Unit type battery compartment integrating multiple monitoring parameters and predictive safety management and control system

Abstract

The invention relates to a unit type battery compartment integrated with multiple monitoring parameters and a predictive safety management and control system. Comprising a compartment body, a target battery pack arranged in the compartment body, a multi-parameter monitoring subsystem, a predictive safety early warning model subsystem and an integrated abnormity and fire safety management and control subsystem.

Description

Technical Field

[0001] This application belongs to the field of battery technology, and specifically relates to a modular battery compartment that integrates multiple monitoring parameters and a predictive safety management system. Background Technology

[0002] The safe operation of battery packs is crucial for the safe and stable operation of battery energy storage systems. How to monitor batteries during operation and take appropriate measures when abnormal conditions are detected is a problem that needs to be solved. Summary of the Invention

[0003] This application provides a modular battery compartment integrating multiple monitoring parameters and a predictive safety management system. The modular battery compartment includes: The system includes: the storage unit, the target battery pack located inside the storage unit, a multi-parameter monitoring subsystem, a predictive safety early warning model subsystem, and an integrated anomaly and fire safety control subsystem. A multi-parameter monitoring subsystem is used to collect sensor data during the operation of the target battery pack; The predictive safety early warning model subsystem is used to determine the parameters for judgment based on sensor data. The parameters for judgment include: the predicted voltage of the target battery pack, the predicted temperature of the target battery pack, the predicted SOC of the target battery pack, and the predicted SOH of the target battery pack. Among them, determining the parameters for judgment based on sensor data includes: generating a deviation term for each individual cell in the target battery pack according to an inverse Gaussian distribution, and determining the predicted voltage of the target battery pack based on the monitored voltage of each individual cell in the sensor data and the deviation term of each individual cell. The integrated anomaly and fire safety management subsystem is used to determine whether each of the multiple risk conditions is met based on the parameters used for judgment, and when it is determined that the risk condition is met, to provide early warning and protective measures corresponding to the risk condition.

[0004] In one possible implementation, determining the predicted voltage of the target battery pack based on the monitored voltage of each individual cell in the sensor data and the deviation term of each individual cell includes: Based on the monitored voltage of each individual cell and the deviation term of each individual cell, the candidate voltage of the target battery pack is determined; The predicted voltage of the target battery pack is determined based on the candidate voltages of the target battery pack.

[0005] In one possible implementation, determining the predicted voltage of the target battery pack based on the candidate voltages of the target battery pack includes: The temperature compensation factor is determined based on the predicted temperature of the target battery pack and the preset reference temperature. The predicted voltage of the target battery pack is determined based on the candidate voltage and temperature compensation factor.

[0006] In one possible implementation, the parameters used for the judgment, determined based on sensor data, include: Using a strong tracking Kalman filter, based on the coupled electro-thermal model, the predicted SOC and predicted temperature of each individual cell in the target battery pack are predicted. Based on the predicted SOC of each individual cell, the predicted SOC of the target battery pack is determined, and based on the predicted temperature of each individual cell, the predicted temperature of the target battery pack is determined.

[0007] In one possible implementation, the parameters used for judgment also include: the thermal runaway probability corresponding to the target battery pack; and the parameters used for judgment determined based on sensor data include: The probability of electrical signal anomalies is determined based on the measured voltage and predicted voltage of the target battery pack in the sensor data. The probability of thermal signal anomalies is determined based on the highest predicted temperature among all individual cells in the target battery pack and the temperature change rate coefficient of the target battery pack. The probability of chemical signal anomalies is determined based on the concentrations of various gases in the sensor data. These gases include electrolyte vapor, H2, and CO. The abnormal probabilities of the electrical signal, thermal signal, and chemical signal are fused to obtain the thermal runaway probability of the target battery pack.

[0008] In one possible implementation, multiple risk conditions include: a first consultation-level warning condition, a second consultation-level warning condition, and a third consultation-level warning condition; the first consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than a first thermal runaway probability threshold and the predicted SOH corresponding to the target battery pack is less than the SOH threshold; the second consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than the first thermal runaway probability threshold and the internal resistance of at least one individual cell in the target battery pack continues to increase abnormally; the third consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than the first thermal runaway probability threshold and the amount of electrolyte vapor in the parameters used for judgment is at the trace level; the protective measures corresponding to the target consultation-level warning condition include: limiting the charging and discharging power of the target battery pack, wherein the target consultation-level warning condition is any one of the first consultation-level warning condition, the second consultation-level warning condition, and the third consultation-level warning condition.

[0009] In one possible implementation, multiple risk conditions include: a first severe fault warning condition, a second severe fault warning condition, a third severe fault warning condition, and a fourth severe fault warning condition; the first severe fault warning condition is a sudden voltage drop in the target battery pack; the second severe fault warning condition is overcharging or over-discharging at least one individual cell in the target battery pack; the third severe fault warning condition is that the concentration of H2 or CO in the parameters used for judgment exceeds a first concentration threshold; and the fourth severe fault warning condition is that the probability of thermal runaway is greater than a second thermal runaway probability threshold, wherein the second thermal runaway probability threshold is greater than the first thermal runaway probability threshold. Protective measures corresponding to the target severe fault warning conditions include: triggering a high-speed solid-state switch for electrical isolation, starting a high-power liquid / air cooling system, and enhancing ventilation. The target severe fault warning condition is any one of the first, second, third, or fourth severe fault warning conditions; triggering the high-speed solid-state switch is used for electrical isolation; starting the high-power liquid / air cooling system is used for forced heat dissipation; and enhancing ventilation is used to reduce the concentration of combustible gases.

[0010] In one possible implementation, multiple risk conditions include: a first fire pre-response condition, a second fire pre-response condition, and a third fire pre-response condition; the first fire pre-response condition is that the thermal runaway probability is greater than a third thermal runaway probability threshold; the second fire pre-response condition is that the temperature of the target battery pack rises sharply; and the third fire pre-response condition is that the concentration of H2 or CO in the parameters used for judgment exceeds a second concentration threshold; wherein, the third thermal runaway probability threshold is greater than the second thermal runaway probability threshold, and the protective measures corresponding to the target fire pre-response condition include: releasing the main extinguishing agent and suppressing the measured temperature of the target battery pack below the thermal runaway critical point, wherein the target fire pre-response condition is any one of the first, second, and third fire pre-response conditions, and the main extinguishing agent is released for forced cooling and chemical inerting.

[0011] In one possible implementation, multiple risk conditions include: a first fire confirmation condition and a second fire confirmation condition; the first fire confirmation condition is that the thermal runaway probability is greater than a fourth thermal runaway probability threshold, wherein the fourth thermal runaway probability threshold is greater than a third thermal runaway probability threshold; the second fire confirmation condition is that a smoke detector alarm or a flame detector alarm; the protective measures corresponding to the target fire confirmation condition include: confirming the release of extinguishing agent and activating a backup flame suppression procedure; sending a fire alarm, wherein the target fire confirmation condition is either the first fire confirmation condition or the second fire confirmation condition; any one of the first fire confirmation conditions.

[0012] In one possible implementation, the battery compartment has fireproof and thermal insulation capabilities. The outer shell of the compartment is made of high-strength fire-resistant material, and the inner wall of the compartment is coated with a heat-activated intumescent fire-retardant coating. The heat-activated intumescent fire-retardant coating is used to form a heat insulation layer at high temperatures. The interior of the compartment is divided into partitions by heat-insulating partitions to slow down heat spread. The unit battery compartment integrating multiple monitoring parameters and predictive safety management systems also includes: a non-contact infrared thermal imaging sensor for capturing the two-dimensional temperature distribution of the target battery pack. The capture of the two-dimensional temperature distribution of the target battery pack is used to detect local abnormal hot spots within the target battery pack.

[0013] Beneficial effects: The modular battery compartment integrating multiple monitoring parameters and a predictive safety management system provided in this application embodiment can determine whether each of multiple risk conditions is met based on the parameters used for judgment. When a risk condition is determined to be met, it can provide a warning and corresponding protective measures. Furthermore, when a corresponding risk condition is determined to be met, it can identify the abnormal situation indicated by that risk condition and implement corresponding protective measures to avoid at least one of the following: battery pack failure, battery pack thermal runaway, or reduction of damage caused by abnormal situations. Thus, the battery is monitored during operation, and appropriate measures are taken when corresponding abnormal situations are detected.

[0014] The unit battery compartment integrating multiple monitoring parameters and predictive safety management system provided in this application generates a deviation term for each individual cell in the target battery pack based on an inverse Gaussian distribution. It also determines the predicted voltage of the target battery pack based on the monitored voltage of each individual cell from sensor data and the deviation term of each individual cell. This relatively accurate prediction of the target battery pack's voltage allows for more precise determination of whether abnormal situations have occurred. The predicted voltage can be used for voltage drop detection, early consistency fault warning, polarization anomaly detection, and other abnormal situations.

[0015] The modular battery compartment provided in this application integrates multiple monitoring parameters and a predictive safety management system, constructing a modular battery compartment system that integrates multi-parameter monitoring, predictive early warning, and active management. It constructs a highly integrated and modular safety architecture: monitoring, prediction, control, and execution mechanisms are fully integrated into an independent, physically isolated modular battery compartment, achieving physical isolation and functional autonomy against faults. This ensures that a fault in a single compartment will not cascade to the entire energy storage power station, improving the overall resilience and safety of large-scale BESS (Battery Emergency Storage System). Through modular and integrated design, physical isolation and functional autonomy of a single faulty unit are achieved, preventing cascading reactions and enhancing the overall safety of large-scale energy storage power stations. Accurate prediction is achieved by combining a physical-data fusion algorithm based on inverse Gaussian voltage correction and strong tracking Kalman filtering, and a tiered response mechanism from consultation and early warning to fire prevention and control is established.

[0016] The modular battery compartment of this application consists of a modular compartment body with fireproof and thermal insulation capabilities. The outer shell of the compartment body can be made of high-strength fire-resistant material, and the inner wall of the compartment body can be coated with a heat-activated intumescent fire-retardant coating, which is used to form a heat insulation layer at high temperatures. The interior of the compartment body is divided into zones by heat-insulating partitions to slow down heat spread.

[0017] The highly integrated design of the unit battery compartment, which integrates multiple monitoring parameters and a predictive safety management system, provided in this application embodiment ensures physical isolation of faults and immediacy of safety response. Its modular concept can effectively prevent a single faulty unit from affecting the entire energy storage power station.

[0018] The modular battery compartment integrating multiple monitoring parameters and a predictive safety management system provided in this application embodiment can improve prediction accuracy and robustness: the physical model-data fusion prediction method overcomes the shortcomings of insufficient generalization ability of pure data models and the inaccuracy of pure physical models. The constraints of physical laws make the model prediction more reliable, effectively reducing false alarm and false negative rates, and maintaining high accuracy even when facing new operating conditions. It can make accurate and reliable probabilistic predictions of the future safety status of the battery, effectively reducing false alarm and false negative rates. The modular battery compartment integrating multiple monitoring parameters and a predictive safety management system provided in this application embodiment can capture very early signs of thermal runaway through a multi-dimensional sensor network, achieving very early fault detection. By sensitively monitoring the chemical precursor gases of thermal runaway (such as DMC and H2), this invention can capture the chemical signals of the initial fault minutes or even hours before significant electrical or thermal anomalies occur in the battery, significantly advancing the warning window and gaining valuable time for intervention. It achieves proactive and preventative safety management, shifting the safety management paradigm from post-event response to intervention when risks occur. Based on prediction results, it can automatically execute measures such as electrical isolation, forced cooling, and even pre-release of fire extinguishing agents before thermal runaway occurs, effectively preventing or delaying the occurrence and spread of thermal runaway, shifting the safety defense line from post-event fire suppression to pre-event explosion and fire prevention. It can capture chemical precursor signals occurring in the very early stages of thermal runaway that traditional BMS cannot detect, significantly advancing the warning window. It can automatically execute graded intervention measures, including electrical isolation and forced cooling, based on prediction results before thermal runaway occurs. It can activate the fire suppression system before an open flame appears, proactively suppressing thermal runaway from a chemical and physical perspective. Attached Figure Description To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a schematic diagram of a second-order equivalent circuit model. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. It should be noted that, unless otherwise specified, the implementation methods and features in the implementation methods in this disclosure can be combined, separated, interchanged, and / or rearranged. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] In the accompanying drawings, the dimensions and relative dimensions of components may be exaggerated for clarity and / or descriptive purposes. When exemplary embodiments can be implemented differently, a specific process sequence may be performed in a different order than that described. For example, two consecutively described processes may be performed substantially simultaneously or in the reverse order of their description. Furthermore, the same reference numerals denote the same components.

[0022] When a component is referred to as being "on" or "above" another component, "connected to," or "joined to" another component, the component may be directly on, directly connected to, or directly joined to the other component, or there may be intermediate components. However, when a component is referred to as being "directly on" another component, "directly connected to," or "directly joined to" another component, there are no intermediate components. Therefore, the term "connection" can refer to a physical connection, an electrical connection, etc., and may or may not have intermediate components.

[0023] For descriptive purposes, this disclosure may use spatial relative terms such as “top,” “bottom,” “below,” “under,” “under,” “below,” “above,” “above,” “higher,” etc., which are relative to components, to describe the relationship between one component and another (other) component as shown in the accompanying drawings.

[0024] The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “a” and “the” are intended to include the plural forms as well. Furthermore, when the terms “comprising” and / or “including” and variations thereof are used in this specification, it indicates the presence of the stated features, integrals, steps, operations, parts, components, and / or groups thereof, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, parts, components, and / or groups thereof. It should also be noted that, as used herein, the terms “substantially,” “about,” and other similar terms are used as approximate terms rather than as terms of degree, thus explaining the inherent biases in measurements, calculated values, and / or provided values ​​that would be recognized by one of ordinary skill in the art.

[0025] This application provides a unit battery compartment integrating multiple monitoring parameters and a predictive safety management system, including: a compartment body, a target battery pack installed inside the compartment body, a multi-parameter monitoring subsystem, a predictive safety early warning model subsystem, and an integrated anomaly and fire safety management subsystem; A multi-parameter monitoring subsystem is used to collect sensor data during the operation of the target battery pack; The predictive safety early warning model subsystem is used to determine the parameters for judgment based on sensor data. The parameters for judgment include: the predicted voltage of the target battery pack, the predicted temperature of the target battery pack, the predicted state of charge (SOC) of the target battery pack, and the predicted state of health (SOH) of the target battery pack. Among them, the parameters for judgment determined based on sensor data include: generating the deviation term of each individual cell in the target battery pack according to the inverse Gaussian distribution, and determining the predicted voltage of the target battery pack based on the monitored voltage of each individual cell in the sensor data and the deviation term of each individual cell. The integrated anomaly and fire safety management subsystem is used to determine whether each of the multiple risk conditions is met based on the parameters used for judgment, and when it is determined that the risk condition is met, to provide early warning and protective measures corresponding to the risk condition.

[0026] The target battery pack includes multiple individual cells, which can be connected in series.

[0027] The interior of the compartment is used to house the target battery pack, and the predictive safety early warning model subsystem runs on the central processing unit.

[0028] In one possible implementation, the battery compartment has fireproof and thermal insulation capabilities. The outer shell of the compartment is made of high-strength fire-resistant material, and the inner wall of the compartment is coated with a heat-activated intumescent fire-retardant coating. The heat-activated intumescent fire-retardant coating is used to form a heat insulation layer at high temperatures. The interior of the compartment is divided into partitions by heat-insulating partitions to slow down heat spread. The unit battery compartment integrating multiple monitoring parameters and predictive safety management systems also includes: a non-contact infrared thermal imaging sensor for capturing the two-dimensional temperature distribution of the target battery pack. The capture of the two-dimensional temperature distribution of the target battery pack is used to detect local abnormal hot spots within the target battery pack.

[0029] The multi-parameter monitoring subsystem includes various sensors located within the storage chamber.

[0030] Sensor data includes data collected by each sensor. The data collected by the sensors can be used to determine whether corresponding risk conditions are met and / or to generate parameter items in the parameters used for that determination.

[0031] It should be noted that the data collected by the sensor, the data monitored by the sensor, and the measured data are equivalent.

[0032] As an example, the various sensors inside the chamber include: electrochemical and electrical sensors, thermal sensors, gas composition analysis sensor arrays, and physical sensors.

[0033] The thermal sensors may include: contact temperature sensors positioned at key locations. They may also include: non-contact infrared thermal imaging sensors used to capture the two-dimensional temperature distribution across the entire battery array, enabling earlier detection of localized hot spots.

[0034] Physical sensors may include pressure sensors. They may also include smoke detectors or flame detectors.

[0035] The sensor data collected during the operation of the target battery pack includes: data collected by electrochemical and electrical sensors, such as the voltage, current, and impedance of each individual cell; data collected by thermal sensors; and data collected by the gas composition analysis sensor array.

[0036] Electrochemical and electrical sensors are used to monitor the voltage, current, impedance, and other properties of individual cells in a target battery pack in real time.

[0037] Among them, the voltage, current, impedance and other data of each cell in the target battery pack monitored by electrochemical and electrical sensors can be used to determine the predicted voltage of the target battery pack, the predicted SOC of the target battery pack, and the predicted SOH of the target battery pack.

[0038] A gas composition analysis sensor array is used to capture gases released at different stages of battery thermal runaway. The system analyzes the concentrations of gases released at different stages of battery thermal runaway. A gas composition analysis sensor array can sensitively detect electrolyte solvent vapors (such as DMC) generated in the very early stages of thermal runaway, as well as hydrogen (H2) and carbon monoxide (CO) released due to SEI membrane decomposition in the critical early stages. Through sequential analysis of these precursor gases, the system can accurately determine the stage of the fault, advancing the warning window by minutes or even hours.

[0039] The predictive safety early warning model subsystem is used to determine the parameters for judgment based on sensor data. The parameters for judgment include: the predicted voltage of the target battery pack, the predicted temperature of the target battery pack, the predicted SOC of the target battery pack, and the predicted SOH of the target battery pack.

[0040] If a monitored data point from a sensor can be used to determine whether the corresponding risk conditions are met, then the monitored data point is directly used as a parameter in the parameters used for the determination.

[0041] If a monitored data point in a sensor dataset is used to generate a parameter item in a set of parameters for judgment, then the monitored data is used to generate the parameter item in the set of parameters for judgment.

[0042] It should be noted that a battery pack can also be called a battery module. The voltage of the target battery pack can refer to the terminal voltage of the target battery pack.

[0043] The unit battery compartment integrating multiple monitoring parameters and a predictive safety management system provided in this application takes into account the inconsistencies between battery cells due to differences in manufacturing processes and aging paths when obtaining the predicted voltage of the target battery pack. Therefore, in this application embodiment, the voltage of the battery pack monitored by the sensors is not directly used as the predicted voltage of the target battery pack. Instead, a voltage distribution correction based on an inverse Gaussian distribution is performed to obtain the predicted voltage of the target battery pack.

[0044] The unit battery compartment integrating multiple monitoring parameters and predictive safety management system provided in this application embodiment also considers that: if it is assumed that the voltage deviation of a single cell follows a Gaussian distribution, it is impossible to accurately describe the asymmetric, long-tailed distribution characteristics caused by a few short-board cells in the later stage of battery pack aging.

[0045] Therefore, the unit battery compartment integrating multiple monitoring parameters and predictive safety management system provided in this application embodiment uses an inverse Gaussian distribution to statistically correct the distribution deviation of the individual cell voltage. Assuming that the voltage deviation ΔUi of the individual cell follows an inverse Gaussian distribution with position parameter μ=Umean and shape parameter λ=μ³ / Var(Ui), the probability density function of the inverse Gaussian distribution is: It should be noted that the right-skewed characteristics of the inverse Gaussian distribution are highly consistent with the asymmetry of performance degradation in battery packs.

[0046] It should be noted that the voltage deviation of a single cell and the deviation term of a single cell are equivalent. Both the voltage deviation of a single cell and the deviation term of a single cell represent the difference between the monitored voltage of a single cell and the actual voltage of the single cell.

[0047] Based on the inverse Gaussian distribution, generating the deviation term for each individual cell in the target battery pack can include: for individual cell A in the target battery pack, generating random numbers according to the inverse Gaussian distribution, and using the generated random numbers as the deviation term for individual cell A. In other words, the deviation term for individual cell A is sampled from the inverse Gaussian distribution. Individual cell A can be any individual cell in the target battery pack.

[0048] As an example, the monitored voltages of all individual cells in the target battery pack are summed to obtain the voltage sum. The deviation terms of all individual cells in the target battery pack are summed to obtain the deviation term sum. The voltage sum is then added to the deviation term sum to obtain the predicted voltage of the target battery pack.

[0049] In other words, the predicted voltage of the target battery pack is the sum of the voltages and the sum of the deviation terms.

[0050] In one possible implementation, determining the predicted voltage of the target battery pack based on the monitored voltage of each individual cell and the deviation term of each individual cell in the sensor data includes: determining the candidate voltage of the target battery pack based on the monitored voltage of each individual cell and the deviation term of each individual cell; and determining the predicted voltage of the target battery pack based on the candidate voltage of the target battery pack.

[0051] As an example, the monitored voltages of all individual cells in the target battery pack are summed to obtain the sum of voltages. The deviation terms of all individual cells in the target battery pack are summed to obtain the sum of deviation terms. The sum of voltages and the sum of deviation terms are then added together. The predicted voltage of the target battery pack is the sum of the sum of voltages and the sum of deviation terms, which is determined as the candidate voltage of the target battery pack.

[0052] As another example, the following formula can be used to determine the candidate voltage of the target battery pack based on the monitored voltage of each individual cell in the sensor data, the deviation term of each individual cell, and the distribution correction coefficient: Among them, U F Let κ represent the candidate voltage of the target battery pack, and let κ denote the distribution correction coefficient. Let N represent the deviation term of the i-th cell in the target battery pack generated by the inverse Gaussian distribution, and let N represent the number of cells in the target battery pack.

[0053] As an example of the distribution correction factor, the distribution correction factor is preset.

[0054] As another example of the distribution correction coefficient, κ=σpack / σcell, where σpack is the standard deviation of the measured voltage of the battery pack in the energy storage system in which the target battery pack is installed, and σcell is the standard deviation of the measured voltage of the individual cells in the target battery pack.

[0055] It should be noted that Monte Carlo simulations show that the model improves the characterization ability of the battery pack voltage distribution in the later stages of aging (capacity decay > 20%) by 41%, confirming the effectiveness of the statistical correction mechanism and thus providing a more realistic physical prior for the data-driven model. In one possible implementation, the candidate voltage of the target battery pack is directly used as the predicted voltage of the target battery pack.

[0056] In another possible implementation, when determining the predicted voltage of the target battery pack, the relationship between the temperature characteristics of the battery pack and the variation in the polarization resistance of the edge cells is also considered. The variation in the polarization resistance of the edge cells affects the accuracy of the predicted battery pack voltage. For example, finite element simulations show that during natural convection, the maximum internal temperature difference of the battery pack reaches 8.3℃, causing a 19% increase in the polarization resistance of the edge cells, thus affecting the accuracy of the predicted battery pack voltage. Therefore, a temperature compensation factor β = 1 + 0.05(T - Tref) is used to dynamically correct the predicted battery pack terminal voltage. Specifically, determining the predicted voltage of the target battery pack based on the candidate voltage includes: determining the temperature compensation factor based on the predicted temperature of the target battery pack and a pre-set reference temperature; and determining the predicted voltage of the target battery pack based on the candidate voltage and the temperature compensation factor.

[0057] As an example, the following formula can be used to determine the temperature compensation factor based on the predicted temperature and reference temperature corresponding to the target battery pack: β =1+0.05(T-Tref) Where β is the temperature compensation factor, T is the predicted temperature corresponding to the target battery pack, and Tref is the reference temperature, for example, Tref is 298.15K (25℃).

[0058] As an example, the following formula can be used to determine the predicted voltage of the target battery pack based on the candidate voltage and temperature compensation factor: in, The predicted voltage of the target battery pack. The candidate voltage for the target battery pack.

[0059] It should be noted that the predicted voltage of the target battery pack can be called the predicted terminal voltage of the target battery pack after multiple corrections (U). pre ) .

[0060] In this embodiment of the application, parameters for judgment are determined based on sensor data. The parameters for judgment include: the predicted voltage of the target battery pack, the predicted temperature of the target battery pack, the predicted SOC of the target battery pack, and the predicted SOH of the target battery pack.

[0061] The predicted SOC for the target battery pack can be the predicted SOC of the single cell with the lowest predicted SOC in the target battery pack. The predicted SOH for the target battery pack can be the predicted SOH of the single cell with the lowest predicted SOH in the target battery pack.

[0062] As an example, any existing algorithm for estimating SOC can be used to estimate the SOC of each individual cell in the target battery pack. For each individual cell in the target battery pack, the estimated SOC of the individual cell is used as the predicted SOC of the individual cell.

[0063] As an example, any existing algorithm for estimating SOH can be used to estimate the SOH of each individual cell in the target battery pack. For each individual cell in the target battery pack, the estimated SOC of the individual cell is used as the predicted SOC of the individual cell.

[0064] As an example, any algorithm used to estimate the temperature of a battery pack can be employed to estimate the temperature of the target battery pack, and the estimated temperature can be used as the predicted temperature corresponding to the target battery pack. For instance, the predicted temperature of the target battery pack can be determined based on the collected temperatures at multiple locations inside the target battery pack, a pre-defined relationship between the temperatures at multiple locations inside the battery pack and the predicted temperature corresponding to the battery pack.

[0065] As another example, the temperature of each individual cell in each target battery pack is estimated using an arbitrary method for estimating the temperature of individual cells, and the lowest temperature among all estimated individual cell temperatures is taken as the predicted temperature corresponding to the target battery pack.

[0066] In one possible implementation, the parameters for judgment are determined based on sensor data, including: using a strong tracking Kalman filter, predicting the predicted state of charge (SOC) and the predicted temperature of each individual cell in the target battery pack according to the coupled electro-thermal model; determining the predicted SOC of the target battery pack based on the predicted SOC of each individual cell; and determining the predicted temperature of the target battery pack based on the predicted temperature of each individual cell.

[0067] The coupled electro-thermal model can include: an improved second-order equivalent circuit model, the Arrhenius equation, and a thermal model.

[0068] The thermal model can be used to predict heat generation and temperature changes inside the battery in real time and dynamically. The thermal model is bidirectionally coupled with the improved second-order equivalent circuit model. The parameters such as current and internal resistance (R0, R1, R2) calculated by the improved second-order equivalent circuit model are the core inputs for calculating Joule heat and entropy heat in the thermal model. The temperature T predicted by the thermal model will in turn affect the parameters in the electrochemical model, thus forming an "electro-thermal coupling".

[0069] The coupled electro-thermal model can serve as the state-space model for a strong tracking Kalman filter (STKF).

[0070] For each individual cell, a strong tracking Kalman filter is used to perform optimal estimation based on the coupled electro-thermal model and the measured values ​​of the parameters used to estimate the SOC collected by the sensor, and the predicted SOC of the individual cell is output.

[0071] For each individual cell, a strong tracking Kalman filter is used to perform an optimal estimate based on the coupled electro-thermal model and the measured values ​​of the parameters used to estimate the temperature of the individual cell collected by the sensor, and the predicted temperature of the individual cell is output.

[0072] As an example of determining the predicted SOC of the target battery pack based on the predicted SOC of each individual cell, the predicted SOC of the individual cell with the smallest predicted SOC in the target battery pack is determined as the predicted SOC of the target battery pack.

[0073] As an example of determining the predicted temperature of the target battery pack based on the predicted temperature of each individual cell, the predicted temperature of the individual cell with the lowest predicted temperature in the target battery pack is determined as the predicted temperature of the target battery pack.

[0074] At the individual cell level, an improved second-order equivalent circuit model (ECM) is employed to accurately characterize its electrochemical dynamics. The state equation of this model consists of the open circuit voltage (OCV), the ohmic internal resistance (R0), and two RC networks (R1 / C1) describing polarization effects. The nonlinear relationship between the open circuit voltage and the state of charge (SOC) is fitted with high precision using a fifth-order polynomial, the mathematical expression of which is: Where z belongs to [0,1] to represent SOC, the polynomial coefficients an are calibrated using the least squares method. Experimental data show that the OCV prediction error at a charge-discharge rate of 0.1C is less than 8mV, which is better than the traditional piecewise linear model.

[0075] refer to Figure 1 It shows a schematic diagram of the second-order equivalent circuit model.

[0076] Where Ra and Rb are both ohmic internal resistances, and R2 is a polarization resistor.

[0077] Considering the significant impact of temperature on the rate of internal chemical reactions in the battery, the coupled electro-thermal model includes the Arrhenius equation to correct for the temperature dependence of polarization resistance. The Arrhenius equation is as follows: Where Ea is the activation energy (kJ / mol), Rg = 8.314 J / (mol·K) is the ideal gas constant, and Tref is 298.15 K (25 °C). The correction term can well describe the diffusion activation process of lithium ions in the electrode material, effectively improving the model's adaptability over a wide temperature range. The thermal model is constructed using the lumped parameter method, comprehensively considering Joule heating, entropic heating, and convective heat dissipation, thus achieving coupling between the temperature field and the electrochemical state. The thermal model is as follows: Where Cth is the battery heat capacity (in J / K), h is the convective heat transfer coefficient (in W / (m²·K)), and A is the effective heat dissipation area. The temperature prediction error of the thermal model under 5C discharge conditions is ±1.8℃, which is 57% more accurate than the model that ignores the entropy heat term. To achieve high-precision online estimation of the coupled states of multiple physics fields, a Strong Tracking Kalman Filter (STKF) algorithm was designed and adopted. The STKF algorithm simultaneously estimates the electrochemical state (SOC) and the thermal state (T). The STKF algorithm dynamically adjusts the prediction covariance matrix through a fading factor, thereby forcing the filter to quickly track the true state of the system when abrupt changes occur in the innovation sequence (i.e., the difference between model predictions and actual measurements). The core update equation of the STKF algorithm is as follows: Φ is the state transition matrix, H is the observation matrix, and Q and R represent the covariance matrices of process noise and observation noise, respectively. P_(k|k-1) is the "prior estimate covariance matrix", which quantifies the uncertainty of the state prediction value. That is, P_(k|k-1) indicates the possible range of error in the prediction of SOC and T based on the model. tr(N_k) is the trace of matrix N_k, which is the sum of its diagonal elements. N_k represents the difference between the "actual mismatch" and the "theoretically expected mismatch". tr(M_k) is the trace of matrix M_k, representing the theoretically expected uncertainty that the state uncertainty inherited from the previous time step will cause to the measurement prediction at the current time step. The fading factor ρk enables the filter to quickly track state changes when battery parameters suddenly change, such as an internal short circuit. Experimental data shows that under conditions of a step change in current (±5C), the SOC estimation error can be controlled within ±1.5%, which is a 62% improvement compared to the traditional extended Kalman filter (EKF) algorithm. It should be noted that the coupled electro-thermal model can also include a three-dimensional heat conduction equation. By solving the three-dimensional heat conduction equation, the temperature value at any location within the entire battery pack can be obtained, i.e., the complete temperature field distribution. The three-dimensional heat conduction equation complements the overall system, and the obtained temperature field provides STKF with more accurate thermal state information, forming a closed loop. The three-dimensional heat conduction equation is as follows: Where qgen is the volumetric heat generation rate and k is the anisotropic thermal conductivity.

[0078] In one possible implementation, the parameters used for judgment also include: the thermal runaway probability corresponding to the target battery pack; and the parameters determined based on sensor data include: determining the electrical signal anomaly probability based on the measured voltage and predicted voltage of the target battery pack in the sensor data; determining the thermal signal anomaly probability based on the highest predicted temperature among all individual cells in the target battery pack and the temperature change rate coefficient of the target battery pack; determining the chemical signal anomaly probability based on the concentrations of various gases in the sensor data, wherein the various gases include: electrolyte vapor, H2, and CO; and fusing the electrical signal anomaly probability, the thermal signal anomaly probability, and the chemical signal anomaly probability to obtain the thermal runaway probability corresponding to the target battery pack.

[0079] The measured voltage of the target battery pack is the sum of the measured voltages of all individual cells in the target battery pack.

[0080] As an example, the following formula can be used to calculate the probability of an electrical signal anomaly based on the measured voltage of the target battery pack and the predicted voltage of the target battery pack: in, This represents the probability of an abnormal electrical signal. The measured voltage of the target battery pack is the voltage that is monitored. Predicted voltage of the target battery pack The noise standard deviation of the voltage measurement. Sensitivity adjustment coefficient.

[0081] As an example, the following formula can be used to determine the probability of thermal signal anomalies based on the highest predicted temperature among all individual cells in the target battery pack and the temperature change rate coefficient of the target battery pack: in, This represents the probability of thermal signal anomalies. This represents the highest predicted temperature among all individual cells in the target battery pack. The temperature change rate coefficient of the target battery pack. Temperature threshold dT / dt is the temperature rise rate threshold, k is the temperature change rate of the target battery pack, and k is the temperature rise rate threshold. T k d These are pre-set coefficients.

[0082] As an example, the following formula can be used to determine the probability of chemical signal anomalies based on the concentrations of various gases in the sensor data: For each gas (H2, CO, DMC, etc.): Ci(t) is the concentration of a certain gas in the sensor data, Ci(t) is the concentration threshold of that gas, and kgas is the preset coefficient of that gas.

[0083] As an example, the following formula is used to fuse the probabilities of electrical signal anomalies, thermal signal anomalies, and chemical signal anomalies to determine the thermal runaway probability of the target battery pack: in, w represents the thermal runaway probability corresponding to the target battery pack. e w t and w c This is a weighting coefficient that can be adaptively adjusted based on battery type and operating status.

[0084] This application provides a unit-type battery compartment that integrates multiple monitoring parameters and a predictive safety management system, and provides a hierarchical active management system based on multi-source information fusion and predictive output.

[0085] This system deeply integrates battery safety management with fire response through predictive, tiered intervention, achieving a proactive safety paradigm shift from post-event handling to pre-event prevention and in-event mitigation. The embodiments of this application provide a modular battery compartment integrating multiple monitoring parameters and a predictive safety management system. Through closed-loop control of state perception, risk assessment, and execution response, it achieves adaptive and collaborative intervention against battery failure and thermal runaway risks.

[0086] The integrated anomaly and fire safety management subsystem can determine whether each of the multiple risk conditions is met based on the parameters used for judgment, and when it is determined that the risk condition is met, it can provide early warning and protective measures corresponding to the risk condition.

[0087] The warnings corresponding to the fulfillment of the relevant risk conditions include: displaying text indicating the abnormal situation indicating the fulfillment of the relevant risk conditions, and playing voice messages indicating the abnormal situation indicating the fulfillment of the relevant risk conditions.

[0088] It should be noted that for risk condition A, the parameters used for judgment include parameters for determining whether risk A is met. Specifically, when risk condition A is determined to be met, a warning and corresponding protective measures are implemented. Risk condition A can be any risk condition. The protective measures corresponding to the met risk condition are used to prevent accidents caused by the abnormal situation indicated by the met risk condition or to reduce the damage caused by the abnormal situation indicated by the met risk condition.

[0089] As an example, the following are risk conditions: the predicted SOH of the target battery pack is less than the SOH threshold, the voltage of the target battery pack suddenly drops, at least one cell in the target battery pack is overcharged or over-discharged, and the temperature of the target battery pack rises sharply.

[0090] Specifically, the overcharge or over-discharge of individual cells in the target battery pack is determined based on either a parameter used to determine whether a single cell is overcharged or over-discharged. As an example, the overcharge or over-discharge of individual cells in the target battery pack can be determined based on the predicted State of Charge (SOC) of the individual cells in the target battery pack, as specified in the parameters used for judgment.

[0091] A voltage drop in the target battery pack can refer to a rate of voltage decrease in the target battery pack that exceeds a rate threshold. If the predicted voltage of the target battery pack determined at the current moment is less than the predicted voltage of the target battery pack determined at a previous moment, the predicted voltage of the target battery pack at the previous moment can be subtracted from the predicted voltage of the target battery pack determined at the current moment to obtain a voltage difference. Dividing this voltage difference by the time difference between the current moment and the previous moment yields the rate of voltage decrease of the target battery pack.

[0092] In one possible implementation, multiple risk conditions include: a first consultation-level warning condition, a second consultation-level warning condition, and a third consultation-level warning condition; the first consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than a first thermal runaway probability threshold and the predicted SOH corresponding to the target battery pack is less than the SOH threshold; the second consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than the first thermal runaway probability threshold and the internal resistance of at least one individual cell in the target battery pack continues to increase abnormally; the third consultation-level warning condition includes: the thermal runaway probability corresponding to the target battery pack is less than the first thermal runaway probability threshold and the amount of electrolyte vapor in the parameters used for judgment is at the trace level; the protective measures corresponding to the target consultation-level warning condition include: limiting the charging and discharging power of the target battery pack, wherein the target consultation-level warning condition is any one of the first consultation-level warning condition, the second consultation-level warning condition, and the third consultation-level warning condition.

[0093] It should be noted that the amount of electrolyte vapor in the parameters used for judgment is actually the amount of electrolyte vapor in the sensor data.

[0094] As an example, the first thermal runaway probability threshold is 0.01.

[0095] As an example, the charging and discharging power of the target battery pack is limited to below 80% of the rated charging and discharging power by limiting the charging and discharging power of the target battery pack through EMS.

[0096] In one possible implementation, multiple risk conditions include: a first severe fault warning condition, a second severe fault warning condition, a third severe fault warning condition, and a fourth severe fault warning condition; the first severe fault warning condition is a sudden voltage drop in the target battery pack; the second severe fault warning condition is overcharging or over-discharging at least one individual cell in the target battery pack; the third severe fault warning condition is that the concentration of H2 or CO in the parameters used for judgment exceeds a first concentration threshold; and the fourth severe fault warning condition is that the probability of thermal runaway is greater than a second thermal runaway probability threshold, wherein the second thermal runaway probability threshold is greater than the first thermal runaway probability threshold. Protective measures corresponding to the target severe fault warning conditions include: triggering a high-speed solid-state switch for electrical isolation, starting a high-power liquid / air cooling system, and enhancing ventilation. The target severe fault warning condition is any one of the first, second, third, or fourth severe fault warning conditions; triggering the high-speed solid-state switch is used for electrical isolation; starting the high-power liquid / air cooling system is used for forced heat dissipation; and enhancing ventilation is used to reduce the concentration of combustible gases.

[0097] It should be noted that the concentration of H2 or CO in the parameters used for judgment is actually the same as the concentration of H2 or CO in the sensor data.

[0098] As an example, the second thermal runaway probability threshold is 0.1.

[0099] In one possible implementation, multiple risk conditions include: a first fire pre-response condition, a second fire pre-response condition, and a third fire pre-response condition; the first fire pre-response condition is that the probability of thermal runaway is greater than a third thermal runaway probability threshold; the second fire pre-response condition is that the temperature of the target battery pack rises sharply; and the third fire pre-response condition is that the concentration of H2 or CO in the parameters used for judgment exceeds a second concentration threshold. Wherein, the third thermal runaway probability threshold is greater than the second thermal runaway probability threshold. Protective measures corresponding to the target fire pre-response condition include: releasing the main extinguishing agent and suppressing the measured temperature of the target battery pack below the thermal runaway critical point. The target fire pre-response condition is any one of the first, second, and third fire pre-response conditions, and the main extinguishing agent is released for forced cooling and chemical inerting.

[0100] As an example, the third thermal runaway probability threshold is 0.8.

[0101] A sharp rise in temperature can be defined as the rate of temperature change (dT / dt) of the target battery pack exceeding a temperature change rate threshold. The rate of temperature change of the target battery pack can be determined based on the predicted temperature of the target battery pack from the parameters used for this assessment. The predicted temperature of the target battery pack is determined at the current moment.

[0102] As an example, if the predicted temperature of the target battery pack determined at the current moment is higher than the predicted temperature of the target battery pack determined at a previous moment, the temperature difference can be obtained by subtracting the predicted temperature of the target battery pack determined at the current moment from the predicted temperature of the target battery pack determined at a previous moment. The temperature difference can then be divided by the time difference between the current moment and the previous moment to obtain the temperature change rate of the target battery pack.

[0103] Among them, the measured temperature of the target battery pack, i.e. the monitored temperature, can be the measured temperature of the single cell with the lowest measured temperature in the target battery pack.

[0104] In one possible implementation, multiple risk conditions include: a first fire confirmation condition and a second fire confirmation condition; the first fire confirmation condition is that the probability of thermal runaway is greater than a fourth thermal runaway probability threshold, wherein the fourth thermal runaway probability threshold is greater than a third thermal runaway probability threshold; the second fire confirmation condition is that a smoke detector alarm or a flame detector alarm; the protective measures corresponding to the target fire confirmation condition include: confirming the release of extinguishing agent and activating a backup flame suppression procedure; sending a fire alarm, wherein the target fire confirmation condition is either the first fire confirmation condition or the second fire confirmation condition; any one of the first fire confirmation conditions.

[0105] As an example, the primary extinguishing agent is pre-released as heptafluoropropane.

[0106] It should be noted that a thermal runaway probability greater than the fourth thermal runaway probability threshold can mean that the thermal runaway probability is approximately 1.

[0107] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0108] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0109] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A modular battery compartment integrating multiple monitoring parameters and a predictive safety management system, characterized in that: The modular battery compartment, which integrates multiple monitoring parameters and a predictive safety management system, includes: The container body, the target battery pack located inside the container body, the multi-parameter monitoring subsystem, the predictive safety early warning model subsystem, and the integrated anomaly and fire safety control subsystem; A multi-parameter monitoring subsystem is used to collect sensor data during the operation of the target battery pack; The predictive safety early warning model subsystem is used to determine the parameters for judgment based on sensor data. The parameters for judgment include: the predicted voltage of the target battery pack, the predicted temperature of the target battery pack, the predicted SOC of the target battery pack, and the predicted SOH of the target battery pack. Among them, determining the parameters for judgment based on sensor data includes: generating a deviation term for each individual cell in the target battery pack according to an inverse Gaussian distribution, and determining the predicted voltage of the target battery pack based on the monitored voltage of each individual cell in the sensor data and the deviation term of each individual cell. The integrated anomaly and fire safety management subsystem is used to determine whether each of the multiple risk conditions is met based on the parameters used for judgment, and when it is determined that the risk condition is met, to provide early warning and protective measures corresponding to the risk condition.

2. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 1, characterized in that: The predicted voltage of the target battery pack is determined based on the monitored voltage of each individual cell in the sensor data and the deviation term of each individual cell, including: Based on the monitored voltage of each individual cell and the deviation term of each individual cell, the candidate voltage of the target battery pack is determined; The predicted voltage of the target battery pack is determined based on the candidate voltages of the target battery pack.

3. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 2, characterized in that: Determining the predicted voltage of the target battery pack based on the candidate voltages includes: The temperature compensation factor is determined based on the predicted temperature of the target battery pack and the preset reference temperature. The predicted voltage of the target battery pack is determined based on the candidate voltage and temperature compensation factor.

4. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 1, characterized in that: Based on the sensor data, the parameters used for the judgment include: Using a strong tracking Kalman filter, based on the coupled electro-thermal model, the predicted SOC and predicted temperature of each individual cell in the target battery pack are predicted. Based on the predicted SOC of each individual cell, the predicted SOC of the target battery pack is determined, and based on the predicted temperature of each individual cell, the predicted temperature of the target battery pack is determined.

5. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 1, characterized in that: The parameters used for the judgment also include: the thermal runaway probability corresponding to the target battery pack; and the parameters used for the judgment, determined based on sensor data, include: The probability of electrical signal anomalies is determined based on the measured voltage and predicted voltage of the target battery pack in the sensor data. The probability of thermal signal anomalies is determined based on the highest predicted temperature among all individual cells in the target battery pack and the temperature change rate coefficient of the target battery pack. The probability of chemical signal anomalies is determined based on the concentrations of various gases in the sensor data. These gases include electrolyte vapor, H2, and CO. The abnormal probabilities of the electrical signal, thermal signal, and chemical signal are fused to obtain the thermal runaway probability of the target battery pack.

6. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 5, characterized in that: Multiple risk conditions include: a first consultation-level warning condition, a second consultation-level warning condition, and a third consultation-level warning condition. The first consultation-level warning condition includes: the thermal runaway probability of the target battery pack is less than the first thermal runaway probability threshold and the predicted state of equilibrium (SOH) of the target battery pack is less than the SOH threshold. The second consultation-level warning condition includes: the thermal runaway probability of the target battery pack is less than the first thermal runaway probability threshold and the internal resistance of at least one individual cell in the target battery pack continues to increase abnormally. The third consultation-level warning condition includes: the thermal runaway probability of the target battery pack is less than the first thermal runaway probability threshold and the amount of electrolyte vapor in the parameters used for judgment is at the trace level. Protective measures corresponding to the target consultation-level warning conditions include: limiting the charging and discharging power of the target battery pack. The target consultation-level warning condition is any one of the first, second, or third consultation-level warning conditions.

7. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 5, characterized in that: Multiple risk conditions include: first serious fault warning condition, second serious fault warning condition, third serious fault warning condition, and fourth serious fault warning condition; The first severe fault warning condition is a sudden voltage drop in the target battery pack; the second severe fault warning condition is overcharging or over-discharging of at least one individual cell in the target battery pack; the third severe fault warning condition is that the concentration of H2 or CO in the parameters used for judgment exceeds a first concentration threshold; and the fourth severe fault warning condition is that the thermal runaway probability is greater than a second thermal runaway probability threshold, wherein the second thermal runaway probability threshold is greater than the first thermal runaway probability threshold. Protective measures corresponding to the target severe fault warning conditions include: triggering a high-speed solid-state switch for electrical isolation, starting a high-power liquid / air cooling system, and enhancing ventilation. The target severe fault warning condition is any one of the first, second, third, or fourth severe fault warning conditions; triggering the high-speed solid-state switch is used for electrical isolation; starting the high-power liquid / air cooling system is used for forced heat dissipation; and enhancing ventilation is used to reduce the concentration of combustible gases.

8. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 5, characterized in that: Multiple risk conditions include: a first fire pre-response condition, a second fire pre-response condition, and a third fire pre-response condition; the first fire pre-response condition is that the probability of thermal runaway is greater than the third thermal runaway probability threshold; the second fire pre-response condition is that the temperature of the target battery pack rises sharply; the third fire pre-response condition is that the concentration of H2 or CO in the parameters used for judgment exceeds the second concentration threshold; wherein, the third thermal runaway probability threshold is greater than the second thermal runaway probability threshold, and the protective measures corresponding to the target fire pre-response condition include: releasing the main extinguishing agent and suppressing the measured temperature of the target battery pack below the thermal runaway critical point, wherein the target fire pre-response condition is any one of the first fire pre-response condition, the second fire pre-response condition, and the third fire pre-response condition, and releasing the main extinguishing agent for forced cooling and chemical inerting.

9. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 5, characterized in that: Multiple risk conditions include: a first fire confirmation condition and a second fire confirmation condition; the first fire confirmation condition is that the probability of thermal runaway is greater than a fourth thermal runaway probability threshold, wherein the fourth thermal runaway probability threshold is greater than a third thermal runaway probability threshold; the second fire confirmation condition is that a smoke detector alarm or a flame detector alarm; the protective measures corresponding to the target fire confirmation condition include: confirming the release of extinguishing agent and activating the backup flame suppression procedure; sending a fire alarm, wherein the target fire confirmation condition is either the first fire confirmation condition or the second fire confirmation condition; any one of the first fire confirmation conditions.

10. The unit-type battery compartment with integrated multi-monitoring parameters and predictive safety management system according to claim 1, characterized in that: The silo has fireproof and thermal insulation capabilities. The outer shell of the silo is made of high-strength fire-resistant material, and the inner wall of the silo is coated with heat-activated intumescent fire-retardant coating. The heat-activated intumescent fire-retardant coating is used to form a heat insulation layer at high temperatures. The interior of the silo is divided into partitions by heat-insulating partitions to slow down heat spread. The unit-type battery compartment, which integrates multiple monitoring parameters and a predictive safety management system, also includes a non-contact infrared thermal imaging sensor for capturing the two-dimensional temperature distribution of the target battery pack. This capture of the two-dimensional temperature distribution of the target battery pack is used to detect localized abnormal hot spots within the target battery pack.

Images