Energy storage battery module in-cavity self-suppression staged fire extinguishing system

Abstract

The application provides an energy storage battery module cavity self-inhibition hierarchical fire extinguishing system, which comprises a battery module level cavity self-inhibition unit, a cluster level strong counter-jet injection fire extinguishing unit, a cabin level full-submersion cooling anti-reignition unit and an intelligent linkage control unit; the intelligent linkage control unit comprises a signal acquisition module, a digital twin prediction module and a fire linkage controller; the signal acquisition module acquires battery cell states and battery environment information; the digital twin prediction module constructs a full-scale model of "battery cell-module-cluster-cabin" based on battery thermal physical parameters, and predicts a thermal runaway spreading path; the FIC receives the SAM signal and the digital twin prediction result, executes a multi-level linkage discharge strategy of the battery module level cavity self-inhibition unit, the cluster level strong counter-jet injection fire extinguishing unit and the cabin level full-submersion cooling anti-reignition unit, and realizes early warning, accurate prevention and control and complete blocking of the battery energy storage fire.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electrochemical energy storage safety protection, and particularly relates to a self-inhibition hierarchical fire extinguishing system in a battery module cavity. BACKGROUND

[0002] With the wide application of lithium ion batteries in the field of energy storage, the risk of thermal runaway has become an important safety problem faced by the industry. The existing fire extinguishing system relies on external sensors or fixed fire extinguishing paths. If the lithium battery fire is not continuously cooled after the open fire is extinguished, the internal electrochemical reaction of the battery will continue, and the temperature of the battery will rise again and reignite. The temperature of thermal runaway is relatively low (about 140℃), the battery combustion temperature is high (the highest exceeds 700℃), and the open fire will be extinguished. The module itself is slow to dissipate heat. A large amount of flammable gas is generated during the thermal runaway process, which has an explosion risk in a closed space. The larger the capacity of the module, the greater the risk of fire. The traditional fire extinguishing system adopts a cabin-level full submersion method, which has a large response delay and lacks multi-level linkage control at the module level, cluster level and cabin level, resulting in low fire extinguishing efficiency and the risk of reignition. There is a lack of intelligent prevention and control scheme in the prior art that can achieve early self-inhibition in the battery module and form multi-level linkage with the cluster level and cabin level system. Therefore, there is an urgent need for a fire prevention and control system that can achieve multi-level cooperation from the module to the cabin, precise inhibition and prevention of reignition. SUMMARY

[0003] The purpose of the present application is to propose a multi-module-based multi-level linkage control to achieve a self-inhibition hierarchical fire extinguishing system for early warning, precise prevention and control and complete blocking of energy storage battery fires.

[0004] To achieve the above purpose, the present application provides a self-inhibition hierarchical fire extinguishing system in a battery module cavity, characterized in that it comprises a battery module level cavity self-inhibition unit, a cluster level strong opposing jet fire extinguishing unit, a cabin level full submersion cooling and reignition prevention unit and an intelligent linkage control unit.

[0005] The intelligent linkage control unit comprises a signal acquisition module, a digital twin prediction module and a fire linkage controller. The signal acquisition module acquires cell state and battery environment information. The digital twin prediction module constructs a full-scale model of "cell-module-cluster-cabin" based on battery thermal physical parameters, and predicts the thermal runaway time and spread path through the LSTM neural network in the digital twin module and the finite element analysis method. The fire linkage controller receives the SAM signal and the digital twin prediction result, and executes the multi-level linkage discharge strategy of the battery module level cavity self-inhibition unit, the cluster level strong opposing jet fire extinguishing unit and the cabin level full submersion cooling and reignition prevention unit.

[0006] Further, the battery module level cavity self-inhibition unit comprises at least three distributed thermally-driven phase change inhibition units embedded in the gaps between the battery module cells.

[0007] The thermal-force driven phase change suppression unit is a multi-layer composite structure, including an outer thermal sensitive trigger layer, a secondary outer layer, a middle reaction layer, and an inner buffer layer; the reaction layer contains perfluorohexanone precursor material;

[0008] The phase change material layer, i.e. the reaction layer, of the thermal-force driven phase change suppression unit is composed of 68% paraffin, 12% expanded graphite, 4.5% high thermal conductivity graphite film, and 15.5% composite flame retardant, and is packaged in an aluminum plastic bag; the flame retardant is a composite flame retardant system composed of ammonium polyphosphate and aluminum hypophosphite in a mass ratio of 2:1.

[0009] Further, the automatic trigger condition of the thermal-force driven phase change suppression unit is that the local ambient temperature ≥ T0 and the pressure ≥ P0; the T0 is dynamically adjusted with the number of battery cycles;

[0010] The adjustment model is: (℃);

[0011] In the formula, n is the number of battery cycles;

[0012] The pressure threshold P0 is fixed at 0.3 MPa ± 0.05 MPa.

[0013] Further, the cluster level strong counter-jet injection fire extinguishing unit is a plurality of high pressure nozzles arranged at the top or side of the battery cluster layer, the plurality of high pressure nozzles are all raised at 30°-45°, the high pressure nozzles on both sides are asymmetrically and crossly arranged; the outlet of the high pressure nozzle is provided with a cyclone, and the injection flow forms a strong counterflow field with a speed ≥ 2.5 m / s in the cluster, and the single flow rate of the nozzle can be dynamically adjusted between 5 L / min-12 L / min based on the measured heat spread rate;

[0014] The injection flow rate of the high pressure nozzle is dynamically adjusted based on the heat spread rate, and the adjustment formula is:

[0015] ;

[0016] In the formula, is the control parameter of the execution unit (nozzle), is the reference flow rate 5 L / min, is the measured heat spread rate, is the reference rate 0.5 m / s, when ≥ 1.5 m / s, the system will automatically switch to the maximum flow rate 12 L / min;

[0017] The dynamic adjustment formula of the fire extinguishing agent discharge amount is:

[0018] ;

[0019] wherein, is a system-level management parameter for total dose control, is a proportional coefficient (determined according to experiments), is a real-time temperature change rate, is a basic spray amount.

[0020] Further, the cabin-level full-submersion cooling anti-reburning unit comprises a full-submersion gas fire extinguishing device and a continuous cooling fine water mist system; the full-submersion gas fire extinguishing device is a fire extinguishing agent circulating module connected to the battery module; the continuous cooling fine water mist system is an atomizer and a spray head arranged in the battery module;

[0021] When the continuous cooling fine water mist system is working, the water mist conductivity is monitored based on the conductivity sensor, and if it exceeds the safety threshold, the full-submersion gas fire extinguishing is automatically switched to prevent electrical short circuit;

[0022] The safety threshold is that the conductivity σ of the atomized medium in the cabin is greater than 50 μS / cm ± 5 μS / cm.

[0023] Further, the atomized particle size of the continuous cooling fine water mist system is 35 μm ± 2 μm, and the spray intensity is adjusted in real time according to the temperature gradient in the cabin, and the specific adjustment strategy is:

[0024] (1) When the temperature gradient is less than 20 ℃ / m, the spray intensity is set to 4 L / (min·m²);

[0025] (2) When the temperature gradient is between 20-50 ℃ / m, the spray intensity is set to 8 L / (min·m²);

[0026] (3) When the temperature gradient is greater than 50 ℃ / m, the spray intensity is set to 12 L / (min·m²).

[0027] Further, the digital twin prediction module adopts a combination of digital twin prediction, LSTM algorithm and finite element analysis method to form a feedforward closed loop, and the linkage of each level is dynamically triggered based on the thermal runaway prediction signal;

[0028] The thermal runaway prediction algorithm based on the LSTM neural network has an input including voltage drop rate, temperature rise rate and flammable gas concentration change rate, and an output being a thermal runaway time prediction value ;

[0029] The thermal runaway prediction formula is: ;

[0030] wherein, is the predicted thermal runaway time, is the voltage change rate, is the temperature change rate, is the combustible gas concentration change rate;

[0031] Further, the finite element analysis method divides the battery module into cubic grid units with a side length of 5mm, and calculates the temperature change trend of each sub-region;

[0032] The heat spread path prediction algorithm is based on the following heat conduction equation:

[0033]

[0034] In the formula, is the thermal diffusivity, is the unit volume heat generation rate, is the material density, is the specific heat capacity, the prediction lead time is greater than or equal to 30s, and the prediction error is less than or equal to 7%.

[0035] Further, the fire linkage controller executes a three-level linkage spraying strategy of the module level, the cluster level and the cabin level, and the specific strategy is: when receiving a module level trigger signal, starting the module level intracavity self-inhibition unit; when the module level trigger signal lasts more than 30 seconds, starting the cluster level strong collision jet fire extinguishing unit; when the cluster level trigger signal lasts more than 60 seconds, starting the cabin level full submersion cooling anti-reignition unit.

[0036] Further, the battery module is provided with a battery cell, a battery cell monitoring unit and a radiator connected to the bottom of the sealed shell in the shell.

[0037] The battery cell monitoring unit is a controller integrating battery cell voltage, temperature, smoke and combustible gas monitoring units, and includes a battery cell voltage acquisition unit, a battery cell temperature acquisition unit, an intracabinet temperature acquisition unit, an intracabinet smoke acquisition unit, an intracabinet combustible gas acquisition unit, a power supply and a communication module connected with the microprocessor.

[0038] Compared with the prior art, the advantages of the present application are:

[0039] 1. The present application realizes early warning, accurate prevention and control and complete blocking of energy storage battery fire through multi-level linkage control of module level intracavity self-inhibition, cluster level intensified jet and cabin level full submersion cooling.

[0040] 2. The battery module level intracavity self-inhibition unit of the present application releases fire extinguishing gas based on the phase change principle, realizes early inhibition in the battery cavity, and the trigger threshold T0 is dynamically adjusted with the number of battery cycles, which fully guarantees the safety of the battery based on the battery habit.

[0041] ​3、The cluster level reinforced jet of the application is designed to make the extinguishing agent form a cyclone field in the cluster, enhancing the contact efficiency of the extinguishing agent with the heat source, while the arrangement of the nozzle makes the jet stream form a strong convection field inside the battery cluster, which can be dynamically adjusted based on the heat spread rate, for cluster level reinforced suppression and fire barrier for the fire that is still not controlled after intracavity suppression.

[0042] 4、The cabin level full submersion cooling anti-reignition unit of the application contains a full submersion gas extinguishing device and a continuous cooling fine water mist system, with an electrical conductivity sensor to monitor the electrical conductivity of the water mist, and if it exceeds the safety threshold, it will automatically switch to full perfluorohexone gas extinguishing to prevent electrical short circuits, further ensuring the safety of the battery circuit on the basis of fully realizing complete fire blocking. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 It is a schematic diagram of the overall architecture of the intracavity self-suppression grading fire extinguishing system of the energy storage battery module in the embodiment of the application.

[0044] Figure 2 It is a control flowchart of the intracavity self-suppression grading fire extinguishing system of the energy storage battery module in the embodiment of the application.

[0045] Figure 3 It is a schematic diagram of the digital twin model in the embodiment of the application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical scheme and advantages of the application more clear, the technical scheme of the application will be further described below.

[0047] As shown in Figure 1 The application proposes an intracavity self-suppression grading fire extinguishing system for energy storage battery module, which is composed of an intracavity self-suppression grading fire extinguishing system for energy storage battery module, a cluster level reinforced jet fire extinguishing unit, a cabin level full submersion cooling anti-reignition unit and an intelligent linkage control unit.

[0048] Specifically, the battery module level intracavity self-suppression unit contains three distributed heat-power driven phase change suppression units (TSD-PCMU) embedded in the gaps between the battery module cells, the TSD-PCMU in the module level intracavity self-suppression unit adopts a multi-layer composite structure, including an outer heat-sensitive trigger layer, a secondary outer layer, a middle reaction layer and an inner buffer layer, the reaction layer contains perfluorohexone precursor material which releases extinguishing gas after phase change from solid to gas after triggering; the heat-sensitive trigger layer of the TSD-PCMU is the outermost layer in direct contact with the cells, which is generally made of low melting point alloy (preferably bismuth-based alloy), and its precise melting point is the trigger threshold T0 under the current cycle number, and adopts a porous film or microcapsule packaging form to increase the thermal contact area.

[0049] The secondary outer layer adopts a pressure sensitive film, which is a composite material composed of microporous polytetrafluoroethylene (ePTFE) and a thermally responsive hydrogel.

[0050] The reaction layer is a phase change material layer, which is generally composed of a mesoporous SiO2 nanocomposite of perfluorohexanone precursor, and conforms to a "core-shell" structure. In this embodiment, it is mainly composed of 68% paraffin, 12% expanded graphite, 4.5% high thermal conductivity graphite film, and 15.5% composite flame retardant, and is packaged in an aluminum plastic bag; the flame retardant is a composite flame retardant system composed of ammonium polyphosphate and aluminum hypophosphite in a mass ratio of 2:1; the phase change material layer is a solid-gas phase change fire extinguishing reaction triggered by thermal-mechanical coupling, which is not a simple storage tank release.

[0051] As shown in Figure 2 , the TSD-PCMU is configured to be automatically triggered in response to the "and" logic condition of local ambient temperature ≥ T0 and pressure ≥ P0, to release fire extinguishing medium through a solid-gas conversion chemical reaction to achieve early suppression in the cavity, and T0 is dynamically adjusted with the number of battery cycles; the TSD-PCMU links a mechanical signal generator (MSG), which outputs a passive switching signal to the battery management system (BMS) when triggered;

[0052] The adjustment model is: (℃);

[0053] In the formula, n is the number of battery cycles;

[0054] The pressure threshold P0 is fixed at 0.3 MPa ± 0.05 MPa.

[0055] In this embodiment, the cluster level strong counter-jet injection fire extinguishing unit includes 8 groups of high pressure nozzles arranged at the top / side of the battery cluster layer, the nozzles are arranged in an asymmetric cross shape at an angle of 30°-45°, and the nozzle outlet is provided with a cyclone device to form a cyclone field in the cluster, thereby enhancing the contact efficiency of the fire extinguishing agent and the heat source. The injection flow in the cluster forms a strong convection field with a speed of ≥2m / s, and the arrangement of the nozzles forms a strong convection field in the battery cluster, which can be dynamically adjusted based on the heat spread rate, and is used for cluster level intensive suppression and fire prevention barrier for the fire that has not been controlled after the in-cavity suppression;

[0056] The injection parameter dynamic adjustment formula of the cluster level strong counter-jet injection fire extinguishing unit is:

[0057] ;

[0058] In the formula, is the control parameter of the execution unit (nozzle), is the reference flow rate 5L / min, is the measured heat spread rate, is the reference rate 0.5m / s, and when ≥1.5m / s the system will automatically switch to the maximum flow rate of 12 L / min;

[0059] The dynamic adjustment formula for the discharge amount of the extinguishing agent is:

[0060]

[0061] wherein, is a management parameter at the system level, used for total dose control, is a proportional coefficient (determined according to experiments), is the real-time temperature change rate, is the basic discharge amount.

[0062] In this embodiment, the cabin-level full-submersion cooling and anti-reignition unit comprises a full-submersion gas extinguishing device (the extinguishing agent is perfluorohexanone, and the discharge concentration is 8%-10%) and a continuous cooling water mist system, and the discharge intensity of the water mist system is adjusted in real time according to the temperature gradient in the cabin; the cabin-level water mist system is provided with an electric conductivity sensor for monitoring the conductivity of the water mist, and if the conductivity exceeds a safety threshold, the system will automatically switch to the perfluorohexanone gas extinguishing mode to prevent electrical short circuits.

[0063] Specifically, the perfluorohexanone gas extinguishing device comprises an extinguishing agent storage tank, a circulating pump, and an extinguishing agent recovery tank, the extinguishing agent storage tank is connected to the bottom of the battery module through a pipeline, the inner cavity of the extinguishing agent storage tank is higher than the inner cavity of the battery module, the extinguishing agent recovery tank is connected to the top of the battery module through a pipeline, the extinguishing agent recovery tank is provided with a refrigeration machine to achieve a recycling rate of the extinguishing agent of ≥85%. The extinguishing agent recovery tank is connected to the extinguishing agent storage tank through a recovery pipeline, and the circulating pump is installed on the recovery pipeline. The water mist system comprises an atomizer, a spraying pipeline, and a spray head, and the atomizer has the following parameters: working pressure: 18 MPa; atomized particle size: 35 μm±2 μm; spraying flow rate: 4-12 L / min.

[0064] The water mist system of the cabin-level full-submersion cooling and anti-reignition unit has an atomized particle size of 35 μm±2 μm, and the discharge intensity is adjusted in real time according to the temperature gradient in the cabin, and the specific adjustment strategy is as follows:

[0065] (1) When the temperature gradient is less than 20℃ / m, the discharge intensity is set to 4 L / (min·m²);

[0066] (2) When the temperature gradient is between 20-50℃ / m, the discharge intensity is set to 8 L / (min·m²);

[0067] (3) When the temperature gradient is greater than 50℃ / m, the discharge intensity is set to 12 L / (min·m²).

[0068] ​To further ensure the safety of fire suppression, the cabin-level fine water mist system is equipped with a conductivity sensor to monitor the conductivity of the water mist. If the conductivity exceeds the safety threshold, it will automatically switch to perfluorohexanone gas for fire suppression to prevent electrical short circuits.

[0069] In this embodiment, the intelligent linkage control unit includes a signal acquisition module (SAM), a digital twin prediction module, and a fire alarm linkage controller (FIC); the SAM communicates with all MSGs via a CAN bus (500kbps) to acquire trigger signals; such as Figure 2 As shown, the digital twin prediction module constructs a full-scale model of "cell-module-cluster-compartment" based on battery thermal properties to predict the thermal runaway propagation path. The purpose of predicting the thermal runaway propagation path is to achieve the dual functions of prediction and early warning, and precise prevention and control. Through the prediction of thermal runaway time (TP) using the digital twin model, the system can issue early warning signals when early fault symptoms such as voltage drop and rapid temperature rise appear in the battery. Furthermore, the predicted propagation path guides the precise response of cluster-level and compartment-level fire suppression units. For example, the fire control interlocking controller (FIC) can prioritize and enhance the flow rate of cluster-level nozzles and the cooling intensity of compartment-level nozzles in the direction of thermal spread based on the predicted path, achieving optimal allocation of fire suppression resources. The FIC receives SAM signals and digital twin prediction results, executes a multi-level linkage spray strategy of module-cluster-compartment, and controls the start-up, shutdown, and spray parameters of the latter two levels of units. The digital twin prediction module uses a thermal runaway prediction algorithm based on an LSTM neural network. Inputs include voltage drop rate, temperature rise rate, and combustible gas concentration change rate; the output is the predicted thermal runaway time. This is used to initiate suppression measures in advance.

[0070] Specifically, the digital twin prediction module of the intelligent linkage control unit constructs a full-scale model of "cell-module-cluster-cabin" based on battery thermal property parameters. The model uses a combination of digital twin prediction, LSTM algorithm, and finite element analysis to form a feedforward closed loop. The linkage of each level is dynamically triggered based on thermal runaway prediction signals, without manually set thresholds. In addition, the prediction error is ≤7%, and the prediction lead time is ≥30s.

[0071] The finite element method divides the battery module into cubic mesh elements with a side length of 5mm and calculates the temperature change trend of each sub-region. The heat spread path prediction algorithm is based on the following heat conduction equation:

[0072] ;

[0073] In the formula, Where is the thermal diffusivity, Heat production rate per unit volume For material density, For specific heat capacity, the prediction lead time is ≥30s;

[0074] The thermal runaway prediction formula is: ;

[0075] In the formula, is the predicted thermal runaway time, is the voltage change rate, is the temperature change rate, is the combustible gas concentration change rate.

[0076] In the embodiment, as shown in Figure 2 , the intelligent linkage control unit receives the SAM signal and the digital twin prediction result, and executes the module-cluster-cabin three-level linkage discharge strategy, and the specific strategy is as follows:

[0077] (1) When receiving the module-level trigger signal, start the module-level intracavity self-inhibition unit;

[0078] (2) When the module-level trigger signal lasts for more than 30 seconds, start the cluster-level strong-collision jet fire extinguishing unit;

[0079] (3) When the cluster-level trigger signal lasts for more than 60 seconds, start the cabin-level full-submersion cooling anti-reignition unit.

[0080] In the embodiment, as shown in Figure 3 , the shell of the battery module is internally provided with battery cells, a battery cell monitoring unit and a heat sink connected to the bottom of the sealed shell. The heat sink is either an air-cooled heat sink or a water-cooled heat exchanger. The battery cell monitoring unit is a controller integrating battery cell voltage, temperature, smoke and combustible gas monitoring units, including a battery cell voltage acquisition unit, a battery cell temperature acquisition unit, an intracabinet temperature acquisition unit, an intracabinet smoke acquisition unit and an intracabinet combustible gas acquisition unit connected to a microprocessor, a power supply and a communication module. The battery cell monitoring unit integrates a multi-parameter acquisition module, including: voltage acquisition accuracy ±5mV, sampling frequency 1kHz; temperature acquisition range -40℃~120℃, accuracy ±1℃; lower limit of combustible gas (CO, H2) detection ≤50ppm, response time ≤3s.

[0081] In the embodiment, the intelligent linkage control unit further includes an interface module in communication with the energy storage system BMS (battery management system) for real-time acquisition of battery SOC, SOH and charge-discharge current data, and prediction of thermal runaway risk level in combination with the digital twin model, and dynamic adjustment of each level of fire extinguishing.

[0082] In order to facilitate the understanding of those skilled in the art, the control method of the hierarchical fire extinguishing system of the present application is as follows:

[0083] S1: Module-level dual-parameter perception and intracavity self-inhibition. The distributed TSD-PCMU monitors the cell gap temperature and pressure in real time. When the dual thresholds of temperature ≥ 70℃ and pressure ≥ 90kPa are reached at the same time, the heat-sensitive trigger layer is instantly melted, triggering the solid-gas conversion of the phase change material, and completing the directional injection of perfluorohexone within 0.2s, achieving early intracavity inhibition of the module. The TSD-PCMU-linked mechanical signal generator (MSG) synchronously outputs passive switching signals to the signal acquisition module, ensuring that the fire signal can still be reliably transmitted when the circuit is interrupted.

[0084] S2: Digital twin prediction and cluster-level prediction. While the signal acquisition module receives the MSG signal, the digital twin module fuses multi-dimensional data such as voltage drop rate, temperature rise rate, and gas concentration change rate in real time, runs the LSTM prediction algorithm and finite element heat spread model to accurately predict the thermal runaway time Tp, and outputs the heat spread path. The system intelligently determines that either "module trigger duration ≥ 30 seconds" or "Tp < 60 seconds" is true, and immediately generates a cluster-level start command, realizing the leap from passive response to predictive intervention.

[0085] S3: Cluster-level dynamic rotational flow injection. The fire linkage controller activates the asymmetrically arranged high-pressure nozzles, dynamically adjusts the injection flow rate based on the real-time heat spread rate V, and forms a three-dimensional strong rotational flow field with a speed ≥ 2.5m / s through the design of 35°-40° inclination and rotational flow, accurately covering the complex structure of the battery cluster, and performing cluster-level intensive inhibition and fire prevention barrier, effectively blocking the heat spread between modules.

[0086] S4: Cabin-level safety anti-reignition. When the cluster-level trigger duration ≥ 60 seconds or Tp < 30 seconds, the system starts full submersion protection, first sprays 8%-10% concentration perfluorohexone to achieve full-area inertization and fire extinguishing, then starts fine water mist according to the cabin temperature gradient G, and adjusts the spraying intensity according to the cabin temperature gradient G; crucially, the conductivity sensor monitors the water mist conductivity in real time, and automatically switches to gas fire extinguishing when the conductivity exceeds the safety threshold of 55μS / cm, completely eliminating the risk of electrical short circuit.

[0087] S5: Closed-loop optimization of fire extinguishing system. The system continues to run until the strict termination condition of "cabin temperature ≤ 60℃ and maintained for 30 minutes" is met, and the actual heat spread path, extinguishing agent efficiency, etc. Data are fed back to the digital twin platform, and through the self-learning optimization of the LSTM network parameters, the system performance is continuously evolved, and an intelligent energy storage safety fire extinguishing system with growth is built.

[0088] The above merely describes the preferred embodiments of the present application and does not limit the present application in any way. Any person skilled in the art can make any form of equivalent replacement or modification to the technical solutions and technical contents disclosed by the present application without departing from the scope of the technical solutions of the present application, and such changes still belong to the protection scope of the present application.

Claims

1. A self-suppressing graded fire extinguishing system for the cavity of an energy storage battery module, characterized in that, This includes a battery module-level in-cavity self-suppression unit, a cluster-level high-pressure counter-current jet fire extinguishing unit, a cabin-level total flooding cooling and anti-reignition unit, and an intelligent linkage control unit; The intelligent linkage control unit includes a signal acquisition module, a digital twin prediction module, and a fire-fighting linkage controller. The signal acquisition module collects information on the cell status and battery environment. The digital twin prediction module constructs a full-scale model of "cell-module-cluster-cabin" based on the battery's thermal properties and predicts the thermal runaway time and propagation path using the LSTM neural network and finite element analysis method in the digital twin module. The fire-fighting linkage controller receives signals from the information acquisition module and the digital twin prediction results, and executes a multi-level linkage discharge strategy consisting of the battery module-level cavity self-suppression unit, the cluster-level strong counter-current jet extinguishing unit, and the cabin-level total flooding cooling and anti-reignition unit. The linkage at each level is triggered by predicting the thermal runaway time. The battery module-level cavity self-suppression unit includes a thermally driven phase change suppression unit; the thermally driven phase change suppression unit releases extinguishing gas based on the phase change principle. The automatic triggering conditions for the thermo-mechanically driven phase change suppression unit are: local ambient temperature ≥ T0 and pressure ≥ P0; T0 is dynamically adjusted according to the number of battery cycles. The model was adjusted as follows: ; In the formula, n is the number of battery cycles. The unit is ℃; The pressure threshold P0 is fixed at 0.3 MPa ± 0.05 MPa; The cluster-level high-pressure jet fire extinguishing unit consists of multiple sets of high-pressure nozzles arranged on the top or side of the battery cluster. The high-pressure nozzles on both sides are asymmetrically and cross-arranged. The outlet of the high-pressure nozzle is equipped with a cyclone separator, and the jet stream forms a strong convection field with a velocity ≥2.5m / s within the cluster. The jet flow rate of the high-pressure nozzle is dynamically adjusted based on the thermal spread rate, and the adjustment formula is: ； In the formula, These are the control parameters for the execution unit, which is a high-pressure nozzle; The baseline flow rate is 5 L / min. To measure the rate of heat spread, With a reference speed of 0.5 m / s, when When the speed is ≥1.5m / s, the system will Automatically switches to maximum flow rate of 12L / min; The formula for dynamically adjusting the extinguishing agent release rate is: ； In the formula, These are system-level management parameters used for total dose control. This is the proportionality coefficient, which is determined experimentally. This represents the real-time temperature change rate. Based on the basic discharge volume.

2. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 1, characterized in that, The battery module-level cavity self-suppression unit includes at least three distributed thermo-mechanically driven phase change suppression units embedded in the gaps between battery module cells. The thermo-mechanically driven phase change suppression unit is a multilayer composite structure, including an outer thermosensitive trigger layer, a second outer layer, a middle reaction layer, and an inner buffer layer; the reaction layer contains perfluorohexanone precursor material; The phase change material layer, or reaction layer, of the thermally driven phase change suppression unit is composed of 68% paraffin, 12% expanded graphite, 4.5% high thermal conductivity graphite film, and 15.5% composite flame retardant by mass percentage, and is packaged in an aluminum-plastic bag; the flame retardant is a composite flame retardant system composed of ammonium polyphosphate and aluminum hypophosphite in a mass ratio of 2:

1.

3. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 1, characterized in that, All the high-pressure nozzles are raised at 30°-45°, and the flow rate of each nozzle is dynamically adjusted between 5L / min and 12L / min based on the measured heat spread rate.

4. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 1, characterized in that, The cabin-level total flood cooling and reignition prevention unit includes a total flood gas fire extinguishing device and a continuous cooling fine water mist system; the total flood gas fire extinguishing device is a fire extinguishing agent circulation module connected to the battery module; the continuous cooling fine water mist system is an atomizer and nozzle installed in the battery module; When the continuous cooling fine water mist system is working, it monitors the conductivity of the water mist based on a conductivity sensor. If the conductivity exceeds the safety threshold, it automatically switches to perfluorohexanone gas for fire extinguishing to prevent electrical short circuits. The safety threshold is defined as the conductivity of the atomized medium inside the chamber, σ > 50 μS / cm ± 5 μS / cm.

5. The self-suppressing graded fire extinguishing system within the energy storage battery module cavity according to claim 4, characterized in that, The atomized particle size of the continuous cooling fine water mist system is 35μm±2μm, and the spray intensity is adjusted in real time according to the temperature gradient inside the chamber. The specific adjustment strategy is as follows: (1) When the temperature gradient is less than 20℃ / m, the spray intensity is set to 4L / (min·m²); (2) When the temperature gradient is between 20-50℃ / m, the spray intensity is set to 8L / (min·m²); (3) When the temperature gradient is greater than 50℃ / m, the spray intensity is set to 12L / (min·m²).

6. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 1, characterized in that, The digital twin prediction module uses a combination of digital twin prediction, LSTM algorithm and finite element analysis to form a feedforward closed loop, and the linkage of each level is dynamically triggered based on the thermal runaway prediction signal. The thermal runaway prediction algorithm based on LSTM neural network takes voltage drop rate, temperature rise rate, and combustible gas concentration change rate as inputs and outputs the predicted thermal runaway time. ; The formula for predicting thermal runaway is: ; In the formula, To predict thermal runaway time, The rate of change of voltage. For the rate of temperature change, This represents the rate of change in the concentration of combustible gas.

7. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 6, characterized in that, The finite element analysis method divides the battery module into cubic mesh elements with a side length of 5mm and calculates the temperature change trend of each sub-region. The heat propagation path prediction algorithm is based on the following heat conduction equation: ； In the formula, Where is the thermal diffusivity, Heat production rate per unit volume For material density, For specific heat capacity, the prediction lead time is ≥30s, and the prediction error is ≤7%.

8. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 7, characterized in that, The fire alarm control system implements a three-level linkage discharge strategy at the module level, cluster level, and compartment level. Specifically, when a module-level trigger signal is received, the module-level in-cavity self-suppression unit is activated; when the predicted thermal runaway time is less than 60 seconds, the cluster-level strong counter-current jet extinguishing unit is activated; and when the predicted thermal runaway time is less than 30 seconds, the compartment-level total flooding cooling and anti-reignition unit is activated.

9. The self-suppressing graded fire extinguishing system within the cavity of the energy storage battery module according to claim 1, characterized in that, The battery module housing contains battery cells, a battery cell monitoring unit, and a heat sink connected to the bottom of the sealed housing. The battery cell monitoring unit is a controller that integrates battery cell voltage, temperature, smoke, and combustible gas monitoring units. It includes a battery cell voltage acquisition unit, a battery cell temperature acquisition unit, an internal temperature acquisition unit, an internal smoke acquisition unit, and an internal combustible gas acquisition unit, all connected to a microprocessor, as well as a power supply and a communication module.

Images