway location prediction model to determine the energy storage system thermal runaway location.

[0029] For example, through the operation log of the energy storage system, the data of 10 historical thermal runaway events in the past year were obtained, including the specific location of each thermal runaway event, such as the third battery cell in module A, and the operation data of the energy storage system when the thermal runaway event occurred, such as current, voltage and power. If the current recorded in a thermal runaway event is 150A, the voltage is 48V, and the power is 7200W. At the same time, the operation data of the energy storage system under normal operating conditions are extracted from the operation log, such as the current range of 50A to 100A, the voltage of 45V to 50V, and the power of 2250W to 5000W under normal conditions. These data are preprocessed to remove noise and outliers, and the decision tree algorithm is used for model training to construct a thermal runaway risk assessment model for the energy storage system. The current operating data of the energy storage system is obtained in real time through data acquisition equipment. If the model determines that the current current is 130A, the voltage is 47V, and the power is 6110W, and there is a risk of thermal runaway, then the model is trained using a recurrent neural network based on historical thermal runaway data to build a thermal runaway location prediction model for the energy storage system. Finally, it is predicted that the possible thermal runaway location of the energy storage system is in the second battery cell of module B.

[0030] Step S102, a thermal runaway position management module is used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect.

[0031] Through the data acquisition equipment, the real-time energy storage system operation data is obtained, and the energy storage system thermal runaway location prediction model is used to predict the energy storage system thermal runaway location. The infrared thermal imager is used to monitor the temperature of the location in real time to obtain the temperature data of the thermal runaway location, including the current temperature and the temperature rise rate. If the real-time monitored temperature is greater than the preset temperature threshold or the temperature rise rate is greater than the preset rate threshold, the control system sends a command to start the local cooling device for emergency cooling operation. The real-time temperature monitoring data of the thermal runaway location is obtained through the infrared thermal imager to determine the cooling effect, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect. The cooling effects include excellent, good and poor. If the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to the relevant maintenance personnel.

[0032] Exemplarily, by using a power quality analyzer, the operating data of the energy storage system is obtained in real time, and using the thermal runaway location prediction model, it is predicted that the location of thermal runaway of the energy storage system is the 5th battery cell of module C. The infrared thermal imager is used to monitor the temperature of the location in real time, and it is found that the current temperature is 85°C and the temperature rise rate is 2°C / minute. Since the real-time monitored temperature exceeds the preset temperature threshold of 80°C, and the temperature rise rate is also higher than the preset rate threshold of 1.5°C / minute, the control system immediately sends an instruction to start the local cooling device for emergency cooling operation. During the cooling process, the temperature data of the location is continuously monitored by the infrared thermal imager to determine the cooling effect. The initial cooling effect was evaluated as good, but after adjusting the cooling power and cooling time, it was found that the cooling effect still did not meet the expected standard, and the temperature dropped slowly and was still high. Therefore, an alarm signal is sent to the relevant maintenance personnel through the control system to notify them to take further emergency measures.

[0033] Step S103, a load fluctuation thermal runaway risk assessment module is used to construct a load fluctuation thermal runaway risk assessment model based on historical load fluctuation data to determine the thermal runaway risk of the energy storage system under different load fluctuations.

[0034] The load fluctuation frequency of the energy storage system is obtained in real time through data acquisition equipment to determine whether the load fluctuation frequency is greater than the preset frequency threshold. If the load fluctuation frequency of the energy storage system is greater than the preset frequency threshold, the historical load fluctuation data is obtained through the operation log of the energy storage system, and the thermal runaway risk status is marked for the historical load fluctuation data. The load fluctuation data includes average load, load fluctuation frequency, load fluctuation amplitude, maximum load, and minimum load. The thermal runaway risk status includes presence and absence. Based on the historical load fluctuation data, the random forest algorithm is used for model training to construct a load fluctuation thermal runaway risk assessment model to determine the thermal runaway risk of the energy storage system under different load fluctuations. The load fluctuation data of the energy storage system is obtained in real time through data acquisition equipment, and the load fluctuation thermal runaway risk assessment model is used to determine the thermal runaway risk of the energy storage system under the current load fluctuation.

[0035] Exemplarily, by using the power quality analyzer, the load fluctuation frequency data of the energy storage system is obtained in real time, and it is judged that the current load fluctuation frequency is 1.5Hz, which exceeds the preset frequency threshold of 1.0Hz. Since the load fluctuation frequency is greater than the preset frequency threshold, the historical load fluctuation data of the past month is obtained through the operation log of the energy storage system, and the thermal runaway risk status is marked, wherein the historical load fluctuation data 1 includes: the average load is 50kW, the load fluctuation frequency is 0.8Hz, the load fluctuation amplitude is 20kW, the maximum load is 70kW, the minimum load is 30kW, and the thermal runaway risk status is marked as existing. The historical load fluctuation data 2 includes: the average load is 49kW, the load fluctuation frequency is 0.81Hz, the load fluctuation amplitude is 22kW, the maximum load is 80kW, the minimum load is 40kW, and the thermal runaway risk status is marked as existing. The historical load fluctuation data 3 includes: the average load is 40kW, the load fluctuation frequency is 0.6Hz, the load fluctuation amplitude is 15kW, the maximum load is 55kW, the minimum load is 25kW, and the thermal runaway risk status is marked as not existing. Using these historical load fluctuation data, the random forest algorithm is used for model training to build a load fluctuation thermal runaway risk assessment model. The load fluctuation data of the energy storage system is obtained in real time through the power quality analyzer. The current load fluctuation data is obtained as an average load of 60kW, a load fluctuation frequency of 1.5Hz, a load fluctuation amplitude of 25kW, a maximum load of 85kW, and a minimum load of 40kW. The load fluctuation thermal runaway risk assessment model is used to determine that the system has a thermal runaway risk under the current load fluctuation.

[0036] Step S104, a local overheating area migration analysis module is used to obtain historical load fluctuation data, historical local overheating area data and historical local overheating area migration data according to the operation log of the energy storage system, build a local overheating area migration prediction model, and predict the local overheating area position, local overheating area temperature and migration path within a preset time period in the future.

[0037] If the energy storage system under the current load fluctuation has the risk of thermal runaway, the historical load fluctuation data of the energy storage system in different time periods, as well as the historical local overheating area data and historical local overheating area migration data in the corresponding time period are obtained through the operation log of the energy storage system. The local overheating area data includes the local overheating area position, local overheating area temperature, and timestamp, and the migration data includes the migration path and migration speed. If the temperature change rate of the local overheating area is greater than the preset change rate threshold, the temperature of the local overheating area is subjected to temperature compensation. According to the historical load fluctuation data, the compensated historical local overheating area data, and the historical local overheating area migration data, the long short-term memory network is used for model training to construct a local overheating area migration prediction model. The load fluctuation data and local overheating area data of the energy storage system are obtained in real time through data acquisition equipment and infrared thermal imagers, and the temperature of the local overheating area whose temperature change rate is greater than the preset change rate threshold is subjected to temperature compensation. The local overheating area migration prediction model is used to predict the local overheating area position, local overheating area temperature, migration path, and migration speed in the future preset time period under the current load fluctuation.

[0038] Exemplarily, by using a power quality analyzer, the load fluctuation data of the energy storage system is obtained in real time, and it is determined whether the energy storage system is at risk of thermal runaway under the current load fluctuation. Through the operation log of the energy storage system, the historical load fluctuation data of the past month and the historical local overheating area data in the corresponding time period are obtained, including data 4, at 14:00 on March 10, 2023, the temperature of the second battery cell of module A reached 85°C, the temperature change rate was 2.5°C / min, the migration path was from module A to module B, and the migration speed was 1cm / min. Data 5 is that at 16:30 on May 22, 2023, the temperature of the third battery cell of module B reached 92°C, the temperature change rate was 3.2°C / min, the migration path was from module B to module C, and the migration speed was 1.2cm / min. Data point 6 is that at 11:45 on August 5, 2023, the temperature of the 5th battery cell of module C reached 88°C, the temperature change rate was 2.8°C / min, the migration path was from module C to module D, and the migration speed was 0.9cm / min. Data point 7 is that at 09:20 on October 19, 2023, the temperature of the 4th battery cell of module D reached 95°C, the temperature change rate was 3.5°C / min, the migration path was from module D to module E, and the migration speed was 1.5cm / min. The temperature change rate of some local overheating areas is greater than the preset change rate threshold of 2°C / min, so the temperature data of these areas are temperature compensated. Subsequently, using historical load fluctuation data, historical local overheating area data, and historical local overheating area migration data, a long short-term memory network is used for model training to construct a local overheating area migration prediction model. Through the power quality analyzer and infrared thermal imager, the load fluctuation data and local overheating area data of the current energy storage system are obtained in real time. Using the trained local overheating area migration prediction model, it is predicted that under the current load fluctuation, the local overheating area will be formed in the 5th battery cell of module C within the next 30 minutes. The temperature is expected to reach 95°C, and the migration path will move from module B to module C, with a migration speed of 1.5cm / min.

[0039] Wherein, the local overheating area temperature compensation processing unit is used to perform temperature compensation processing on the temperature of the local overheating area if the temperature change rate of the local overheating area is greater than a preset change rate threshold.

[0040] If the temperature change rate of the local overheating area is greater than the preset change rate threshold, the temperature change rate compensation formula Temperature compensation is performed on the temperature of the local overheating area, where T c (t) represents the compensated temperature at time t, T(t) represents the actual measured temperature at time t, represents the rate of temperature change at time t, β is a custom temperature compensation coefficient, and Δt is the time step.

[0041] Exemplarily, by using a power quality analyzer, the load fluctuation data of the energy storage system is obtained in real time, and it is judged that the energy storage system is at risk of thermal runaway under the current load fluctuation. Through the operation log of the energy storage system, the historical load fluctuation data of the past month was obtained, and the historical local overheating area data in the corresponding time period was obtained using an infrared thermal imager. Data 8 shows that at 14:00 on March 10, 2023, the temperature of the second battery cell of module A reached 85°C, the temperature change rate was 2.5°C / min, the migration path was from module A to module B, and the migration speed was 1cm / min. Data 9 shows that at 16:30 on May 22, 2023, the temperature of the third battery cell of module B reached 92°C, the temperature change rate was 3.2°C / min, the migration path was from module B to module C, and the migration speed was 1.2cm / min. Data 10 shows that at 11:45 on August 5, 2023, the temperature of the 5th battery cell of module C reached 88°C, the temperature change rate was 2.8°C / min, the migration path was from module C to module D, and the migration speed was 0.9cm / min. Data 11 shows that at 09:20 on October 19, 2023, the temperature of the 4th battery cell of module D reached 95°C, the temperature change rate was 3.5°C / min, the migration path was from module D to module E, and the migration speed was 1.5cm / min. Therefore, the temperature change rate of the local overheating area corresponding to data 9 and data 11 is greater than the preset change rate threshold of 3°C / min, so the temperature data of these areas are temperature compensated. Use the temperature change rate compensation formula The temperature of the local overheating area is subjected to temperature compensation, where Tc(t) represents the compensation temperature at time t, and T(t) represents the actual measured temperature at time t. Indicates the rate of temperature change at time t, β is a custom temperature compensation coefficient, and Δt is the time step, such as one minute. Perform temperature compensation on data point 1. If β = 1.2 and Δt = 1 minute, the compensated temperature Tc(t) = 85 + 1.2 × 2.5 × 1 = 88°C.

[0042] Step S105, a local overheating area cooling execution module is used to adjust the cooling position, cooling power and cooling time of the local cooling device in real time according to the local overheating area position, local overheating area temperature, migration path and migration speed within a future preset time period of the energy storage system.

[0043] According to the location, temperature, migration path and migration speed of the local overheating area in the future preset time period of the energy storage system, the control system sends instructions to start the local cooling device for cooling operation, and adjusts the cooling position, cooling power and cooling time of the local cooling device in real time. Through the infrared thermal imager, the real-time temperature monitoring data of the local overheating area is obtained to judge the cooling effect, and the cooling power and cooling time of the local cooling device are adjusted according to the cooling effect. If the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to the relevant maintenance personnel.

[0044] Exemplarily, the load fluctuation data and local overheating area data of the current energy storage system are obtained in real time through the power quality analyzer and the infrared thermal imager. The trained local overheating area migration prediction model is used to predict that under the current load fluctuation, the local overheating area will be formed in the 5th battery cell of module C within the next 30 minutes. The temperature is expected to reach 95°C, and the migration path will move from module B to module C, with a migration speed of 1.5cm / min. According to the prediction results, the control system sends an instruction to start the local cooling device for cooling operation. The initial cooling position is set to the 5th battery cell of module C, the cooling power is 500W, and the cooling time is 10 minutes. The temperature monitoring data of the 5th battery cell of module C is obtained in real time through the infrared thermal imager. It is found that the temperature drops to 90°C after initial cooling, but the cooling effect is evaluated as good. Adjust the cooling power to 600W, extend the cooling time to 15 minutes, and perform the cooling operation again. Through subsequent monitoring, if it is found that the temperature drops slowly, only dropping to 88°C, and the cooling effect is assessed as poor, the system will send an alarm signal to the relevant maintenance personnel, notifying them to take further emergency measures immediately to prevent the thermal runaway situation from worsening.

[0045] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. The above exemplary features are replaced with the technical features with similar functions disclosed in the present application (but not limited to) to form a technical solution.

Claims

1. A distributed safety supervision system based on ordered energy storage big data, characterized in that: The system comprises: The thermal runaway risk assessment module is used to build a thermal runaway risk assessment model for the energy storage system based on historical thermal runaway event data and historical energy storage system operation data, judge the thermal runaway risk of the current energy storage system, and build a thermal runaway location prediction model for the energy storage system based on historical thermal runaway data to determine the thermal runaway location of the energy storage system; A thermal runaway position management module is used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect; The load fluctuation thermal runaway risk assessment module is used to build a load fluctuation thermal runaway risk assessment model based on historical load fluctuation data to determine the thermal runaway risk of the energy storage system under different load fluctuations; The local overheating area migration analysis module is used to obtain historical load fluctuation data, historical local overheating area data and historical local overheating area migration data according to the operation log of the energy storage system, build a local overheating area migration prediction model, and predict the local overheating area location, local overheating area temperature and migration path within a preset time period in the future; The local overheating area cooling execution module is used to adjust the cooling position, cooling power and cooling time of the local cooling device in real time according to the local overheating area position, local overheating area temperature, migration path and migration speed within a preset time period in the future of the energy storage system.

2. The system according to claim 1, wherein: The thermal runaway risk assessment module is used to construct a thermal runaway risk assessment model for the energy storage system based on historical thermal runaway event data and historical energy storage system operation data, judge the thermal runaway risk of the current energy storage system, and construct a thermal runaway location prediction model for the energy storage system based on historical thermal runaway data to determine the thermal runaway location of the energy storage system, and is characterized by: Obtain historical thermal runaway event data through the operation log of the energy storage system. The historical thermal runaway event data includes the thermal runaway location of the historical thermal runaway event and the operation data of the energy storage system. The operation data of the energy storage system includes current, voltage and power. Obtain the operation data of the energy storage system under normal operation of the energy storage system through the operation log of the energy storage system, and preprocess the acquired historical thermal runaway event data and the operation data of the energy storage system under normal operation, including removing noise and outliers. Use the decision tree algorithm to train the model based on the preprocessed operation data of the energy storage system under normal operation of the energy storage system and the operation data of the energy storage system in the historical thermal runaway event data, and build a thermal runaway risk assessment model for the energy storage system. Obtain real-time energy storage system operation data through data acquisition equipment, and use the thermal runaway risk assessment model of the energy storage system to determine whether the current energy storage system has a thermal runaway risk. If the current energy storage system has a thermal runaway risk, use a recurrent neural network to train the model based on the historical thermal runaway data, build a thermal runaway location prediction model for the energy storage system, and determine the thermal runaway location of the energy storage system.

3. The system according to claim 1, wherein: The thermal runaway position management module is used to use a local cooling device to perform emergency cooling operations according to the temperature data of the thermal runaway position, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect, and is characterized in that: Through the data acquisition equipment, the real-time energy storage system operation data is obtained, and the energy storage system thermal runaway position prediction model is used to predict the energy storage system thermal runaway position; the infrared thermal imager is used to monitor the temperature of the position in real time to obtain the temperature data of the thermal runaway position, including the current temperature and the temperature rise rate; if the real-time monitored temperature is greater than the preset temperature threshold or the temperature rise rate is greater than the preset rate threshold, the control system sends a command to start the local cooling device for emergency cooling operation; the infrared thermal imager is used to obtain the real-time temperature monitoring data of the thermal runaway position, judge the cooling effect, and adjust the cooling power and cooling time of the local cooling device according to the cooling effect, and the cooling effect includes excellent, good and poor; if the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to the relevant maintenance personnel.

4. The system according to claim 1, wherein: The load fluctuation thermal runaway risk assessment module is used to construct a load fluctuation thermal runaway risk assessment model based on historical load fluctuation data to determine the thermal runaway risk of the energy storage system under different load fluctuations, and is characterized by: The load fluctuation frequency of the energy storage system is obtained in real time through data acquisition equipment to determine whether the load fluctuation frequency is greater than the preset frequency threshold; if the load fluctuation frequency of the energy storage system is greater than the preset frequency threshold, the historical load fluctuation data is obtained through the operation log of the energy storage system, and the thermal runaway risk status is marked for the historical load fluctuation data. The load fluctuation data includes average load, load fluctuation frequency, load fluctuation amplitude, maximum load, and minimum load. The thermal runaway risk status includes presence and absence; based on the historical load fluctuation data, the random forest algorithm is used for model training to build a load fluctuation thermal runaway risk assessment model to determine the thermal runaway risk of the energy storage system under different load fluctuations; the load fluctuation data of the energy storage system is obtained in real time through data acquisition equipment, and the load fluctuation thermal runaway risk assessment model is used to determine the thermal runaway risk of the energy storage system under the current load fluctuation.

5. The system according to claim 1, wherein: The local overheating area migration analysis module is used to obtain historical load fluctuation data, historical local overheating area data and historical local overheating area migration data according to the operation log of the energy storage system, build a local overheating area migration prediction model, and predict the local overheating area position, local overheating area temperature and migration path within a preset time period in the future, and is characterized by: If the energy storage system under the current load fluctuation has the risk of thermal runaway, the historical load fluctuation data of the energy storage system in different time periods, as well as the historical local overheating area data and the historical local overheating area migration data in the corresponding time period are obtained through the operation log of the energy storage system. The local overheating area data includes the local overheating area position, the local overheating area temperature, and the timestamp, and the migration data includes the migration path and the migration speed. If the local overheating area temperature change rate is greater than the preset change rate threshold, the local overheating area temperature is subjected to temperature compensation processing by the local overheating area temperature compensation processing unit. According to the historical load fluctuation data, the compensated historical local overheating area data and the historical local overheating area migration data, the long short-term memory network is used to perform model training and construct a local overheating area migration prediction model. The load fluctuation data and the local overheating area data of the energy storage system are obtained in real time through the data acquisition equipment and the infrared thermal imager, and the temperature compensation processing is performed on the local overheating area temperature whose temperature change rate is greater than the preset change rate threshold. The local overheating area migration prediction model is used to predict the local overheating area position, the local overheating area temperature, the migration path and the migration speed in the future preset time period under the current load fluctuation.

6. The system according to claim 5, wherein: The local overheating area temperature compensation processing unit is used to perform temperature compensation processing on the local overheating area temperature if the local overheating area temperature change rate is greater than a preset change rate threshold, and is characterized in that: If the temperature change rate of the local overheating area is greater than the preset change rate threshold, the temperature change rate compensation formula Temperature compensation is performed on the temperature of the local overheating area, where T c (t) represents the compensated temperature at time t, T(t) represents the actual measured temperature at time t, represents the rate of temperature change at time t, β is a custom temperature compensation coefficient, and Δt is the time step.

7. The system according to claim 1, wherein: The local overheating area cooling execution module is used to adjust the cooling position, cooling power and cooling time of the local cooling device in real time according to the local overheating area position, local overheating area temperature, migration path and migration speed within a future preset time period of the energy storage system, and is characterized by: According to the location of the local overheating area, the temperature of the local overheating area, the migration path and the migration speed within the future preset time period of the energy storage system, the control system sends instructions to start the local cooling device for cooling operation, and adjusts the cooling position, cooling power and cooling time of the local cooling device in real time; obtains real-time temperature monitoring data of the location of the local overheating area through an infrared thermal imager, judges the cooling effect, and adjusts the cooling power and cooling time of the local cooling device according to the cooling effect; if the cooling effect of the adjusted local cooling device still does not meet expectations, an alarm signal is sent to relevant maintenance personnel.

Images