Fire control method and system for liquid cooling energy storage cabinet

Through the multi-dimensional data acquisition of support vector machine algorithm and composite detectors, real-time judgment and step-by-step early warning of complex fire status of energy storage cabinets is achieved, and the problem of inaccurate early warning in the existing technology is solved, and the energy storage cabinet is improved fire prevention and control capabilities and safety.

CN118949333APending Publication Date: 2024-11-15江苏林洋储能技术有限公司
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411457110.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-18
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing energy storage cabinet fire protection system cannot be promptly and accurately warned in complex fire conditions, and cannot effectively identify multi-dimensional fire precursors, resulting in inaccurate early fire warnings or delayed responses, reducing the fire prevention and control capabilities and safety of energy storage cabinets.

Method used

The support vector machine algorithm is used for model training, and the composite detector in the energy storage cabinet collects multi-dimensional data of CO concentration, temperature, smoke concentration and VOC concentration in real time to form an input vector, and conduct real-time judgment and classification of fire status, and gradually upgrade it The warning level triggers the corresponding fire extinguishing measures.

Benefits of technology

Through real-time judgment of multi-dimensional data, the problem of inaccurate or delayed response in early fire warnings is avoided, the fire safety and response speed of energy storage cabinets are improved, and timely early warning and fire extinguishing measures are ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118949333A_ABST
    Figure CN118949333A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of energy storage system fire safety, and discloses a liquid cooling energy storage cabinet fire control method and system. The method comprises the steps that CO concentration, temperature, smoke concentration and VOC concentration data are collected in real time through a composite detector, intelligent classification of a fire state and a non-fire state is conducted through a support vector machine algorithm, and corresponding fire extinguishing measures are triggered according to different early warning levels; compared with a traditional single parameter threshold value judgment method, particularly in the initial stage of a fire, the technical problem that complex fire precursor cannot be effectively recognized is solved. According to the invention, the support vector machine model can carry out real-time judgment according to the change trend of the multi-dimensional data, and gradually upgrade the early warning level, thereby avoiding the problem of inaccurate early fire early warning or delayed response, and improving the fire safety and response speed of the energy storage cabinet.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of fire safety of energy storage systems, and in particular relates to a fire control method and system for a liquid-cooled energy storage cabinet. Background Art

[0002] At present, with the widespread application of energy storage systems in industry, commerce and power systems, the safety of energy storage cabinets has received increasing attention. However, the existing fire protection systems of energy storage cabinets still have shortcomings in dealing with complex fire conditions. For example, traditional fire protection systems are usually based on simple threshold judgments such as single temperature or smoke concentration, and cannot timely and accurately warn of the occurrence of fires. Since fires inside energy storage cabinets are mostly caused by the combined action of multiple factors (such as increased CO concentration, excessive temperature and increased smoke concentration), existing technologies cannot fully meet the needs of comprehensive judgment of multi-dimensional fire precursors. Therefore, there is an urgent need for a fire control method that can perform intelligent analysis and judgment based on multi-dimensional data under complex working conditions of energy storage cabinets. Even if the fire precursors are not fully manifested, timely warnings can still be achieved and corresponding fire extinguishing measures can be taken to improve the fire prevention and control capabilities and safety of energy storage cabinets. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to propose a liquid-cooled energy storage cabinet fire control method, aiming to solve the technical problem that the traditional single parameter threshold judgment method cannot effectively identify complex fire precursors, especially in the early stage of a fire.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a fire control method for a liquid-cooled energy storage cabinet. The fire control method of the liquid-cooled energy storage cabinet comprises: Step S10: The composite detector in the energy storage cabinet collects data on CO concentration, temperature, smoke concentration and VOC concentration in real time to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; Step S20: using a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y

[0005] The following optimization objective function is used during training:

[0006] in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; Step S30: During the operation of the energy storage cabinet, the collected multidimensional data vector X is input into the support vector machine model in real time to calculate the decision function

[0007] in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; Step S40: After the support vector machine determines that the system has entered the first-level warning state, the system further determines whether one of the following conditions is met: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered and the sampling frequency is accelerated; Step S50: After the first-level warning signal is triggered, when the support vector machine decision value continues to increase and meets one of the following conditions, the second-level warning state is entered: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the secondary warning signal is triggered, the fire host starts the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; Step S60: When the support vector machine decision value reaches the preset threshold and meets all the following conditions, the system enters the third-level warning state: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures, starts the fully enclosed fire extinguishing mode, turns off the air intake and exhaust fans, and performs three sprayings, each with a spraying volume of 1kg, with an interval of 15 minutes; Step S70: After the detection warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

[0008] Preferably, in step S10, the composite detector is installed beside each liquid-cooled battery PACK in the energy storage cabinet to collect data once per second.

[0009] Preferably, in step S20, the optimal hyperplane is determined by maximizing the support vector interval.

[0010] Preferably, in step S30, the multidimensional data vector X includes not only the current CO concentration, smoke concentration and temperature data, but also the historical data within a preset time window in the past.

[0011] Preferably, in step S40, when the first-level warning signal is triggered, the currently collected CO concentration, smoke concentration, temperature and VOC concentration data are automatically recorded and stored.

[0012] Preferably, in step S60, the cabinet door is automatically locked in the fully enclosed fire extinguishing mode.

[0013] Preferably, in step S60, the fully enclosed fire extinguishing mode is started to support cooling and fire extinguishing in combination with the water fire fighting interface.

[0014] The present invention also provides a liquid-cooled energy storage cabinet fire control system comprising: The data acquisition module is used to collect data on CO concentration, temperature, smoke concentration and VOC concentration in real time through the composite detector in the energy storage cabinet to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; A model training module, used to use a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y

[0015] The following optimization objective function is used during training:

[0016] in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; The real-time classification module is used to input the collected multi-dimensional data vector X into the support vector machine model in real time during the operation of the energy storage cabinet to calculate the decision function

[0017] in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; The first-level warning judgment module is used to determine whether one of the following conditions is met after the support vector machine determines that the system has entered the first-level warning state: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered and the sampling frequency is increased; The second-level warning judgment module is used to enter the second-level warning state when the support vector machine decision value continues to increase after the first-level warning signal is triggered and one of the following conditions is met: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the second-level warning signal is triggered, the fire host issues a first-level alarm, activates the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; The three-level warning judgment module is used to enter the three-level warning state when the support vector machine decision value reaches the preset threshold and meets all the following conditions: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures. The fire host issues a second-level alarm, shuts down the air intake and exhaust fans, and starts a 30-second countdown. After the countdown ends, the spraying of the agent begins. The loop execution module is used to detect that after the early warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

[0018] The present invention also provides a liquid-cooled energy storage cabinet fire control device comprising: A memory, a processor, and a liquid-cooled energy storage cabinet fire control program stored in the memory and executable on the processor, wherein the liquid-cooled energy storage cabinet fire control method is implemented when the liquid-cooled energy storage cabinet fire control program is executed by the processor.

[0019] The present invention also provides a computer program product, including a liquid-cooled energy storage cabinet fire control program, and the liquid-cooled energy storage cabinet fire control method is implemented when the liquid-cooled energy storage cabinet fire control program is executed by a processor.

[0020] The beneficial effect of the present invention is that compared with the traditional single parameter threshold judgment method, especially in the early stage of a fire, the technical problem of being unable to effectively identify complex fire precursors. Through the present invention, the support vector machine model can make real-time judgments based on the changing trend of multidimensional data and gradually upgrade the warning level, thereby avoiding the problem of inaccurate early fire warnings or delayed responses, and improving the fire safety and response speed of the energy storage cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 This is a schematic flow chart of a first embodiment of a fire control method for a liquid-cooled energy storage cabinet of the present invention.

[0023] Figure 2 It is a schematic diagram of equipment for a fire control method of a liquid-cooled energy storage cabinet of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Embodiment 1: Figure 1 , which is a flow chart of the first embodiment of the fire control method for a liquid-cooled energy storage cabinet of the present invention, and provides the first embodiment of the fire control method for a liquid-cooled energy storage cabinet of the present invention.

[0026] Step S10: The composite detector in the energy storage cabinet collects data on CO concentration, temperature, smoke concentration and VOC concentration in real time to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; It should be noted that The CO concentration at the current moment is displayed. It is used to monitor the combustion byproducts in the battery energy storage cabinet and is collected in real time through a composite detector. The increase in CO concentration is often one of the early manifestations of abnormal chemical reactions in the battery. Indicates the current temperature, which is one of the key parameters inside the energy storage cabinet. When the temperature is too high, it often indicates that the battery is overheating or thermal runaway. Indicates the smoke density at the current moment, which indicates the possible presence of burning or increased volatile materials, usually associated with a fire or combustion. Indicates the concentration of volatile organic compounds at the current moment. Detecting volatiles helps to determine the decomposition or combustion status of the battery, especially when certain organic electrolytes in the battery decompose, a large amount of volatile organic compounds will be released.

[0027] It should be understood that the real-time and accuracy of data are crucial. CO concentration, temperature, smoke concentration and VOC concentration are accurately collected by composite detectors, and the collection frequency needs to be high enough to ensure that the system can capture changes in the early stages of a fire. These parameters are not the sole basis for judging a fire, but are a combination of multi-dimensional features, using the classification capabilities of the support vector machine model to evaluate the fire status in real time and dynamically.

[0028] For example, the combination of CO(t) ≥ 400 ppm and T(t) ≥ 65°C may indicate that the battery is in the early stages of thermal runaway. At this time, the support vector machine will output an early warning signal, prompting the system to enter the first-level warning state. When S(t) ≥ 0.9 dB / m and VOC(t) ≥ 700 ppm, it may mean that the electrolyte has decomposed inside the battery, and the risk of fire has increased. The system will automatically enter the second-level warning state and take appropriate fire extinguishing measures.

[0029] Through the coordination of these parameters and support vector machines, the fire warning capability of the system can be improved and the safe operation of the energy storage cabinet can be ensured.

[0030] Step S20: using a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y

[0031] The following optimization objective function is used during training:

[0032] in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; It should be noted that It is the objective optimization function of the support vector machine, which means finding the optimal hyperplane by minimizing this function to distinguish between the fire state and the non-fire state. is a feature weight vector, which determines the importance of each input feature in the final classification. Through training, the model will assign different weights to each input parameter (such as CO concentration, temperature, smoke concentration, VOC concentration, etc.). It is a bias term used to adjust the offset of the classification hyperplane so that the hyperplane can better adapt to the classification boundary. is a regularization parameter that is used to balance the relationship between classification error and model complexity. A larger CCC value means that the model pays more attention to reducing classification errors, while a smaller C value tends to make the model smoother.

[0033] It should be understood that the objective function Represents the classification error. For correctly classified samples, this value is 0; for incorrectly classified samples, this value is positive, indicating that the larger the error of the sample. The support vector machine ensures that the hyperplane can distinguish the data in the fire state and the non-fire state to the greatest extent by minimizing this loss function.

[0034] Step S30: During the operation of the energy storage cabinet, the collected multidimensional data vector X is input into the support vector machine model in real time to calculate the decision function

[0035] in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; It should be understood that when the system's decision value H(x) ≤ 0, it means that the current data does not show signs of fire, and the system can continue to operate normally and maintain a safe state. When H(x)>0, the system enters the first-level warning state, indicating that there may be signs of fire. This is the initial stage of fire detection, and the system will determine whether to take higher-level emergency measures based on further analysis.

[0036] For example, if the current CO concentration of the energy storage cabinet is 500 ppm, the temperature is 70°C, the smoke concentration is 0.6 dB / m, and the VOC concentration is 700 ppm, the data vector collected by the system is X={500,70,0.6,700}. Through the calculation of the support vector machine, if H(x)>0, the system determines that the energy storage cabinet has entered the first-level warning state, starts to issue an audible and visual alarm, and prompts further monitoring. On the contrary, if the current CO concentration and temperature are low, and the smoke concentration and VOC concentration are also within the safe range, the system calculates that H(x)≤0, then the system is judged to be in a safe state and continues to operate normally without triggering an alarm.

[0037] Step S40: After the support vector machine determines that the system has entered the first-level warning state, the system further determines whether one of the following conditions is met: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered and the sampling frequency is increased; It should be noted that the specific values ​​in the above conditions are determined after high and low temperature tests and offset correction.

[0038] It should be understood that the first-level warning system does not directly indicate the occurrence of a fire, but the system will sound an alarm at the early stage of detecting potential danger, with the aim of providing sufficient reaction time to prevent the fire from actually breaking out. When the first-level warning is triggered, the system will further monitor the relevant data to determine whether a higher-level response measure (such as a second-level warning or fire-fighting measures) is needed.

[0039] For example, suppose the system detects a CO concentration of 450 ppm, a smoke concentration of 0.4 dB / m, a temperature of 64°C, and a VOC concentration of 700 ppm. Although other conditions have not reached critical values, since the CO concentration exceeds 400 ppm, the system will determine that the first-level warning conditions are met and issue an audible and visual alarm signal to alert the operator to abnormal conditions inside the energy storage cabinet. If these conditions continue to deteriorate, the CO concentration continues to rise to 800 ppm, or the temperature rises to 75°C, the system may further enter the second-level warning state and even trigger the fire extinguishing device.

[0040] Through such a graded warning mechanism, the system can take timely preventive measures in the early stages of a fire, thereby reducing the risk of fire and ensuring that operators can respond quickly.

[0041] Step S50: After the first-level warning signal is triggered, when the support vector machine decision value continues to increase and meets one of the following conditions, the second-level warning state is entered: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the secondary warning signal is triggered, the fire host issues a primary alarm, activates the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; It should be noted that the CO concentration reaches 800 ppm and the smoke concentration reaches 0.9 dB / m: When the CO concentration rises to 800 ppm, it indicates that combustion or chemical reactions are intensifying. The increase in smoke concentration (reaching 0.9 dB / m) further proves that there are a large number of combustion substances or by-products in the environment. The combination of these two indicates that the risk of fire is extremely high. The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C: When the smoke concentration reaches a certain threshold and the temperature continues to rise, the system can determine that combustion or chemical reactions may have occurred inside the energy storage cabinet. The temperature is greater than 75°C: The temperature rises above 75°C, which is usually the critical critical point for battery thermal runaway. After exceeding this temperature, the risk of fire is extremely high and may trigger a rapid heating and combustion reaction of the battery.

[0042] Step S60: When the support vector machine decision value reaches the preset threshold and meets all the following conditions, the system enters the third-level warning state: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures. The fire host issues a second-level alarm, shuts down the air intake and exhaust fans, and starts a 30-second countdown. After the countdown ends, the spraying of the agent begins. It should be noted that the CO concentration reaches 800 ppm and the smoke concentration reaches 0.9 dB / m and the state lasts for 10 seconds: This means that the fire situation is very serious, the CO concentration and smoke concentration have reached the limit, and this state lasts for 10 seconds, indicating that the fire continues to deteriorate. The CO concentration reaches 300 ppm and the smoke concentration reaches 0.5 dB / m and the temperature reaches 85°C: Although the CO concentration is low, combined with the increase in smoke concentration and temperature, the system determines that there is a great risk of fire at this time, and a combustion reaction may have occurred.

[0043] It should be understood that when the system enters the third-level warning state, it means that the fire extinguishing agent will be sprayed in a fully enclosed manner, and the air inlet and outlet fans will be shut down to ensure that the fire extinguishing agent can effectively act on the fire source to the greatest extent possible, and prevent external air from entering to avoid intensifying combustion.

[0044] Step S70: After the detection warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

[0045] It should be understood that the real-time monitoring of CO concentration, smoke concentration, temperature and VOC concentration is carried out after the fire extinguishing warning signal is executed. Even after the fire extinguishing signal is executed, the system continues to monitor the environment inside the energy storage cabinet to ensure that the fire state is effectively controlled or eliminated. The steps between S30 and S60 will continue to cycle, which means that in the first, second and third warning states, the system will continue to adjust the warning state according to the real-time monitoring data to ensure that the fire will not reignite or further deteriorate.

[0046] For example, suppose the system has triggered the third-level warning state and executed three spraying operations. However, after the fire extinguishing signal is executed, the system still continues to monitor the CO concentration, smoke concentration and temperature in the energy storage cabinet.

[0047] In addition, the present invention also provides a liquid-cooled energy storage cabinet fire control system, the liquid-cooled energy storage cabinet fire control system comprising: The data acquisition module is used to collect data on CO concentration, temperature, smoke concentration and VOC concentration in real time through the composite detector in the energy storage cabinet to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; A model training module, used to use a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y

[0048] The following optimization objective function is used during training:

[0049] in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; The real-time classification module is used to input the collected multi-dimensional data vector X into the support vector machine model in real time during the operation of the energy storage cabinet to calculate the decision function

[0050] in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; The first-level warning judgment module is used to determine whether one of the following conditions is met after the support vector machine determines that the system has entered the first-level warning state: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered; The second-level warning judgment module is used to enter the second-level warning state when the support vector machine decision value continues to increase after the first-level warning signal is triggered and one of the following conditions is met: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the second-level warning signal is triggered, the fire host issues a first-level alarm, activates the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; The three-level warning judgment module is used to enter the three-level warning state when the support vector machine decision value reaches the preset threshold and meets all the following conditions: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures. The fire host issues a second-level alarm, shuts down the air intake and exhaust fans, and starts a 30-second countdown. After the countdown ends, the spraying of the agent begins. The loop execution module is used to detect that after the early warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

[0051] The fire control system for a liquid-cooled energy storage cabinet provided by the present invention adopts a fire control method for a liquid-cooled energy storage cabinet in the above embodiment, which can solve the technical problem of fire control for a liquid-cooled energy storage cabinet. Compared with the prior art, the beneficial effects of the fire control system for a liquid-cooled energy storage cabinet provided by the present invention are the same as the beneficial effects of the fire control method for a liquid-cooled energy storage cabinet provided by the above embodiment, and the other technical features in the fire control system for a liquid-cooled energy storage cabinet are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0052] The present invention provides a liquid-cooled energy storage cabinet fire control device, please refer to Figure 2, a liquid-cooled energy storage cabinet fire control device includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor so that at least one processor can execute a liquid-cooled energy storage cabinet fire control method in the above-mentioned embodiment 1. A liquid-cooled energy storage cabinet fire control device in an embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. A liquid-cooled energy storage cabinet fire control device is only an example and should not bring any limitations to the functions and scope of use of the embodiments of the present invention. A liquid-cooled energy storage cabinet fire control device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of a liquid-cooled energy storage cabinet fire control device are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow a liquid-cooled energy storage cabinet fire control device to communicate wirelessly or wired with other devices to exchange data. Although a liquid-cooled energy storage cabinet fire control device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0053] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of a liquid-cooled energy storage cabinet fire control method as described above. The computer program product provided by the present invention can solve the technical problem of a liquid-cooled energy storage cabinet fire control. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as the beneficial effects of a liquid-cooled energy storage cabinet fire control method provided in the above embodiment, which will not be repeated here.

[0054] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present invention are executed.

[0055] It should be understood that the various parts disclosed in the present invention can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0056] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A fire control method for a liquid-cooled energy storage cabinet, It is characterized in that The fire control method of the liquid-cooled energy storage cabinet comprises: Step S10: The composite detector in the energy storage cabinet collects data on CO concentration, temperature, smoke concentration and VOC concentration in real time to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; Step S20: using a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y The following optimization objective function is used during training: in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; Step S30: During the operation of the energy storage cabinet, the collected multidimensional data vector X is input into the support vector machine model in real time to calculate the decision function in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; Step S40: After the support vector machine determines that the system has entered the first-level warning state, the system further determines whether one of the following conditions is met: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered and the sampling frequency is increased; Step S50: After the first-level warning signal is triggered, when the support vector machine decision value continues to increase and meets one of the following conditions, the second-level warning state is entered: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the secondary warning signal is triggered, the fire host issues a primary alarm, activates the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; Step S60: When the support vector machine decision value reaches the preset threshold and meets all the following conditions, the system enters the third-level warning state: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures. The fire host issues a second-level alarm, shuts down the air intake and exhaust fans, and starts a 30-second countdown. After the countdown ends, the spraying of the agent begins. Step S70: After the detection warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

2. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S10, the composite detector is installed on each liquid-cooled battery PACK in the energy storage cabinet to collect data once per second.

3. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S20, the optimal hyperplane is determined by maximizing the support vector interval.

4. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S30, the multidimensional data vector X includes not only the current CO concentration, smoke concentration and temperature data, but also the historical data within a preset time window in the past.

5. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S40, when the first-level warning signal is triggered, the currently collected CO concentration, smoke concentration, temperature and VOC concentration data are automatically recorded and stored.

6. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S60, the automatic locking of the cabinet door in the fully enclosed fire extinguishing mode is started.

7. A fire control method for a liquid-cooled energy storage cabinet as claimed in claim 1, It is characterized in that In step S60, the fully enclosed fire extinguishing mode is started to support cooling and fire extinguishing in combination with the water fire fighting interface.

8. A liquid-cooled energy storage cabinet fire control system, It is characterized in that The liquid-cooled energy storage cabinet fire control system includes: The data acquisition module is used to collect data on CO concentration, temperature, smoke concentration and VOC concentration in real time through the composite detector in the energy storage cabinet to form a multi-dimensional input vector ,in is the CO concentration, is the temperature, is the smoke concentration, is the concentration of volatile organic compounds; A model training module, used to use a support vector machine algorithm to perform model training to obtain an optimal hyperplane for segmenting a fire state and a non-fire state; The input data is a historical multidimensional input vector: Y The following optimization objective function is used during training: in, is the optimization objective function, is the feature weight vector, is the regularization parameter, is the total number of training data, i represents the i-th sample, is the bias term, is the category label of the training data, is the input feature vector of the i-th sample; The real-time classification module is used to input the collected multi-dimensional data vector X into the support vector machine model in real time during the operation of the energy storage cabinet to calculate the decision function in, is the decision function, is the weight vector, is the bias term, is the input vector; And classify the fire status according to the decision value: when When ≤0, the system is judged to be in a safe state; when >0, the system enters the first level warning state; The first-level warning judgment module is used to determine whether one of the following conditions is met after the support vector machine determines that it has entered the first-level warning state: (1) CO concentration reaches 400 ppm; (2) The smoke concentration reaches 0.5 dB / m; (3) Temperature greater than 65°C; (4) VOC concentration reaches 700 ppm; If any of the above conditions is met, a first-level warning signal is triggered and the sampling frequency is increased; The second-level warning judgment module is used to enter the second-level warning state when the support vector machine decision value continues to increase after the first-level warning signal is triggered and one of the following conditions is met: (1) CO concentration reaches 800 ppm and smoke density reaches 0.9 dB / m; (2) The smoke concentration reaches 0.5 dB / m and the temperature is greater than 65°C; (3) Temperature greater than 75°C; If any of the above conditions is met, the secondary warning signal is triggered, the fire host issues a primary alarm, activates the sound and light alarm, turns on the air intake and exhaust fans, and uploads the alarm signal to the EMS; The three-level warning judgment module is used to enter the three-level warning state when the support vector machine decision value reaches the preset threshold and meets all the following conditions: (1) The CO concentration reaches 800 ppm and the smoke density reaches 0.9 dB / m, and the above conditions last for 10 seconds; (2) CO concentration reaches 300 ppm, smoke density reaches 0.5 dB / m, and temperature reaches 85°C; If any of the above conditions is met, the third-level warning signal is triggered, and the system executes the third-level warning fire extinguishing measures. The fire host issues a second-level alarm, shuts down the air intake and exhaust fans, and starts a 30-second countdown. After the countdown ends, the spraying of the agent begins. The loop execution module is used to detect that after the early warning signal is executed, the CO concentration, smoke concentration, temperature and VOC concentration are monitored in real time, and steps S30 to S60 are continued.

9. A liquid-cooled energy storage cabinet fire control device, It is characterized in that The liquid-cooled energy storage cabinet fire control equipment includes: A memory, a processor, and a liquid-cooled energy storage cabinet fire control program stored in the memory and executable on the processor, wherein the liquid-cooled energy storage cabinet fire control program, when executed by the processor, implements the liquid-cooled energy storage cabinet fire control method described in any one of claims 1 to 7.

10. A computer program product, It is characterized in that The computer program product includes a liquid-cooled energy storage cabinet fire control program, and when the liquid-cooled energy storage cabinet fire control program is executed by a processor, the liquid-cooled energy storage cabinet fire control method described in any one of claims 1 to 7 is implemented.

Citation Information

Cited By

  • Impurity removal method and system for nickel-cobalt-manganese ternary precursor of lithium battery

    CN120698630A