Multi-parameter early warning monitoring system and method for thermal runaway of energy storage battery

Abstract

The application relates to the technical field of energy storage system monitoring and control, in particular to a multi-parameter early warning monitoring system and method for thermal runaway of energy storage batteries; the system comprises a data acquisition module, an early warning analysis module, a decision control module and an execution feedback module; the data acquisition module acquires state parameters and health parameters; the early warning analysis module establishes a three-level risk assessment mechanism based on the state parameters and the health parameters, and outputs a graded early warning signal; the decision control module generates a protection instruction based on the graded early warning signal, including: an optimized cooling instruction for first-level protection, a composite trigger instruction for second-level protection and a forced start instruction for third-level protection; the execution feedback module executes the protection instruction and monitors the execution state of the protection instruction; the three-level risk assessment mechanism is optimized based on the execution state; the application realizes early pre-control of thermal runaway of energy storage batteries, improves the accuracy of early warning, and enhances the synergy and reliability of the protection system.

Description

Technical Field

[0001] This application relates to the field of energy storage system monitoring and control technology, specifically to a multi-parameter early warning monitoring system and method for thermal runaway of energy storage batteries. Background Technology

[0002] With the rapid development of the new energy industry, energy storage power stations, as core equipment for energy storage and peak shaving, are increasingly widely used in scenarios such as grid absorption improvement, distributed energy support, and microgrid regulation. They have become a key infrastructure supporting energy structure transformation and the safe and stable operation of the power system. Energy storage batteries, as the core energy carrier and core component of energy storage power stations, directly determine the overall reliability, operational safety, and service life of the station, making them a core control target for efficient operation and maintenance of energy storage systems. However, under high-rate charging and discharging, long-term cycle aging, extreme operating conditions, and various fault conditions, energy storage batteries are highly susceptible to triggering electrochemical chain exothermic reactions, i.e., thermal runaway. This is accompanied by safety risks such as sudden temperature rise, release of toxic and flammable gases, fire, and even explosion, seriously threatening the property safety of energy storage power station equipment and the lives of on-site maintenance personnel, and hindering the large-scale and high-quality development of the energy storage industry.

[0003] Existing energy storage battery safety protection systems generally employ traditional single-parameter monitoring and triggering mechanisms. Their protection logic is relatively simple and difficult to adapt to the complex evolution and multi-dimensional characteristics of thermal runaway. Specifically, traditional protection systems rely solely on single physical parameters such as temperature and smoke as early warning and protection trigger conditions. For example, they collect individual cell temperature signals using temperature sensors attached to the battery surface, or capture smoke characteristics in the early stages of combustion using smoke detectors. However, their ability to capture the evolution of thermal runaway is significantly limited. The occurrence and development of thermal runaway exhibit distinct stages, from the early latent state of intensified internal battery reactions and trace gas production, to the violent reaction stage of a sudden temperature rise and smoke release. There exists a critical early warning window for emergency response. Single-parameter monitoring has extremely low sensitivity to the characteristic signals of the early latent stage of thermal runaway, often only triggering an early warning after the thermal runaway has entered the violent reaction stage. By this time, the early warning window has been significantly compressed, making it difficult for protective measures to effectively manage the situation.

[0004] Meanwhile, traditional protection systems lack efficient coordination between the early warning and protection execution stages, resulting in significant response lags. In most existing systems, after an early warning signal is issued, cumbersome procedures such as manual confirmation and command issuance are required before protective measures such as cooling, fire suppression, and battery module isolation can be activated, failing to achieve automated and rapid linkage between early warning and protection actions. Furthermore, various protection modules, such as temperature monitoring modules, smoke monitoring modules, fire suppression systems, and ventilation control systems, often operate independently, lacking cross-module data interaction and collaborative control, making it difficult to construct a comprehensive, three-dimensional protection system. In addition, traditional systems suffer from insufficient parameter monitoring accuracy, making them susceptible to interference from complex operating environments, leading to false alarms and missed alarms: temperature sensors are prone to measurement drift in low-temperature environments, and environmental factors such as dust and moisture can cause smoke detectors to mis-trigger, while weak early signs of thermal runaway may be overlooked. This not only reduces the efficiency of power plant operation and maintenance but also creates serious safety hazards, failing to meet the high safety protection requirements of large-scale energy storage power plants.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this application is to provide a multi-parameter early warning monitoring system and method for thermal runaway of energy storage batteries, which to some extent solves the problems raised in the background technology, realizes early control of thermal runaway of energy storage batteries, not only improves the accuracy of early warning, but also enhances the synergy and reliability of the protection system.

[0007] To achieve the above objectives, this application provides the following technical solution:

[0008] In the first aspect, this application provides a multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries, including a data acquisition module, an early warning analysis module, a decision control module and an execution feedback module;

[0009] The data acquisition module is used to collect the state parameters of the energy storage battery and the health parameters of the protection system, and to preprocess the state parameters and health parameters.

[0010] Based on the aforementioned status parameters and health parameters, the early warning analysis module establishes a three-level risk assessment mechanism and outputs graded early warning signals based on the results of the risk assessment.

[0011] The decision control module generates protection commands based on the graded early warning signals, including: optimized cooling commands for level 1 protection, composite trigger commands for level 2 protection, and forced start commands for level 3 protection.

[0012] The execution feedback module is used to execute the protection instructions and monitor the execution status of the protection instructions; and optimize the three-level risk assessment mechanism based on the execution status.

[0013] As a preferred embodiment of the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries described in this application, the early warning analysis module is configured with a three-level risk assessment mechanism; the three-level risk assessment mechanism includes: a first-level risk assessment based on cell temperature, a second-level risk assessment based on packing pressure, and a third-level risk assessment based on health parameters; the steps of the first-level risk assessment specifically include:

[0014] Set a historical time window; within the historical time window, collect the cell temperature of any battery at a first fixed frequency, and calculate the average temperature of all cells of the corresponding battery to obtain the corresponding temperature baseline value.

[0015] Calculate the difference between the cell temperature of any battery at the current sampling time and the corresponding temperature baseline value to obtain the temperature deviation at the current sampling time;

[0016] The first-level risk index of the corresponding battery is obtained by weighting and summing the rate of change of cell temperature, temperature deviation, and total charge / discharge rate.

[0017] As a preferred embodiment of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries described in this application, the first-level risk assessment step further includes:

[0018] A cell temperature prediction model is constructed, and the predicted cell temperature value within the prediction time period is calculated based on the cell temperature prediction model.

[0019] If the first-level risk index is less than the preset first-level risk index threshold, and the predicted cell temperature within the prediction time period is less than the preset cell temperature threshold, then the corresponding battery will be determined to be risk-free.

[0020] If the first level risk index is greater than or equal to the first level risk index threshold and less than the preset second level risk index threshold, or if the cell temperature prediction value is less than the preset first proportion but greater than or equal to the cell temperature threshold, then the corresponding battery will be classified as medium risk.

[0021] If the first-level risk index is greater than or equal to the second-level risk index threshold, or if the predicted cell temperature value of the first proportion is greater than or equal to the cell temperature threshold, then the corresponding battery is determined to be high-risk.

[0022] As a preferred embodiment of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries described in this application, the steps of the second-level risk assessment specifically include:

[0023] Set a historical statistics window; obtain the air pressure of each battery compartment under normal operating conditions within the historical statistics window, and calculate the average value of the air pressure to obtain the corresponding air pressure baseline value;

[0024] Calculate the difference between the current sampling time's chamber pressure and the corresponding baseline pressure value to obtain the pressure deviation;

[0025] Based on the first-level risk index of all batteries in any battery compartment, calculate the risk correlation factor of the corresponding battery compartment, and then perform a weighted summation of the air pressure deviation, the rate of change of air pressure in the compartment, and the risk correlation factor to obtain the second-level risk index of the corresponding battery compartment.

[0026] Based on the aforementioned air pressure in the insertion chamber and the second-level risk index, a second-level risk assessment is conducted.

[0027] As a preferred embodiment of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries described in this application, the third-level risk assessment steps specifically include:

[0028] Health scores are assigned to all critical components within each battery compartment to obtain the health score for each critical component.

[0029] Weighting coefficients are set based on the importance of each key component, and the health scores of the key components of the same battery compartment are weighted and summed according to the weighting coefficients to obtain the comprehensive health score of the corresponding battery compartment.

[0030] The risk intensity is set based on the results of the second-level risk assessment, and the third-level risk score of each battery compartment is calculated based on the risk intensity and the comprehensive health score.

[0031] A third-level risk assessment is conducted based on the aforementioned third-level risk score.

[0032] As a preferred embodiment of the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries described in this application, the decision control module is configured with a protection command generation strategy; the protection command generation strategy specifically includes:

[0033] When the graded early warning signal is a level one early warning signal, an optimized cooling command for level one protection is generated; the optimized cooling command specifically includes command type, target location and cooling strategy;

[0034] The optimized cooling command is a preventative cooling command, and the target location is the target battery compartment corresponding to the battery compartment number in the first-level warning signal; the cooling strategy includes flow regulation and temperature regulation.

[0035] The flow rate adjustment is to increase the refrigerant circulation flow rate to the target battery compartment to a preset ratio of the rated value; the temperature adjustment is to call the refrigerant stored in the intelligent refrigerant box to reduce the inlet temperature of the refrigerant flowing into the battery compartment to a preset temperature safety threshold.

[0036] As a preferred embodiment of the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries described in this application, the protection command generation strategy further includes:

[0037] When the graded early warning signal is a level two early warning signal, a composite triggering command for level two protection is generated; the composite triggering command specifically includes command type, target location and command content;

[0038] The instruction type of the composite triggering command is composite triggering, and the target location is the target battery box corresponding to the battery box number in the secondary warning signal; the instruction content includes pre-start and mixed liquid injection;

[0039] The pre-start optimization is as follows: set the fire pump corresponding to the target battery box to standby mode, and increase the pressure of the fire pump from the current pressure value to the preset pressure value;

[0040] The mixture injection is performed by injecting a mixture of stabilizing agent and fire water into the target battery compartment under the pressure of the fire pump, thereby submerging the battery cells in the target battery compartment.

[0041] As a preferred embodiment of the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries described in this application, the protection command generation strategy further includes:

[0042] When the tiered early warning signal is a level three early warning signal, an emergency activation command for level three protection is generated; the emergency activation command specifically includes the command type, target location, and command content;

[0043] The emergency start command is a forced emergency start command, and the target location is the target battery compartment corresponding to the battery compartment number in the level 3 warning signal.

[0044] The instructions include forced pressure increase and emergency access protection; the forced pressure increase means: increasing the outlet pressure of the fire pump corresponding to the target battery box and stabilizing it at a preset pressure value;

[0045] The emergency access guarantee is achieved by: opening the main valve of the emergency fire water pipeline leading to the target battery compartment, or sending a heating signal to the temperature sensing safety valve of the target battery compartment, and heating the temperature sensing safety valve to the opening temperature.

[0046] As a preferred embodiment of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries described in this application, the cell temperature prediction model includes a linear trend term, an operating condition correction term, and a historical residual term.

[0047] Calculate the product of the rate of change of the cell temperature at the current sampling moment and the predicted time length, and use the sum of the product and the cell temperature at the current sampling moment as the linear trend term;

[0048] The change in heat generation power is calculated based on the total charge / discharge rate, and the heat generation power correction component is calculated based on the change in heat generation power.

[0049] Obtain the inlet temperature of the refrigerant at the current sampling time, and calculate the heat dissipation correction component based on the inlet temperature; add the heat generation power correction component and the heat dissipation correction component to obtain the operating condition correction term;

[0050] Construct a historical deviation database; extract the operating condition feature vector at the current sampling time, and calculate the Euclidean distance between the operating condition feature vector and all operating condition feature vectors in the historical deviation database;

[0051] Based on the Euclidean distance, similar feature vectors are selected, and the predicted residual values ​​corresponding to the similar feature vectors are weighted and summed to obtain the historical residual term at the current sampling time.

[0052] Secondly, this application provides a multi-parameter early warning and monitoring method for thermal runaway of energy storage batteries, including the following steps:

[0053] Collect the state parameters of the energy storage battery and the health parameters of the protection system, and preprocess the state parameters and health parameters;

[0054] Based on the aforementioned state parameters and health parameters, a three-level risk assessment mechanism is established, and graded early warning signals are output based on the results of the risk assessment.

[0055] Based on the graded early warning signals, protection instructions are generated, including: optimized cooling instructions for level 1 protection, composite triggering instructions for level 2 protection, and emergency activation instructions for level 3 protection.

[0056] Execute the protection instructions and monitor their execution status; optimize the three-level risk assessment mechanism based on the execution status.

[0057] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0058] By collecting the state parameters of the energy storage battery and the health parameters of the protection system, and establishing a three-level risk assessment mechanism based on these parameters, a graded early warning signal is output based on the risk assessment results. This solves the problems of delayed early warning, false alarms, and missed alarms in traditional single-parameter monitoring and early warning, and enables early control of thermal runaway of the energy storage battery. Based on the graded early warning signal, protection commands are generated, including: optimized cooling commands for level 1 protection, composite trigger commands for level 2 protection, and emergency start commands for level 3 protection. The protection commands are executed, and their execution status is monitored. Based on the execution status, the three-level risk assessment mechanism is optimized, achieving three-level protection and precise full-cycle control of thermal runaway of the energy storage battery. This overcomes the limitations of traditional separation between early warning and protection, and improves the synergy and reliability of the protection system. Attached Figure Description

[0059] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0060] Figure 1 This is a structural diagram of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries provided in this application;

[0061] Figure 2 A flowchart of the multi-parameter early warning monitoring method for thermal runaway of energy storage batteries provided in this application. Detailed Implementation

[0062] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0063] Example 1

[0064] like Figure 1 As shown in the figure, this embodiment introduces a multi-parameter early warning monitoring system for thermal runaway of energy storage batteries, including a data acquisition module, an early warning analysis module, a decision control module, and an execution feedback module; the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries is applied to an energy storage battery device equipped with a fully silent liquid cooling system; the fully silent liquid cooling system includes at least an intelligent refrigerant box, a circulation pipeline, and an independent heat exchanger installed inside the battery compartment;

[0065] The data acquisition module is used to collect the state parameters of the energy storage battery and the health parameters of the protection system, and to preprocess the state parameters and health parameters; the state parameters of the energy storage battery include cell temperature and rate of change, pack pressure and rate of change, and battery pack parameters;

[0066] The state parameters of the energy storage battery are acquired as follows: a digital temperature sensor, such as an NTC thermistor, is configured within the energy storage battery module; in this embodiment, an NTC thermistor is installed at the terminal connection of each battery; the NTC thermistor is used to measure the surface temperature of each battery cell at a first fixed frequency, such as 1Hz, to obtain the cell temperature, and the timestamp of each sampling is recorded; the ratio of the difference between the cell temperature at the current sampling moment and the cell temperature at the previous sampling moment to the difference of the corresponding timestamp is calculated to obtain the rate of change of the cell temperature at the current sampling moment; in each battery... A micro differential pressure sensor, such as a MEMS piezoresistive sensor, is installed on the sealed cavity of the battery compartment. The micro differential pressure sensor measures the air pressure inside each battery compartment at a second fixed frequency, such as 2Hz, to obtain the compartment air pressure, and records the timestamp of each sampling. The ratio of the difference between the compartment air pressure at the current sampling moment and the compartment air pressure at the previous sampling moment to the difference of the corresponding timestamp is calculated to obtain the rate of change of the compartment air pressure at the current sampling moment. The battery pack parameters include the total charge / discharge rate and state of charge of the battery pack, which are directly read from the main controller of the battery management system through a standard communication interface.

[0067] The health parameters of the protection system are collected as follows: photoelectric smoke sensors are installed at the top of each battery compartment or at locations where smoke tends to accumulate. The system collects alarm output signals from the passive dry contacts of the photoelectric smoke sensors, 4-20mA analog outputs, and reports self-test status and fault codes via the bus in real time. It also acquires feedback status of the intelligent refrigerant circulation pump, such as start / stop feedback and overload alarms, as well as key valve position signals. Additionally, it acquires feedback status of the fire pump, such as start / stop feedback and abnormal pressure alarms, and valve position switch feedback signals of the pipeline solenoid valves.

[0068] The specific preprocessing method is as follows: apply a first-order low-pass filter or moving average filter to analog signals such as cell temperature and box pressure to suppress high-frequency noise and ensure the stability of subsequent calculations; remove abnormal calculated values ​​caused by signal jumps from the calculated cell temperature change rate and box pressure change rate.

[0069] The early warning analysis module establishes a three-level risk assessment mechanism based on the aforementioned status parameters and health parameters, and outputs graded early warning signals based on the risk assessment results. The three-level risk assessment mechanism specifically includes: a first-level risk assessment based on cell temperature, a second-level risk assessment based on charging box pressure, and a third-level risk assessment based on health parameters. The first-level risk assessment based on cell temperature is as follows:

[0070] Set a historical time window, such as the past month; when the cooling system is running normally and the battery is in a healthy charging and discharging state, collect the cell temperature of any battery at the first fixed frequency within the historical time window, and calculate the average temperature of all cells of the corresponding battery to obtain the temperature baseline value of the corresponding battery.

[0071] The difference between the cell temperature at the current sampling moment and the corresponding temperature baseline value of any battery is calculated, i.e., the cell temperature at the current sampling moment minus the corresponding temperature baseline value, to obtain the temperature deviation at the current sampling moment. The cell temperature change rate, temperature deviation, and total charge / discharge rate are weighted and summed to obtain the first-level risk index of the corresponding battery. The weighting coefficients are set according to the battery type and the importance of the cell temperature change rate, temperature deviation, and total charge / discharge rate. For example, the cell temperature change rate directly reflects the instantaneous abnormal heat generation rate inside the battery, so the weighting coefficient for the cell temperature change rate is set to 0.5; the temperature deviation reflects the degree to which the battery has deviated from its healthy baseline and is a manifestation of cumulative risk, so the weighting coefficient for the temperature deviation is set to 0.3; the total charge / discharge rate reflects the macroscopic operating load of the system and is the background and triggering factor for thermal runaway risk, so the weighting coefficient for the total charge / discharge rate is set to 0.2.

[0072] A cell temperature prediction model is constructed, and the predicted cell temperature value within the prediction time period is calculated based on the cell temperature prediction model; for example, the predicted cell temperature value within the next 30 seconds is calculated; the cell temperature prediction model includes a linear trend term, an operating condition correction term, and a historical residual term; the linear trend term is calculated by multiplying the rate of change of the cell temperature at the current sampling time by the prediction time period, and then summing the product with the cell temperature at the current sampling time as the linear trend term;

[0073] The calculation method for the operating condition correction term is as follows: Calculate the difference between the square of the total charge / discharge rate at the current sampling time and the square of the total charge / discharge rate at the previous sampling time, i.e., the square of the total charge / discharge rate at the current sampling time minus the square of the total charge / discharge rate at the previous sampling time, to obtain the change in heat generation power; calculate the product of the change in heat generation power and a preset first proportional coefficient, and multiply the product by the predicted time length to obtain the heat generation power correction component; the first proportional coefficient is positively correlated with the battery's internal resistance and negatively correlated with the battery's heat capacity; obtain the inlet temperature of the refrigerant at the current sampling time, and calculate the difference between the cell temperature and the coolant inlet temperature at the current sampling time, i.e., the change in heat generation power at the current sampling time. The cell temperature at a given moment is subtracted from the refrigerant inlet temperature to obtain the first instantaneous temperature difference. The refrigerant inlet temperature at the previous sampling moment is obtained, and the difference between the cell temperature and the refrigerant inlet temperature at the previous sampling moment is calculated, i.e., the cell temperature at the previous sampling moment minus the coolant inlet temperature, to obtain the second instantaneous temperature difference. The difference between the first instantaneous temperature difference and the second instantaneous temperature difference is calculated, i.e., the first instantaneous temperature difference minus the second instantaneous temperature difference, and the difference is multiplied by a preset second proportional coefficient to obtain a heat dissipation correction component. The second proportional coefficient is negatively correlated with the equivalent thermal resistance of the fully silent liquid cooling system. The heat generation power correction component and the heat dissipation correction component are added together to obtain the operating condition correction term.

[0074] The historical residual term is calculated as follows: An updatable historical deviation database is constructed and maintained. This database stores multiple historical data records, each including at least the operating condition feature vector at the historical sampling time and the predicted residual value at that time. The operating condition feature vector consists of the total charge / discharge rate, refrigerant inlet temperature, and battery state of charge at the historical sampling time. The predicted residual value is calculated by: calculating the predicted cell temperature at the historical sampling time based on a linear trend term and an operating condition correction term, and obtaining the actual measured cell temperature at the historical sampling time; calculating the difference between the actual measured cell temperature and the predicted cell temperature, i.e., the actual measured cell temperature minus the predicted cell temperature. The predicted value is obtained by calculating the predicted residual value at the corresponding historical sampling time. The operating condition feature vector at the current sampling time is extracted, and the Euclidean distance between the operating condition feature vector and all operating condition feature vectors in the historical deviation database is calculated. The Euclidean distances are sorted in descending order, and the operating condition feature vectors in the historical deviation database corresponding to the top k Euclidean distances are extracted as similar feature vectors; where k is a positive integer, such as k=3. The predicted residual values ​​corresponding to the similar feature vectors are weighted and summed to obtain the historical residual term at the current sampling time. The weight coefficient of the predicted residual value corresponding to the similar feature vector is inversely proportional to the corresponding Euclidean distance; the smaller the Euclidean distance, the larger the weight coefficient of the corresponding predicted residual value.

[0075] A first-level risk assessment is performed based on the first-level risk index and the predicted cell temperature, specifically including: if the first-level risk index is less than a preset first-level risk index threshold, and the predicted cell temperature within the prediction time period is less than the preset cell temperature threshold, then the corresponding battery is determined to be risk-free; if the first-level risk index is greater than or equal to the first-level risk index threshold, and less than a preset second-level risk index threshold, or less than a preset first proportion of the predicted cell temperature is greater than or equal to the cell temperature threshold, then the corresponding battery is determined to be medium-risk; if the first-level risk index is greater than or equal to the second-level risk index threshold, or greater than or equal to the first proportion of the predicted cell temperature is greater than or equal to the cell temperature threshold, then the corresponding battery is determined to be medium-risk. The battery is classified as high-risk; the first-level risk index threshold is less than the second-level risk index threshold; the first-level risk index threshold is used to identify early anomalies, i.e., the cell temperature deviates from the normal range but has not yet reached a dangerous level, and is obtained by statistical analysis of temperature deviation and change rate under historical normal operation; the second-level risk index threshold is used to identify high-risk states, i.e., the temperature anomaly is significant and may be about to trigger thermal runaway, and is obtained by conducting thermal runaway experimental simulation; the cell temperature threshold is determined according to specific monitoring needs; the first ratio is used to determine whether the predicted cell temperature value is generally too high, and is obtained by statistical analysis of the prediction error of the cell temperature prediction model; in this embodiment, the cell temperature threshold is 40℃, and the first ratio is 40%.

[0076] The second-level risk assessment based on the chamber pressure is as follows:

[0077] Set a historical statistics window, such as the past two weeks; obtain the air pressure of all battery compartments under normal operating conditions within the historical statistics window, and calculate the average value of the air pressure of the battery compartments to obtain the air pressure baseline value of the corresponding battery compartments; the normal operating condition refers to a healthy state without triggering any warning signals; calculate the difference between the air pressure of the battery compartment at the current sampling time and the corresponding air pressure baseline value, that is, the air pressure of the battery compartment at the current sampling time minus the corresponding air pressure baseline value to obtain the air pressure deviation;

[0078] The risk correlation factor of the corresponding battery compartment is calculated based on the first-level risk index of all batteries in any battery compartment, and the pressure deviation, the rate of change of the compartment pressure and the risk correlation factor are weighted and summed to obtain the second-level risk index of the corresponding battery compartment; the weight coefficients of the pressure deviation, the rate of change of the compartment pressure and the risk correlation factor are set by professionals according to the monitoring requirements.

[0079] In this embodiment, the risk association factor is calculated as follows: the first-level risk indices of all batteries in the battery compartment are sorted in descending order, and the top N first-level risk indices are extracted; where N is a positive integer, for example, N=3; the average value of the top N first-level risk indices is calculated to obtain the risk association factor; optionally, the risk association factor can also be calculated based on the proportion of medium-risk or high-risk batteries in the battery compartment to all batteries in the battery compartment.

[0080] Based on the battery compartment air pressure and the second-level risk index, a second-level risk assessment is performed, specifically including: if the second-level risk index is less than a preset third-level risk index threshold, and all batteries in the battery compartment are classified as risk-free, then the corresponding battery compartment is determined to be risk-free; if the second-level risk index is greater than or equal to the third-level risk index threshold, but less than a preset fourth-level risk index threshold, and at least one battery in the corresponding battery compartment is classified as medium-risk or high-risk, then the corresponding battery compartment is determined to be medium-risk; if the second-level risk index is greater than or equal to the fourth-level risk index threshold, and at least one battery in the corresponding battery compartment is classified as high-risk, then the corresponding battery compartment is determined to be high-risk. If the second-level risk index is greater than or equal to the fourth risk index threshold, and the gas pressure of the battery compartment at the current sampling time and the previous M sampling times is greater than the preset gas pressure threshold, then the corresponding battery compartment is judged as high-risk; where M is a positive integer, such as M=10; the third risk index threshold is less than the fourth risk index threshold; the third risk index threshold is used to identify early gas pressure anomalies in the battery compartment, and is obtained by statistical analysis of gas pressure deviation and change rate under historical normal operating conditions; the fourth risk index threshold is used to identify serious anomalies in the battery compartment, such as obvious gas production or sealing failure, and is obtained through thermal runaway experiment simulation; the gas pressure threshold is set according to specific monitoring requirements.

[0081] The third-level risk assessment based on health parameters is as follows:

[0082] A health score is calculated for all critical components in each battery compartment to obtain a health score for each critical component. The critical components include photoelectric smoke sensors, fire pumps, and fire pipeline valves. The health score ranges from [0,1]. A health score of 1 indicates that the corresponding critical component is completely healthy, and a health score of 0 indicates that the corresponding critical component is completely faulty.

[0083] In this embodiment, the specific method for health scoring is as follows: If the photoelectric smoke sensor reports a fault code or communication with the data acquisition module is interrupted, the health score of the photoelectric smoke sensor is set to 0; if the photoelectric smoke sensor has no fault code and its self-test status is passed, the basic health score of the photoelectric smoke sensor is set to 1, and the deduction items for the photoelectric smoke sensor are calculated: if the 4-20mA analog output signal of the photoelectric smoke sensor remains at the lower or upper limit of the range for a preset time period of more than 30 seconds, 0.4 will be deducted; if the photoelectric smoke sensor has generated false alarms in the past week... If there is no thermal runaway but an alarm is triggered, or if a thermal runaway occurs but no alarm is recorded, 0.2 points will be deducted for each false alarm or missed alarm, with a maximum deduction of 0.6 points. The cumulative sum of deductions for the photoelectric smoke sensor will be calculated, and the difference between the photoelectric smoke sensor's base health score and the cumulative sum will be calculated. That is, the photoelectric smoke sensor's base health score minus the cumulative sum will yield the photoelectric smoke sensor's health score. If a fire pump experiences a fault alarm, the fire pump's health score will be set to 0. If the fire pump does not experience a fault alarm, the fire pump's base health score will be set to 1, and the fire pump's deductions will be calculated. After receiving a start or stop command, if the fire pump fails to provide the corresponding operating or stop status signal within the specified time (e.g., within 2 seconds), 0.3 points will be deducted; if the outlet pressure of the fire pump fails to reach the target pressure value required by the corresponding command (e.g., 0.3 MPa) within the set time after starting, 0.4 points will be deducted; the cumulative deductions for the fire pumps will be calculated, and the difference between the fire pump's basic health score and the cumulative sum will be calculated, i.e., the fire pump's basic health score minus the cumulative sum will yield the fire pump's health score; if a fire pipeline valve experiences a fault alarm, the fire pipeline valve's health score will be set to 0; if the fire... If there is no fault alarm in the fire protection pipeline valve, the basic health score of the fire protection pipeline valve is set to 1, and the deduction items for the fire protection pipeline valve are calculated: when an open or close command is received, if the valve position switch does not change to the corresponding open or closed state within the set time, such as not changing to the corresponding open or closed state within 2 seconds, 0.5 points are deducted; if the response time of the fire protection pipeline valve exceeds the normal range, such as 1 second, 0.2 points are deducted; the cumulative sum of the deduction items of the fire protection pipeline valve is calculated, and the difference between the basic health score of the fire protection pipeline valve and the cumulative sum is calculated, that is, the basic health score of the fire protection pipeline valve minus the cumulative sum, to obtain the health score of the fire protection pipeline valve;

[0084] Weighting coefficients are set based on the importance of each key component, and the health scores of key components in the same battery box are weighted and summed according to the weighting coefficients to obtain the comprehensive health score of the corresponding battery box. In this embodiment, the weighting coefficient of the photoelectric smoke sensor is set to 0.5, the weighting coefficient of the fire pump is set to 0.3, and the weighting coefficient of the fire pipeline valve is set to 0.2.

[0085] The risk intensity is set based on the results of the second-level risk assessment; if the battery compartment is risk-free, the corresponding risk intensity is set as the first risk intensity; if the battery compartment is of medium risk, the corresponding risk intensity is set as the second risk intensity; if the battery compartment is of high risk, the corresponding risk intensity is set as the third risk intensity; the first risk intensity is less than the second risk intensity, and the second risk intensity is less than the third risk intensity; the value range of the risk intensity is [0,1]; in this embodiment, the first risk intensity is 0.1, the second risk intensity is 0.6, and the third risk intensity is 1;

[0086] Based on the overall health score and risk intensity, the third-level risk score of each battery compartment is calculated. The third-level risk score is calculated as follows: calculate the difference between 1 and the overall health score of each battery compartment, that is, subtract the overall health score of each battery compartment from 1, and multiply the difference by the corresponding risk intensity to obtain the third-level risk score of the corresponding battery compartment.

[0087] A third-level risk assessment is performed based on the third-level risk score: if the average cell temperature of all batteries in any battery compartment is greater than a preset second temperature threshold, the corresponding battery compartment is determined to be high-risk; in this embodiment, the second temperature threshold is 63℃; when the third-level risk score is greater than a preset fifth risk score threshold, if the rate of change of cell temperature of any battery in the corresponding battery compartment is greater than a preset temperature change rate threshold, such as 2℃ / s, the corresponding battery compartment is determined to be high-risk; otherwise, the corresponding battery compartment is determined to be medium-risk; when the third-level risk score is less than or equal to the fifth risk score threshold, the corresponding battery compartment is determined to be risk-free; the fifth risk score threshold is determined based on the minimum acceptable health level and risk intensity of the system.

[0088] The specific generation method of the graded early warning signal is as follows: When any battery is classified as medium or high risk, a first-level early warning signal is generated for that battery; the first-level early warning signal includes the early warning level, battery number and the battery pack number it belongs to, and risk type; when the result of the second-level risk assessment of any battery pack is medium or high risk, a second-level early warning signal is generated for that battery pack; the second-level early warning signal includes the early warning level, battery pack number, and risk type; when the result of the third-level risk assessment of any battery pack is high risk, a third-level early warning signal is generated for that battery pack; the third-level early warning signal includes the early warning level, battery pack number, and risk type; when the results of the second-level risk assessment and the third-level risk assessment of the same battery pack are both high risk, only a third-level early warning signal is generated for that battery pack, and the corresponding third-level early warning signal indicates that the second-level risk also exists.

[0089] The decision control module generates protection instructions based on the graded early warning signals, including: optimized cooling instructions for level 1 protection, composite triggering instructions for level 2 protection, and emergency activation instructions for level 3 protection;

[0090] When the graded early warning signal is a Level 1 early warning signal, an optimized cooling command for Level 1 protection is generated. The optimized cooling command specifically includes the command type, target location, and cooling strategy. The command type is preventative cooling, and the target location is the target battery compartment corresponding to the battery compartment number in the Level 1 early warning signal. The cooling strategy includes flow regulation and temperature regulation. The flow regulation involves increasing the refrigerant circulation flow to the target battery compartment to 120%-150% of the rated value. The temperature regulation involves using refrigerant stored in the intelligent refrigerant tank to reduce the inlet temperature of the refrigerant flowing into the battery compartment to a preset temperature safety threshold. The temperature safety threshold is set according to specific monitoring requirements.

[0091] When the graded early warning signal is a level two early warning signal, a composite triggering command for level two protection is generated. The composite triggering command specifically includes command type, target location, and command content. The command type of the composite triggering command is composite triggering, and the target location is the target battery box corresponding to the battery box number in the level two early warning signal. The command content includes pre-start and mixed liquid injection. The pre-start optimization is: setting the fire pump corresponding to the target battery box to standby state, and simultaneously increasing the pressure of the fire pump from 0.3MPa to 0.6MPa. The mixed liquid injection is: injecting a mixture of stabilizing fluid and fire water into the target battery box under the pressure of the fire pump, and submerging the battery cells of the target battery box with the mixed liquid to achieve low-temperature cooling and fire suppression, while preventing thermal runaway.

[0092] When the tiered early warning signal is a Level 3 early warning signal, an emergency activation command for Level 3 protection is generated. The emergency activation command specifically includes the command type, target location, and command content. The command type of the emergency activation command is forced emergency activation, and the target location is the target battery compartment corresponding to the battery compartment number in the Level 3 early warning signal. The command content includes forced pressure increase and emergency access protection. The forced pressure increase is to increase and stabilize the outlet pressure of the fire pump corresponding to the target battery compartment to 0.6 MPa to prepare for high-pressure water injection. The emergency access protection is to open the main valve of the emergency fire water pipeline leading to the target battery compartment, or send a heating signal to the temperature sensing safety valve of the target battery compartment to heat the temperature sensing safety valve to the opening temperature.

[0093] The execution feedback module is used to execute the protection command and monitor its execution status; it optimizes the three-level risk assessment mechanism based on the execution status; the specific method for monitoring the execution status of the protection command is as follows: if the protection command is an optimized cooling command, the actual flow rate and inlet temperature of the refrigerant pipeline of the target battery box are monitored; if the actual flow rate reaches 120%-150% of the rated value and the inlet temperature drops to the temperature safety threshold within a preset first time after the optimized cooling command is issued, the execution is considered successful; otherwise, the execution is considered unsuccessful; if the protection command is a composite trigger command, the outlet pressure of the corresponding fire pump is monitored. Force and operational status feedback; if the outlet pressure reaches 0.6MPa within a preset second time after the composite trigger command is issued, and the fire pump's operational status feedback indicates it is running, then the execution is considered successful; otherwise, the execution is considered unsuccessful. If the protection command is an emergency start command, then the outlet pressure of the fire pump and the valve switch feedback of the emergency fire water pipeline are monitored. If the outlet pressure stabilizes at 0.6MPa within a preset third time after the emergency start command is issued, and the valve switch feedback indicates it is fully open, then the execution is considered successful; otherwise, the execution is considered unsuccessful. The first, second, and third times are all set according to specific detection requirements.

[0094] The specific method for optimizing the three-level risk assessment mechanism based on the execution status is as follows: When the execution status of any protection command is execution failure, locate the component that failed to respond correctly, and multiply the health score of the component in the third-level risk assessment by a preset penalty coefficient; the penalty coefficient is less than 1 and greater than 0; for example, if the outlet pressure of the fire pump is not stable at 0.6MPa, multiply the health score of the fire pump by 0.8 to obtain a new health score of the fire pump; if the execution status of the protection command corresponding to any first-level warning signal is execution success, and the status parameters of the target battery box recover to the risk-free level within a preset fourth time after the completion of the protection command, multiply the corresponding risk index threshold by a preset first adjustment ratio; the fourth time is based on the specific The detection requirements of the body are set; when the protection command is successfully executed and the state is restored, it indicates that the corresponding risk index threshold is too conservative, and the risk index threshold can be appropriately increased to reduce false alarms. Therefore, the first adjustment ratio is greater than 1, and is set according to the sensitivity requirements of the system; if the execution status of the protection command corresponding to any level of warning signal is successful, and the target battery box triggers a medium or high risk level again within a preset fourth time after the protection command is completed, then the corresponding risk index threshold is multiplied by the preset second adjustment ratio; when the protection command is successfully executed but the risk recurs, it indicates that the corresponding risk index threshold is too lenient, and the risk index threshold should be reduced to improve sensitivity. Therefore, the second adjustment ratio is less than 1, and is determined by statistically analyzing the probability of the protection command being successfully executed but the risk recurs.

[0095] Example 2

[0096] This embodiment is the second embodiment of this application; it is based on the same inventive concept as Embodiment 1, and refers to... Figure 2 This embodiment introduces a multi-parameter early warning monitoring method for thermal runaway of energy storage batteries, including the following steps:

[0097] The state parameters of the energy storage battery and the health parameters of the protection system are collected, and the state parameters and health parameters are preprocessed; the state parameters of the energy storage battery include cell temperature and rate of change, inlet pressure and rate of change, and battery pack parameters.

[0098] Based on the aforementioned state parameters and health parameters, a three-level risk assessment mechanism is established, and graded early warning signals are output based on the results of the risk assessment. The three-level risk assessment mechanism includes a first-level risk assessment based on cell temperature, a second-level risk assessment based on the charging box pressure, and a third-level risk assessment based on health parameters.

[0099] The protection command generated based on the graded early warning signal includes: when the graded early warning signal is a level 1 early warning signal, generating an optimized cooling command for level 1 protection; when the graded early warning signal is a level 2 early warning signal, generating a composite trigger command for level 2 protection; and when the graded early warning signal is a level 3 early warning signal, generating an emergency activation command for level 3 protection.

[0100] Execute the protection instructions and monitor their execution status; optimize the three-level risk assessment mechanism based on the execution status.

[0101] The specific functions of each of the above steps are described in the relevant content of the multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries described in Example 1, and will not be repeated here.

[0102] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. A multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries, characterized in that, It includes a data acquisition module, an early warning and analysis module, a decision control module, and an execution feedback module; The data acquisition module is used to collect the state parameters of the energy storage battery and the health parameters of the protection system, and to preprocess the state parameters and health parameters. Based on the aforementioned status parameters and health parameters, the early warning analysis module establishes a three-level risk assessment mechanism and outputs graded early warning signals based on the results of the risk assessment. The decision control module generates protection commands based on the graded early warning signals, including: optimized cooling commands for level 1 protection, composite trigger commands for level 2 protection, and forced start commands for level 3 protection. The execution feedback module is used to execute the protection instructions and monitor the execution status of the protection instructions; and optimize the three-level risk assessment mechanism based on the execution status.

2. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 1, characterized in that, The early warning analysis module is equipped with a three-level risk assessment mechanism; the three-level risk assessment mechanism includes: a first-level risk assessment based on cell temperature, a second-level risk assessment based on charging box pressure, and a third-level risk assessment based on health parameters; the steps of the first-level risk assessment specifically include: Set a historical time window; within the historical time window, collect the cell temperature of any battery at a first fixed frequency, and calculate the average temperature of all cells of the corresponding battery to obtain the corresponding temperature baseline value. Calculate the difference between the cell temperature of any battery at the current sampling time and the corresponding temperature baseline value to obtain the temperature deviation at the current sampling time; The first-level risk index of the corresponding battery is obtained by weighting and summing the rate of change of cell temperature, temperature deviation, and total charge / discharge rate.

3. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 2, characterized in that, The steps of the first-level risk assessment also include: A cell temperature prediction model is constructed, and the predicted cell temperature value within the prediction time period is calculated based on the cell temperature prediction model. If the first-level risk index is less than the preset first-level risk index threshold, and the predicted cell temperature within the prediction time period is less than the preset cell temperature threshold, then the corresponding battery will be determined to be risk-free. If the first level risk index is greater than or equal to the first level risk index threshold and less than the preset second level risk index threshold, or if the cell temperature prediction value is less than the preset first proportion but greater than or equal to the cell temperature threshold, then the corresponding battery will be classified as medium risk. If the first-level risk index is greater than or equal to the second-level risk index threshold, or if the predicted cell temperature value of the first proportion is greater than or equal to the cell temperature threshold, then the corresponding battery is determined to be high-risk.

4. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 3, characterized in that, The steps of the second-level risk assessment specifically include: Set a historical statistics window; obtain the air pressure of each battery compartment under normal operating conditions within the historical statistics window, and calculate the average value of the air pressure to obtain the corresponding air pressure baseline value; Calculate the difference between the current sampling time's chamber pressure and the corresponding baseline pressure value to obtain the pressure deviation; Based on the first-level risk index of all batteries in any battery compartment, calculate the risk correlation factor of the corresponding battery compartment, and then perform a weighted summation of the air pressure deviation, the rate of change of air pressure in the compartment, and the risk correlation factor to obtain the second-level risk index of the corresponding battery compartment. Based on the aforementioned air pressure in the insertion chamber and the second-level risk index, a second-level risk assessment is conducted.

5. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 4, characterized in that, The steps of the third-level risk assessment specifically include: Health scores are assigned to all critical components within each battery compartment to obtain the health score for each critical component. Weighting coefficients are set based on the importance of each key component, and the health scores of the key components of the same battery compartment are weighted and summed according to the weighting coefficients to obtain the comprehensive health score of the corresponding battery compartment. The risk intensity is set based on the results of the second-level risk assessment, and the third-level risk score of each battery compartment is calculated based on the risk intensity and the comprehensive health score. A third-level risk assessment is conducted based on the aforementioned third-level risk score.

6. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 5, characterized in that, The decision control module is configured with a protection command generation strategy; the protection command generation strategy specifically includes: When the graded early warning signal is a level one early warning signal, an optimized cooling command for level one protection is generated; the optimized cooling command specifically includes command type, target location and cooling strategy; The optimized cooling command is a preventative cooling command, and the target location is the target battery compartment corresponding to the battery compartment number in the first-level warning signal; the cooling strategy includes flow regulation and temperature regulation. The flow rate adjustment is to increase the refrigerant circulation flow rate to the target battery compartment to a preset ratio of the rated value; the temperature adjustment is to call the refrigerant stored in the intelligent refrigerant box to reduce the inlet temperature of the refrigerant flowing into the battery compartment to a preset temperature safety threshold.

7. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 6, characterized in that, The protection command generation strategy also includes: When the graded early warning signal is a level two early warning signal, a composite triggering command for level two protection is generated; the composite triggering command specifically includes command type, target location and command content; The instruction type of the composite triggering command is composite triggering, and the target location is the target battery box corresponding to the battery box number in the secondary warning signal; the instruction content includes pre-start and mixed liquid injection; The pre-start optimization is as follows: set the fire pump corresponding to the target battery box to standby mode, and increase the pressure of the fire pump from the current pressure value to the preset pressure value; The mixture injection is performed by injecting a mixture of stabilizing agent and fire water into the target battery compartment under the pressure of the fire pump, thereby submerging the battery cells in the target battery compartment.

8. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 7, characterized in that, The protection command generation strategy also includes: When the tiered early warning signal is a level three early warning signal, an emergency activation command for level three protection is generated; the emergency activation command specifically includes the command type, target location, and command content; The emergency start command is a forced emergency start command, and the target location is the target battery compartment corresponding to the battery compartment number in the level 3 warning signal. The instructions include forced pressure increase and emergency access protection; the forced pressure increase means: increasing the outlet pressure of the fire pump corresponding to the target battery box and stabilizing it at a preset pressure value; The emergency access guarantee is achieved by: opening the main valve of the emergency fire water pipeline leading to the target battery compartment, or sending a heating signal to the temperature sensing safety valve of the target battery compartment, and heating the temperature sensing safety valve to the opening temperature.

9. The multi-parameter early warning and monitoring system for thermal runaway of energy storage batteries as described in claim 8, characterized in that, The cell temperature prediction model includes a linear trend term, an operating condition correction term, and a historical residual term. Calculate the product of the rate of change of the cell temperature at the current sampling moment and the predicted time length, and use the sum of the product and the cell temperature at the current sampling moment as the linear trend term; The change in heat generation power is calculated based on the total charge / discharge rate, and the heat generation power correction component is calculated based on the change in heat generation power. Obtain the inlet temperature of the refrigerant at the current sampling time, and calculate the heat dissipation correction component based on the inlet temperature; The heat generation power correction component and the heat dissipation correction component are added together to obtain the operating condition correction term; Construct a historical deviation database; extract the operating condition feature vector at the current sampling time, and calculate the Euclidean distance between the operating condition feature vector and all operating condition feature vectors in the historical deviation database; Based on the Euclidean distance, similar feature vectors are selected, and the predicted residual values ​​corresponding to the similar feature vectors are weighted and summed to obtain the historical residual term at the current sampling time.

10. A multi-parameter early warning monitoring method for thermal runaway of energy storage batteries, which is implemented based on the multi-parameter early warning monitoring system for thermal runaway of energy storage batteries as described in any one of claims 1-9, characterized in that, Includes the following steps: Collect the state parameters of the energy storage battery and the health parameters of the protection system, and preprocess the state parameters and health parameters; Based on the aforementioned state parameters and health parameters, a three-level risk assessment mechanism is established, and graded early warning signals are output based on the results of the risk assessment. Based on the graded early warning signals, protection instructions are generated, including: optimized cooling instructions for level 1 protection, composite triggering instructions for level 2 protection, and emergency activation instructions for level 3 protection. Execute the protection instructions and monitor their execution status; optimize the three-level risk assessment mechanism based on the execution status.

Images