Intelligent safety protection system and method for power supply charging of underground coal mine power lithium battery and computer readable storage medium

Abstract

The invention relates to the technical field of coal mine underground lithium battery thermal runaway monitoring and early warning, in particular to a coal mine underground power lithium battery power supply charging intelligent safety protection system and method. The system comprises a multi-element signal monitoring module which is used for collecting information and uploading the collected information; the intelligent early warning module is used for receiving the information uploaded by the multi-distance signal monitoring module, preprocessing the received information and carrying out feature mining and multi-distance data fusion analysis on historical data so as to realize dynamic grading early warning on the lithium battery thermal runaway, and the emergency disposal linkage module is used for receiving early warning information sent by the intelligent early warning module and sending the early warning information to the intelligent early warning module. Corresponding grading emergency disposal measures are executed according to the grading of the early warning information; and the central control module is used for receiving a user command and controlling the multi-element signal monitoring module, the intelligent early warning module and the emergency disposal linkage module according to the user command. According to the invention, the timeliness, accuracy and reliability of safety protection are improved.

Description

Technical Field

[0001] This invention relates to the field of monitoring and early warning technology for thermal runaway of lithium batteries in underground coal mines, and in particular to an intelligent safety protection system, method and computer-readable storage medium for charging power lithium batteries in underground coal mines. Background Technology

[0002] With the advancement of intelligent coal mine construction, equipment such as trackless rubber-wheeled vehicles and monorail cranes powered by lithium batteries are widely used underground. However, during the charging and discharging process of lithium batteries, especially under conditions of abuse such as overcharging and internal short circuits, lithium batteries are highly susceptible to thermal runaway, ultimately leading to fires and explosions. The toxic and flammable gases released by the combustion of lithium batteries, such as hydrogen fluoride, sulfur dioxide, and carbon monoxide, directly expose coal miners in the confined underground environment to highly toxic and carcinogenic substances, posing a serious threat to their health.

[0003] Currently, safety protection for lithium battery charging in coal mines largely relies on simple voltage and temperature threshold alarms and conventional fire-fighting facilities, which has the following shortcomings: Single monitoring: It only monitors a few parameters (such as total voltage and surface temperature), which cannot fully perceive the multi-dimensional risk signals of the battery's internal state and external environment, making it difficult to detect early hidden dangers in a timely manner.

[0004] Delayed early warning: Alarm methods based on fixed thresholds cannot identify the evolution of thermal runaway, resulting in low accuracy and high false alarm rate. Furthermore, they cannot achieve tiered early warning, which can easily lead to over- or under-response in emergency response.

[0005] The response was crude: there was a lack of precise emergency response plans linked to the warning level, the extinguishing agent was inefficient, the spraying method was limited, and it was difficult to effectively extinguish deep-seated lithium battery fires and prevent reignition.

[0006] Therefore, there is an urgent need for an intelligent security protection system that can achieve closed-loop management of the entire chain from "monitoring and perception" to "intelligent early warning" and then to "precise handling". Summary of the Invention

[0007] The present invention aims to at least partially solve one of the technical problems in the related art.

[0008] Therefore, the first objective of this invention is to provide an intelligent safety protection system for charging underground power lithium batteries in coal mines, comprising: The multi-signal monitoring module is used to acquire internal state information, external safety information, and environmental state information of the lithium battery, and upload the collected information. The intelligent early warning module is used to receive information uploaded by the multi-range signal monitoring module, preprocess the received information, and perform feature mining and multi-range data fusion analysis on historical data to achieve dynamic graded early warning for lithium battery thermal runaway. The graded early warning includes attention level, warning level and danger level. When the lithium battery only experiences temperature abnormality, the intelligent early warning module issues a attention level warning. When the concentration of combustible gas in the coal mine exceeds the threshold, the intelligent early warning module issues a warning level warning. When an open flame is present, the intelligent early warning module issues a danger level warning. The emergency response linkage module is used to receive early warning information from the intelligent early warning module and execute corresponding graded emergency response measures according to the grade of the early warning information. The central control module is used to receive user commands and control the multi-signal monitoring module, intelligent early warning module, and emergency response linkage module according to the user commands.

[0009] In one embodiment of the present invention, the multi-signal monitoring module further includes: The internal condition monitoring unit is used to acquire the voltage, current, temperature and SOC data of the lithium battery through the battery management system; An external safety monitoring unit is used to monitor the battery status in real time. The external safety monitoring unit includes a temperature sensor, a smoke sensor, a gas concentration sensor, and a video monitoring device installed on the surface of the battery pack. An environmental condition monitoring unit is used to monitor the environmental condition in real time. The environmental condition monitoring unit includes an environmental temperature sensor, a humidity sensor, and a dust concentration sensor installed inside the charging cabinet, on the top of the charging chamber, and on the side walls.

[0010] In one embodiment of the present invention, the intelligent early warning module includes: The data preprocessing submodule is used to preprocess the monitoring data, and the preprocessing methods include cleaning, noise reduction and normalization. Thermal runaway database, used to store characteristic data of downhole thermal runaway; The feature mining and index construction submodule uses the Apriori association rule mining algorithm to construct a thermal runaway early warning index system based on the feature data of thermal runaway. The multi-source data fusion early warning submodule uses machine learning algorithms to dynamically determine the early warning threshold and monitors the thermal runaway of lithium batteries in real time based on information uploaded by the multi-source signal monitoring module. When there is no abnormal information, the multi-source data fusion early warning submodule outputs a normal signal. When there is abnormal information, the multi-source data fusion early warning submodule outputs a warning at the attention level, alert level, or danger level according to the specific abnormal information.

[0011] In one embodiment of the present invention, the tiered response measures executed by the emergency response linkage module include: In response to the attention-level warning, an audible and visual alarm is activated, and the lithium battery is controlled to reduce current charging. In response to the alert level warning, the lithium battery is disconnected from other devices, and a powerful exhaust system is activated; In response to a hazard warning, the lithium battery is disconnected from other devices, and the intelligent high-voltage pulse fire suppression system is activated. The intelligent high-pressure pulse fire extinguishing device uses a jet injector based on the pulse jet principle and integrates AI machine vision for automatic fire source location and remote spraying of aerogel fire extinguishing agent.

[0012] In one embodiment of the present invention, a display module is further included for displaying the status of underground lithium batteries in coal mines to users in real time and receiving user commands in real time. The display module is connected to the central control module.

[0013] To achieve the above objectives, a second aspect of the present invention proposes an intelligent safety protection method for charging underground power lithium batteries in coal mines, characterized by comprising the following steps: S1 collects battery internal status information, external safety information and environmental status information in real time through multiple types of sensors deployed in charging equipment and the environment; S2, using a machine learning model to include thermal runaway early warning signals at four levels: normal, attention, alert, and danger; when the lithium battery only experiences temperature abnormalities, an attention-level warning is issued; when the concentration of combustible gas in the coal mine exceeds the threshold, an alert-level warning is issued; and when an open flame is present, a danger-level warning is issued. The machine learning model is a trained random forest classification model. S3, automatically trigger the corresponding graded emergency response process according to the warning level.

[0014] In one embodiment of the present invention, the method for obtaining a trained random forest classification model is as follows: A1 removes outliers from the collected multi-source data and uses the Kalman filter algorithm to denoise the collected multi-source data, eliminating sensor noise interference. A2: Based on historical data, thermal runaway characteristic data is obtained, and the Apriori association rule mining algorithm is used to perform association analysis on the historical data to obtain data association rules and construct a thermal runaway early warning index system. A3 uses preprocessed multi-source data and association rules to train a random forest classification model, resulting in a well-trained random forest classification model.

[0015] In one embodiment of the present invention, S3 further includes: S31, when outputting a warning at the attention level, alerts staff through sound and light, and controls the lithium battery to reduce current charging. S32, when an alarm-level warning is issued, disconnects the lithium battery from other devices and starts strong ventilation to reduce the concentration of combustible gas in the environment; S33, when outputting a hazard warning, disconnects the lithium battery from other devices and activates the intelligent high-voltage pulse fire extinguishing device.

[0016] In one embodiment of the present invention, S33 further includes: AI machine vision is used to automatically locate the fire source, and the aerogel extinguishing agent is remotely sprayed by the injector based on the pulse jet principle in the intelligent high-pressure pulse fire extinguishing device.

[0017] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0018] The method, system, and storage medium of this invention enable dynamic graded early warning and precise emergency response linkage for thermal runaway during the charging process of underground power lithium batteries in coal mines, significantly improving the timeliness, accuracy, and reliability of safety protection.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a dynamic classification and early warning method for thermal runaway of lithium batteries in coal mines according to an embodiment of the present invention; Figure 2 This is a structural diagram of a dynamic graded early warning system for thermal runaway of lithium batteries in coal mines according to an embodiment of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The following describes a dynamic graded early warning method for thermal runaway of lithium batteries in coal mines according to an embodiment of the present invention, with reference to the accompanying drawings.

[0024] Example 1 Figure 1 This is a flowchart of a dynamic classification and early warning method for thermal runaway of lithium batteries in coal mines according to an embodiment of the present invention.

[0025] like Figure 1 As shown, the dynamic classification and early warning method for thermal runaway of lithium batteries in underground coal mines includes the following steps: S1 collects real-time battery internal status information, external safety information, and environmental status information through various types of sensors deployed in charging equipment and the environment.

[0026] Specifically, in some implementations, the real-time acquisition of internal battery status information, external safety information, and environmental status information by deploying multiple types of sensors in the charging equipment and environment is the first key step in the "monitoring-early warning-response" closed-loop management of the intelligent safety protection system platform of this invention. This step, based on the collaborative work of a multi-source heterogeneous sensor network, achieves comprehensive, real-time, and high-precision monitoring of the charging process of underground power lithium batteries in coal mines.

[0027] At the technical implementation level, the multi-type sensor network includes a Battery Management System (BMS) interface, a PT100 temperature sensor, an electrochemical gas sensor (for detecting CO, H2, CH4, etc.), an infrared thermal imager, a dust sensor, and an ambient temperature and humidity sensor. The BMS acquires key parameters such as voltage, current, temperature, internal resistance, and state of charge (SOC) of the battery pack or individual cells in real time via a CAN bus or wireless communication methods (such as ZigBee or LoRa). The external safety monitoring unit collects abnormal signals such as smoke and gas concentrations through sensors deployed around the charging equipment. The environmental condition monitoring unit collects data such as ambient temperature, humidity, and dust concentration through sensors located on the top and side walls of the charging chamber to assess the overall environmental impact on battery safety.

[0028] At the application level, this step is suitable for underground charging chambers or fixed charging points in coal mines, especially during the lithium battery charging process of equipment such as trackless rubber-tired vehicles and monorail cranes. The sensor network is optimized according to the chamber structure and equipment layout to ensure coverage of all areas with potential thermal runaway risks. For example, infrared thermal imagers are placed in the corners of the chamber to achieve panoramic scanning of the surface temperature distribution of the charging equipment; AI video monitoring devices work in conjunction with thermal imagers for fire source identification and location.

[0029] The technical advantage of this step lies in the fact that, through the collaborative acquisition of data from multiple types of sensors, the system can achieve multi-dimensional state perception of the lithium battery charging process, providing high-quality, real-time data support for subsequent intelligent early warning and emergency response. Its innovation lies in the fusion of BMS internal data and external environmental data, breaking through the limitations of traditional single-parameter monitoring, significantly improving the accuracy and response speed of thermal runaway identification, and providing a solid data foundation for the safe charging of lithium batteries in coal mines.

[0030] S2 utilizes a machine learning model to include four levels of thermal runaway warning signals: normal, attention, alert, and danger. When the lithium battery only experiences temperature anomalies, an attention-level warning is issued; when the concentration of combustible gas in the coal mine exceeds the threshold, an alert-level warning is issued; and when an open flame is detected, a danger-level warning is issued.

[0031] As one implementation method, the machine learning model used in this invention is a pre-trained random forest classification model.

[0032] The method for obtaining a well-trained random forest classification model is as follows: A1 removes outliers from the collected multi-source data and uses the Kalman filter algorithm to denoise the collected multi-source data, eliminating sensor noise interference. A2: Based on historical data, thermal runaway characteristic data is obtained, and the Apriori association rule mining algorithm is used to perform association analysis on the historical data to obtain data association rules and construct a thermal runaway early warning index system. A3 uses preprocessed multi-source data and association rules to train a random forest classification model, resulting in a well-trained random forest classification model.

[0033] Specifically, in some implementations, the process of cleaning, denoising, and normalizing the collected multi-source data, and constructing a thermal runaway early warning index system based on association rule mining algorithms, is a key technical step for the intelligent early warning module to achieve high-precision thermal runaway identification. This step first involves quality control of the raw data from the multi-source signal monitoring module through a data preprocessing submodule. The data cleaning stage employs outlier removal methods based on statistical thresholds. For example, a 3σ principle is set for time-series data such as voltage and temperature, removing outliers exceeding three standard deviations from the mean to ensure data reliability. The denoising process uses a Kalman filter algorithm, which, through state prediction and observation update mechanisms, effectively suppresses the impact of sensor noise on data characteristics and improves the signal-to-noise ratio.

[0034] Furthermore, the feature mining and index construction submodule employs the Apriori association rule mining algorithm to extract key feature combinations in the thermal runaway evolution process from the preprocessed multidimensional data. For example, by analyzing the slight decrease in voltage after the voltage plateau period, accompanied by a decrease in CO concentration... Rise to The association rules are used to construct physically meaningful early warning indicators. This indicator system provides input features for the subsequent multi-source data fusion early warning submodule, supports model training based on machine learning algorithms such as random forests, and thus realizes early identification and dynamic hierarchical early warning of thermal runaway.

[0035] This step plays a crucial role in the safety protection system for lithium battery charging in underground coal mines. Through data quality assurance and feature mining, it provides high-confidence input data for intelligent early warning, significantly improving the generalization ability and response speed of the early warning model, and providing a scientific basis for subsequent emergency response coordination.

[0036] S3, automatically trigger the corresponding graded emergency response process according to the warning level.

[0037] Further, step S3 includes: S31, when outputting a warning at the attention level, alerts staff through sound and light, and controls the lithium battery to reduce current charging. S32, when an alarm-level warning is issued, disconnects the lithium battery from other devices and starts strong ventilation to reduce the concentration of combustible gas in the environment; S33, when outputting a hazard warning, disconnects the lithium battery from other devices and activates the intelligent high-voltage pulse fire extinguishing device.

[0038] Specifically, in some implementations, the "automatic triggering of corresponding graded emergency response procedures based on the warning level" is the core control logic of the emergency response linkage module in the intelligent security protection system platform of this invention. Its technical implementation is based on a closed-loop linkage between multi-source data fusion analysis results and a preset graded response mechanism. This step receives the warning level signals (including "normal," "attention," "alert," and "danger") output by the intelligent warning module through the central control and display module, and automatically activates the corresponding response units according to the preset emergency response strategy, achieving seamless connection from warning identification to emergency response.

[0039] In terms of specific operation, when the system detects a "Caution" level warning, the emergency response linkage module first sends a warning signal to on-site personnel through audible and visual alarm devices (such as LED warning lights and buzzers), and simultaneously sends a current reduction command to the charging control system to reduce the charging current below the safe threshold to slow down the heat accumulation process. When a "Alert" level warning is triggered, the system further executes a power cut-off operation, cutting off the charging power supply through a relay or circuit breaker to prevent further deterioration of thermal runaway; at the same time, the forced ventilation system is activated to quickly remove any flammable gases (such as CO and H2) that may have accumulated in the charging chamber, reducing the risk of explosion. When the system identifies a "dangerous" level warning, the emergency response linkage module immediately activates the intelligent high-pressure pulse fire extinguishing device. This device automatically locates the fire source through an AI machine vision system (integrating thermal imaging and visible light recognition) and sends the fire extinguishing command to the high-pressure jet controller to achieve remote and precise spraying of aerogel fire extinguishing agent. The spraying distance is not less than 10 meters, the volume is not less than 50 liters, the spraying interval is not more than 3 seconds, and the linkage control response time is less than 3 seconds, ensuring that the fire extinguishing agent can quickly cover the fire source and effectively suppress reignition.

[0040] This step plays a crucial role in the entire system, and its technical value lies in achieving precise and efficient emergency response through a tiered response mechanism. Compared to traditional single-threshold alarms and passive fire suppression methods, this invention significantly improves the system's response speed and effectiveness to thermal runaway events by combining dynamic early warning levels with differentiated response strategies, thereby ensuring the safety of underground coal mine workers and the continuity of equipment operation.

[0041] The dynamic graded early warning method for thermal runaway during underground lithium battery charging in coal mines, as described in this invention, achieves multi-source data fusion and dynamic graded early warning for thermal runaway during the charging process of underground power lithium batteries. This improves the accuracy and timeliness of the early warning, effectively supports graded emergency response, and enhances the safety protection capability of underground lithium battery charging.

[0042] Example 2 Example 1 Figure 2 This is a flowchart of a dynamic classification and early warning method for thermal runaway of lithium batteries in coal mines according to an embodiment of the present invention.

[0043] like Figure 2 As shown, an intelligent safety protection system for charging underground power lithium batteries in coal mines includes: The multi-signal monitoring module is used to acquire internal state information, external safety information, and environmental state information of the lithium battery, and upload the collected information. The intelligent early warning module is used to receive information uploaded by the multi-range signal monitoring module, preprocess the received information, and perform feature mining and multi-range data fusion analysis on historical data to achieve dynamic graded early warning for lithium battery thermal runaway. The graded early warning includes attention level, warning level and danger level. When the lithium battery only experiences temperature abnormality, the intelligent early warning module issues a attention level warning. When the concentration of combustible gas in the coal mine exceeds the threshold, the intelligent early warning module issues a warning level warning. When an open flame is present, the intelligent early warning module issues a danger level warning. The emergency response linkage module is used to receive early warning information from the intelligent early warning module and execute corresponding graded emergency response measures according to the grade of the early warning information. The central control module is used to receive user commands and control the multi-signal monitoring module, intelligent early warning module, and emergency response linkage module according to the user commands.

[0044] Furthermore, the multi-signal monitoring module also includes: The internal condition monitoring unit is used to acquire the voltage, current, temperature and SOC data of the lithium battery through the battery management system; An external safety monitoring unit is used to monitor the battery status in real time. The external safety monitoring unit includes a temperature sensor, a smoke sensor, a gas concentration sensor, and a video monitoring device installed on the surface of the battery pack. An environmental condition monitoring unit is used to monitor the environmental condition in real time. The environmental condition monitoring unit includes an environmental temperature sensor, a humidity sensor, and a dust concentration sensor installed inside the charging cabinet, on the top of the charging chamber, and on the side walls.

[0045] Furthermore, the intelligent early warning module includes: The data preprocessing submodule is used to preprocess the monitoring data, and the preprocessing methods include cleaning, noise reduction and normalization. Thermal runaway database, used to store characteristic data of downhole thermal runaway; The feature mining and index construction submodule uses the Apriori association rule mining algorithm to construct a thermal runaway early warning index system based on the feature data of thermal runaway. The multi-source data fusion early warning submodule uses machine learning algorithms to dynamically determine the early warning threshold and monitors the thermal runaway of lithium batteries in real time based on information uploaded by the multi-source signal monitoring module. When there is no abnormal information, the multi-source data fusion early warning submodule outputs a normal signal. When there is abnormal information, the multi-source data fusion early warning submodule outputs a warning at the attention level, alert level, or danger level according to the specific abnormal information.

[0046] Furthermore, the tiered response measures executed by the emergency response linkage module include: In response to the attention-level warning, an audible and visual alarm is activated, and the lithium battery is controlled to reduce current charging. In response to the alert level warning, the lithium battery is disconnected from other devices, and the chamber's forced ventilation system is activated; In response to a hazard warning, the lithium battery is disconnected from other devices, and the intelligent high-voltage pulse fire suppression system is activated. The intelligent high-pressure pulse fire extinguishing device uses a jet injector based on the pulse jet principle and integrates AI machine vision for automatic fire source location and remote spraying of aerogel fire extinguishing agent.

[0047] As one implementation method, the aerogel fire extinguishing agent is a composite formulation based on silica aerogel and high-efficiency fire extinguishing additives. The key technical parameters of the aerogel fire extinguishing agent include: bulk density ≤80kg / m³, specific surface area ≥600m² / g, porosity ≥95%, and thermal stability ≥1000℃.

[0048] Furthermore, the system also includes a display module for displaying the status of underground lithium batteries to users in real time and receiving user commands in real time. The display module is connected to the central control module.

[0049] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned intelligent safety protection method for charging power lithium batteries in coal mines.

[0050] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0051] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A smart safety protection system for charging underground power lithium batteries in coal mines, characterized in that, include: The multi-signal monitoring module is used to acquire internal state information, external safety information, and environmental state information of the lithium battery, and upload the collected information. The intelligent early warning module is used to receive information uploaded by the multi-range signal monitoring module, preprocess the received information, and perform feature mining and multi-range data fusion analysis on historical data to achieve dynamic graded early warning for lithium battery thermal runaway. The graded early warning includes attention level, warning level and danger level. When the lithium battery only experiences temperature abnormality, the intelligent early warning module issues a attention level warning. When the concentration of combustible gas in the coal mine exceeds the threshold, the intelligent early warning module issues a warning level warning. When an open flame is present, the intelligent early warning module issues a danger level warning. The emergency response linkage module is used to receive early warning information from the intelligent early warning module and execute corresponding graded emergency response measures according to the grade of the early warning information. The central control module is used to receive user commands and control the multi-signal monitoring module, intelligent early warning module, and emergency response linkage module according to the user commands.

2. The system according to claim 1, characterized in that, The multi-signal monitoring module also includes: The internal condition monitoring unit is used to acquire the voltage, current, temperature and SOC data of the lithium battery through the battery management system; An external safety monitoring unit is used to monitor the battery status in real time. The external safety monitoring unit includes a temperature sensor, a smoke sensor, a gas concentration sensor, and a video monitoring device installed on the surface of the battery pack. An environmental condition monitoring unit is used to monitor the environmental condition in real time. The environmental condition monitoring unit includes an environmental temperature sensor, a humidity sensor, and a dust concentration sensor installed inside the charging cabinet, on the top of the charging chamber, and on the side walls.

3. The system according to claim 1, characterized in that, The intelligent early warning module includes: The data preprocessing submodule is used to preprocess the monitoring data, and the preprocessing methods include cleaning, noise reduction and normalization. Thermal runaway database, used to store characteristic data of downhole thermal runaway; The feature mining and index construction submodule uses the Apriori association rule mining algorithm to construct a thermal runaway early warning index system based on the feature data of thermal runaway. The multi-source data fusion early warning submodule uses machine learning algorithms to dynamically determine the early warning threshold and monitors the thermal runaway of lithium batteries in real time based on information uploaded by the multi-source signal monitoring module. When there is no abnormal information, the multi-source data fusion early warning submodule outputs a normal signal. When there is abnormal information, the multi-source data fusion early warning submodule outputs a warning at the attention level, alert level, or danger level according to the specific abnormal information.

4. The system according to claim 1, characterized in that, The tiered response measures executed by the emergency response linkage module include: In response to the attention-level warning, an audible and visual alarm is activated, and the lithium battery is controlled to reduce current charging. In response to the alert level warning, the lithium battery is disconnected from other devices, and a powerful exhaust system is activated; In response to a hazard warning, the lithium battery is disconnected from other devices, and the intelligent high-voltage pulse fire suppression system is activated. The intelligent high-pressure pulse fire extinguishing device uses a jet injector based on the pulse jet principle and integrates AI machine vision for automatic fire source location and remote spraying of aerogel fire extinguishing agent.

5. The system according to claim 1, characterized in that, It also includes a display module for displaying the status of underground lithium batteries to users in real time and receiving user commands in real time, the display module being connected to the central control module.

6. A method for intelligent safety protection of charging power lithium batteries in coal mines, characterized in that, Includes the following steps: S1 collects battery internal status information, external safety information and environmental status information in real time through multiple types of sensors deployed in charging equipment and the environment; S2, using a machine learning model to include thermal runaway early warning signals at four levels: normal, attention, alert, and danger; when the lithium battery only experiences temperature abnormalities, an attention-level warning is issued; when the concentration of combustible gas in the coal mine exceeds the threshold, an alert-level warning is issued; and when an open flame is present, a danger-level warning is issued. The machine learning model is a trained random forest classification model. S3, automatically trigger the corresponding graded emergency response process according to the warning level.

7. The system method according to claim 6, characterized in that, The method for obtaining a well-trained random forest classification model is as follows: A1 removes outliers from the collected multi-source data and uses the Kalman filter algorithm to denoise the collected multi-source data, eliminating sensor noise interference. A2: Based on historical data, thermal runaway characteristic data is obtained, and the Apriori association rule mining algorithm is used to perform association analysis on the historical data to obtain data association rules and construct a thermal runaway early warning index system. A3 uses preprocessed multi-source data and association rules to train a random forest classification model, resulting in a well-trained random forest classification model.

8. The system method according to claim 6, characterized in that, S3 further includes: S31, when outputting a warning at the attention level, alerts staff through sound and light, and controls the lithium battery to reduce current charging. S32, when an alarm-level warning is issued, disconnects the lithium battery from other devices and starts strong ventilation to reduce the concentration of combustible gas in the environment; S33, when outputting a hazard warning, disconnects the lithium battery from other devices and activates the intelligent high-voltage pulse fire extinguishing device.

9. The system method according to claim 8, characterized in that, S33 further includes: AI machine vision is used to automatically locate the fire source, and the aerogel extinguishing agent is remotely sprayed by the injector based on the pulse jet principle in the intelligent high-pressure pulse fire extinguishing device.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of any one of claims 6-9.

Images