CN-121971818-A - Fire control combined control system for multiple energy storage devices

CN121971818ACN 121971818 ACN121971818 ACN 121971818ACN-121971818-A

Abstract

The invention discloses a fire control combined control system for multiple energy storage devices, which belongs to the technical field of fire control of the energy storage devices, and comprises an information acquisition module, a fire execution module, an integrated multi-channel spraying system, an inert gas injection module and a physical isolation mechanism, wherein the information acquisition module comprises a basic perception layer for judging a situation after a fire occurrence, a prediction perception layer data analysis decision module for predicting the situation before the fire occurrence, a fire situation assessment unit for assessing a fire situation development situation and a potential spreading path in real time, and a thermal runaway risk assessment unit for performing risk index calculation after assessing by adopting an improved entropy weight-TOPSIS model, the cooperative control module is used for optimizing fire extinguishing resource scheduling based on a dynamic game theory, generating a multi-device linkage strategy by a priority weight algorithm, and executing grading response fire control. The intelligent coordinated control of the multiple energy storage devices, accurate prediction of thermal runaway risk and optimization of fire extinguishing resource allocation can be realized.

Inventors

GU YI
WU JIAN
ZHAO JUNLIANG
YANG CHAORAN
LIU WEI
CHENG QIAN
PEI JIE
CAO CHUANZHAO
PING XIAOFAN
SUN ZHOUTING
CHEN WENBO
LEI HAODONG
LIU JINGHAN
ZHANG DE
ZHANG JIANGTAO
WANG JIAYUN
CAO XI
ZHOU LIREN
LIU MINGYI
LIU BO
LI XIAOCHEN
TONG GUOPING
WANG YANLING

Assignees

华能国际电力股份有限公司上海石洞口第二电厂
中国华能集团清洁能源技术研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20251022

Claims (9)

1. A fire control system for a multi-energy storage device, comprising: The information acquisition module comprises a basic sensing layer and a prediction sensing layer, wherein the basic sensing layer is used for acquiring temperature signals, smoke signals, VOC gas signals and pressure signals in real time and judging situations after fire occurrence; The data analysis decision module comprises a fire situation assessment unit and a thermal runaway risk assessment unit, wherein the fire situation assessment unit is used for assessing the fire situation development situation and the potential propagation path in real time, and the thermal runaway risk assessment unit fuses the characteristics based on a dynamic weighted fusion model Mapping to a thermal runaway key parameter space, and calculating a risk index after evaluating by adopting an improved entropy weight-TOPSIS model; the cooperative control module optimizes fire extinguishing resource scheduling based on a dynamic game theory and generates a multi-equipment linkage strategy through a priority weight algorithm; and the fire control execution module is integrated with the multi-channel spraying system, the inert gas injection module and the physical isolation mechanism and is used for executing graded response fire control.
2. The fire control system for a multi-energy storage device of claim 1, wherein the fire control method of the fire control system comprises the steps of: s1, creating a virtual game agent for each energy storage device and initializing a policy space , wherein, The fire control execution module is registered and comprises a spraying unit arranged at the top of the energy storage equipment Inert gas injection unit disposed around energy storage device Fire barrier unit arranged between two adjacent energy storage devices And sets the upper limit of the fire control execution module capacity of each unit ; S2, setting a benefit function comprising the benefit function of the energy storage equipment And a fire execution module revenue function : In the formula, As an index of the risk (risk) of, For the severity of the fire of the device i, The amount of fire extinguishing resources requested for device i, For the thermal runaway critical threshold of device i, As an initial value of the weight, the weight is, For the resource consumption penalty factor, In order to implement the cost factor of the present invention, For the total number of associated devices, To allocate fire extinguishing resources from the execution unit j to the transmission loss factor of the device i, To perform the fire extinguishing efficiency coefficient of unit j for device i. S3, real-time game strategy flow, namely when the prediction perception layer detects the thermal runaway risk, evaluating a risk index R through a thermal runaway risk evaluation unit, adjusting a benefit function according to the risk index R, and outputting an optimal combination strategy of each unit in the fire control execution module Performing accurate resource allocation based on game results Executing closed loop detection once in a preset time period to perform effect feedback and strategy correction, and outputting all units in the fire control execution module to adopt all combined strategies when the basic sensing layer detects fire And the fire extinguishing efficiency is evaluated by a fire situation evaluation unit.
3. The fire control system for multiple energy storage devices according to claim 2, wherein in step S3, the policy optimization method is as follows: In the formula, In order for the rate of learning to be high, For the purpose of instant rewards, As a discount factor, the number of times the discount is calculated, To take action a in state s, To be in the next state All possible actions below Maximum of (3) Values.
4. The fire control system for a multi-energy storage device of claim 3, wherein the fire situation assessment unit comprises the following steps: a1, temperature is measured Smoke concentration VOC concentration Pressure variation The sensor data of the system is normalized and weighted fused to generate a comprehensive fire index F: In the formula, Is a weight coefficient of the temperature parameter, Is a weight coefficient of the smoke concentration, Is a weight coefficient of the concentration of the VOC, Is a weight coefficient of the pressure and is used for the pressure, As a result of the ambient initial temperature, Is the air pressure of the air, and is the air pressure of the air, 、、、 The upper limit of the sensor range of the temperature sensor, the smoke sensor, the VOC gas sensor and the pressure sensor is respectively; a2, confirming fire occurrence through threshold judgment and Bayesian probability model, and dividing fire grades: In the formula, Is the posterior probability of the occurrence of a fire, When a fire occurs, the sensor data meet the conditional probability of the characteristic comprehensive fire index F, For the prior probability of a fire occurrence, When a fire occurs, the data of each sensor meets the edge probability of the characteristic comprehensive fire index F; A3, quantifying fire energy based on a heat release rate model: In the formula, In order to be able to achieve a heat release rate from the fire, For the coefficient of integration to be the same, Is the temperature of the fire source; a4, calculating heat flux based on Stefan-Boltzmann law: In the formula, In order to achieve a heat flux density, For the Stefan-Boltzmann constant, For the emissivity of the battery material, Is the absolute temperature of the surface of the fire source, The temperature of the air is the absolute temperature of the environment, Is the convective heat transfer coefficient; a5, predicting gas diffusion by adopting a Gaussian plume model: In the formula, Is the concentration of the fire heat release in the three-dimensional space, Is a three-dimensional coordinate axis, In order to be able to achieve a heat release rate from the fire, For the average wind speed, For the diffusion length of the fire heat release in the y-axis direction, For the diffusion length of the fire heat release in the z-axis direction, Is the height of the smoke discharging channel of the energy storage equipment.
5. The fire control system for a multi-energy storage device of claim 4, wherein the method of assessing a thermal runaway risk assessment unit comprises the steps of: b1, constructing a fusion model to execute real-time feature extraction and fusion calculation: In the formula, In order to fuse the features of the features, Is a time-frequency matrix of sound waves, Is a gas concentration gradient matrix, In the case of a thermodynamic temperature field, In the case of a matrix of image pixels, As a function of the non-linear mapping, As the mode identifier of the user, For the fused weight coefficient of each mode, Is that The norm is normalized and the result is that, Is a matrix of trainable parameters specific to a modality, In order to modify the linear cell activation function, A convolution kernel is extracted for the modal feature, In order to be the original multi-modal data, Modal bias parameters; the method is convolution operation and is used for cross-modal feature extraction; B2. mapping the fusion feature F to a thermal runaway key parameter space: In the formula, In order to provide a rate of change of temperature, For the gradient of the change in the concentration of hydrogen, A second derivative for thermal imaging; b3 risk index Calculation of B4 dynamic threshold adjustment B5 risk level mapping and decision making In the formula, Is a coefficient of proportionality and is used for the control of the power supply, As a matrix of weights, the weight matrix, Is a bias term.
6. The fire control system for multiple energy storage devices of claim 1, wherein the base sensing layer comprises distributed temperature sensors, smoke detectors, VOC gas sensors and pressure sensors, wherein the sensors are arranged in a redundant manner to cover key monitoring points of the energy storage devices.
7. The fire control combined control system for the multi-energy storage device according to claim 1, wherein the acoustic signal acquisition of the predictive sensing layer is performed by an ultrasonic detection device for identifying abnormal acoustic characteristics of lithium precipitation or diaphragm rupture inside the battery.
8. The fire control combined control system for the multi-energy storage device according to claim 1, wherein the multi-channel spraying system of the fire control execution module comprises a high-pressure water mist mode for rapid cooling, a fine water mist mode for suppressing re-combustion and a directional spraying mode for precisely covering a fire source core area.
9. The fire control system for a multi-energy storage device of claim 1, wherein the physical isolation mechanism comprises a fire barrier, an explosion vent, and a fuse protector, wherein the fuse protector actively cuts off electrical connection when an abnormality in voltage between the battery packs is detected.

Description

Fire control combined control system for multiple energy storage devices Technical Field The invention belongs to the technical field of fire control of energy storage equipment, and particularly relates to a fire control combined control system for multiple energy storage equipment. Background With the large-scale application of renewable energy sources, electrochemical energy storage power stations are rapidly developing as important energy conditioning facilities. However, the risk of thermal runaway of energy storage batteries (particularly lithium ion batteries) has been a major concern limiting their safety applications. Most of the existing energy storage power station fire-fighting systems adopt passive response type design, namely, a fire extinguishing program is started after open fire or high temperature is detected, and the hysteresis fire-fighting measures are difficult to effectively control the chain reaction of thermal runaway of the battery. Particularly in large-scale energy storage power stations, when a plurality of energy storage devices simultaneously or sequentially generate thermal runaway, the existing fire-fighting system lacks effective global coordination control capability, and can cause problems of uneven fire-fighting resource distribution, untimely response and the like. In addition, different types of energy storage devices (such as lithium ion batteries, sodium sulfur batteries, flow batteries and the like) have different demands on fire extinguishing agents, and the existing systems often adopt a single fire extinguishing agent, so that the optimal fire extinguishing effect is difficult to realize. Disclosure of Invention The present invention aims to solve at least one of the technical problems in the related art to some extent. In order to achieve the above purpose, the present invention provides the following technical solutions: The invention provides a fire control combined control system for multiple energy storage devices, which comprises an information acquisition module, a prediction sensing layer and a control module, wherein the basic sensing layer is used for acquiring temperature signals, smoke signals, VOC gas signals and pressure signals in real time and judging the situation after a fire condition occurs; The data analysis decision module comprises a fire situation assessment unit and a thermal runaway risk assessment unit, wherein the fire situation assessment unit is used for assessing the fire situation development situation and the potential propagation path in real time, and the thermal runaway risk assessment unit fuses the characteristics based on a dynamic weighted fusion model Mapping to a thermal runaway key parameter space, and calculating a risk index after evaluating by adopting an improved entropy weight-TOPSIS model; the cooperative control module optimizes fire extinguishing resource scheduling based on a dynamic game theory and generates a multi-equipment linkage strategy through a priority weight algorithm; and the fire control execution module is integrated with the multi-channel spraying system, the inert gas injection module and the physical isolation mechanism and is used for executing graded response fire control. Further, the fire control method of the fire control combined control system comprises the following steps: s1, creating a virtual game agent for each energy storage device and initializing a policy space , wherein,The fire control execution module is registered and comprises a spraying unit arranged at the top of the energy storage equipmentInert gas injection unit disposed around energy storage deviceFire barrier unit arranged between two adjacent energy storage devicesAnd sets the upper limit of the fire control execution module capacity of each unit; S2, setting a benefit function comprising the benefit function of the energy storage equipmentAnd a fire execution module revenue function: In the formula,As an index of the risk (risk) of,For the severity of the fire of the device i,The amount of fire extinguishing resources requested for device i,For the thermal runaway critical threshold of device i,As an initial value of the weight, the weight is,For the resource consumption penalty factor,In order to implement the cost factor of the present invention,For the total number of associated devices,To allocate fire extinguishing resources from the execution unit j to the transmission loss factor of the device i,To perform the fire extinguishing efficiency coefficient of unit j for device i. S3, real-time game strategy flow, namely when the prediction perception layer detects the thermal runaway risk, evaluating a risk index R through a thermal runaway risk evaluation unit, adjusting a benefit function according to the risk index R, and outputting an optimal combination strategy of each unit in the fire control execution modulePerforming accurate resource allocation based on game resultsExecuting closed loop detection once in