Mine fire data acquisition and monitoring system

Abstract

The invention provides a mine fire data acquisition and monitoring system, which comprises a data acquisition module, a data transmission module, a central processing module, a monitoring and alarming module and a power supply module, and is characterized in that the data acquisition module is used for acquiring mine environment parameters; the data transmission module is used for transmitting mine environment parameters; the central processing module performs fire risk assessment and trend prediction based on a random forest algorithm; the monitoring and alarming module is used for monitoring mine environment parameters in real time and giving an alarm in time when a fire risk assessment result triggers early warning; and the power supply module is used for providing stable power supply for the system. According to the mine fire data acquisition and monitoring system, mine environment data is acquired through the sensor network, fire risk assessment is performed by using a random forest algorithm, early warning is given out in time through the monitoring and alarm module, and intelligent monitoring and emergency response of mine fire are realized.

Description

Technical Field

[0001] The invention belongs to the field of mine safety, and in particular relates to a mine fire data acquisition and monitoring system. Background Art

[0002] Mine fire is one of the most serious disasters in the coal mine production process. It is highly concealed, has a low probability of occurrence, but is extremely destructive. Once a fire occurs, it not only threatens the lives of miners, but also causes damage to mine equipment, production stagnation, and even secondary disasters such as gas explosions. According to statistics, mine fire accidents account for more than 30% of major coal mine accidents. The main reasons for their frequent occurrence include complex mine environments, diverse gas types, and flammable and explosive gases.

[0003] At present, the traditional method of monitoring mine fires mainly relies on manual inspections and single sensor detection, which is not only inefficient but also has great limitations. Manual inspections rely on human experience and judgment and are prone to missed detections, while single sensors cannot accurately identify fire precursors due to the inability to achieve comprehensive data analysis, which often leads to delayed warnings. In addition, the harsh environmental conditions in mines, such as high temperature, high humidity, and high dust concentration, can easily interfere with the normal operation of sensors and affect the reliability of the monitoring system. Therefore, there is an urgent need for a more intelligent, automated, and reliable mine fire monitoring system.

[0004] In recent years, with the rapid development of Internet of Things technology, big data analysis and artificial intelligence technology, fire monitoring methods based on multi-sensor data fusion have gradually emerged. Using a variety of sensors (such as temperature, gas concentration, smoke, etc.) to collect data, combined with advanced machine learning algorithms (such as random forests), dynamic monitoring, trend prediction and risk assessment of fire risks can be achieved. Compared with traditional methods, this technology can significantly improve the accuracy and real-time performance of fire monitoring, while supporting system automated emergency response. However, the fire monitoring functions of most systems are currently limited to data collection and simple threshold alarms, and the ability to identify and assess fire precursors in complex environments is insufficient, making it difficult to meet the actual needs of mine production.

[0005] Therefore, researching and developing a mine fire monitoring system based on multi-sensor data acquisition, reliable data transmission and machine learning algorithms can not only solve the defects of existing technologies, but also provide strong technical support for coal mine safety production. Summary of the invention

[0006] The present invention aims to provide a mine fire data acquisition and monitoring system, which realizes real-time acquisition, analysis and monitoring of mine fire parameters by integrating multiple sensor devices, and improves the early warning and response capabilities of mine fires.

[0007] The technical solution of the present invention is as follows:

[0008] A mine fire data acquisition and monitoring system, including a data acquisition module, a data transmission module, a central processing module and a monitoring and alarm module.

[0009] The data acquisition module is used to collect mine environment parameters;

[0010] The data transmission module is responsible for the transmission of mine environment parameters;

[0011] The central processing module performs fire risk assessment and trend prediction based on the random forest algorithm;

[0012] The monitoring and alarm module is used to monitor the mine environmental parameters in real time and issue an alarm in time when the fire risk assessment results trigger an early warning.

[0013] Furthermore, the data acquisition module includes a sensor layer, an abnormality detection unit and a data calibration unit.

[0014] The sensor layer includes a temperature sensor, a CO sensor, a CH4 sensor, an O2 sensor and a smoke sensor.

[0015] The anomaly detection unit is used to identify unnatural and violent fluctuations and signal interruptions in a short period of time, and to remove abnormal data;

[0016] The data calibration unit periodically calibrates the sensor based on preset reference environmental parameters.

[0017] Furthermore, the data transmission module includes a communication unit and a routing management unit.

[0018] The communication unit is used to transmit the mine environment parameters in the data acquisition module to the central processing module through the LoRa technology;

[0019] The routing management unit is used to optimize and manage the data transmission path.

[0020] Furthermore, the central processing module includes a data processing unit, a fire risk assessment and prediction unit, and a decision and instruction output unit.

[0021] The data processing unit receives, decodes and pre-processes the mine environment data from the data acquisition module;

[0022] The fire risk assessment unit is used to perform fire risk analysis on the mine environment data, assess the fire risk level, and predict future fire trends;

[0023] The decision and instruction output unit is used to generate corresponding emergency instructions or warning information according to the risk assessment results and prediction information.

[0024] Furthermore, the fire risk levels include:

[0025] When the fire probability output by the fire risk assessment and prediction unit is less than 10%, it is a low risk. At this time, no alarm is required, only the status is recorded.

[0026] When the fire probability output by the fire risk assessment and prediction unit is 10%-50%, it is a medium risk and triggers an early warning signal.

[0027] When the fire probability output by the fire risk assessment and prediction unit is >50%, it is a high risk, and an alarm is immediately issued and emergency measures are implemented.

[0028] Furthermore, the monitoring and alarm module includes a real-time monitoring unit, a fire warning and alarm unit, and an emergency response and execution unit.

[0029] The real-time monitoring unit continuously monitors various environmental parameters of the mine;

[0030] The fire warning and alarm unit is used to trigger alarm information according to the fire risk assessment result;

[0031] The emergency response and execution unit is used to automatically trigger relevant emergency response measures and control the emergency equipment of the mine according to the fire risk assessment results and alarm information.

[0032] Furthermore, the method of triggering the alarm information includes sound and light alarm, SMS notification and APP push.

[0033] Furthermore, the system also includes a power supply module,

[0034] The power supply module includes a main power supply and a backup power supply. The main power supply is a mine-specific power supply, and the backup power supply is a 50Ah lithium battery pack, which supports the system to run continuously for 48 hours in the event of a power outage and automatically switches to charging mode after the main power supply is restored.

[0035] Compared with the prior art, the present invention has the following advantages:

[0036] The mine fire data acquisition and monitoring system of the present invention collects mine environment data through a sensor network, uses a random forest algorithm to evaluate fire risks, and issues early warnings in a timely manner through a monitoring and alarm module, thereby realizing intelligent monitoring and emergency response of mine fires. The system design focuses on the accuracy of data acquisition, the stability of transmission, the reliability of evaluation, and the timeliness of alarms, fully ensuring the safety and efficiency of mine operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings generally illustrate various embodiments by way of example and not limitation, and together with the description and claims, serve to illustrate the embodiments of the invention. Where appropriate, the same reference numerals are used throughout the drawings to refer to the same or similar parts. Such embodiments are illustrative and are not intended to be exhaustive or exclusive embodiments of the present apparatus or method.

[0038] Figure 1 This is a structural diagram of the mine fire data acquisition and monitoring system.

[0039] Figure 2 Data analysis flow chart for the central processing module. DETAILED DESCRIPTION

[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0041] like Figure 1-Figure 2 As shown, the embodiment of the present invention provides a mine fire data acquisition and monitoring system, including a data acquisition module, a data transmission module, a central processing module, a monitoring and alarm module, and a power supply module. The system realizes real-time monitoring, evaluation, and alarm of mine fire risks through the collaborative work of multiple modules to ensure mine safety.

[0042] 1. The data acquisition module consists of a sensor layer, anomaly detection unit and data calibration unit, and is responsible for the collection, monitoring and preprocessing of key environmental parameters inside the mine.

[0043] The sensor layer includes:

[0044] Temperature sensor: range -40℃ to 150℃, accuracy ±0.5℃, used to detect changes in temperature inside the mine, arranged in the main ventilation ducts and key operating areas of the mine.

[0045] CO sensor (electrochemical type): detection range 0-1000ppm, accuracy ±3%, used to monitor carbon monoxide concentration in mines. CO concentration is an important feature of the early stage of mine fire.

[0046] CH4 sensor (catalytic combustion type): detection range 0-5% (volume ratio), accuracy ±2%, used to detect methane concentration, arranged in coal seams and areas where gas is prone to accumulate.

[0047] O2 sensor (zirconia type): detection range 0-25%, accuracy ±0.2%, used to monitor oxygen concentration in mines and help determine oxygen consumption during fires.

[0048] Smoke sensor: Based on the photoelectric principle, it detects the concentration of smoke particles and is used for smoke warning of early fires.

[0049] The anomaly detection unit includes:

[0050] Adaptive anomaly detection algorithms are used to identify data with unnatural and drastic fluctuations in a short period of time (such as an instantaneous temperature rise of more than 20°C or a sharp increase of 50ppm in CO concentration in a short period of time).

[0051] When the sensor signal is interrupted or experiences abnormal fluctuations, the system discards the data and records the abnormal event in the log.

[0052] The data calibration unit includes:

[0053] The sensor is calibrated every 48 hours according to preset baseline environmental parameters.

[0054] The calibration process includes zero drift adjustment and full-scale range correction to ensure the accuracy of the collected data.

[0055] After the calibration is completed, the data calibration unit stores the calibration log in the internal memory and reports it to the central processing module.

[0056] 2. The data transmission module consists of a communication unit and a routing management unit, and is responsible for efficiently and stably transmitting the data collected by the data acquisition module to the central processing module.

[0057] The communication unit adopts LoRa communication technology, the communication frequency band is 868MHz or 915MHz, and the maximum transmission distance can reach 10 kilometers.

[0058] It supports point-to-multipoint transmission mode, ensuring that data from multiple sensor nodes can be transmitted to the central processing module in parallel. It has a low power consumption mode with a transmission power of ≤100mW.

[0059] The routing management unit is used to dynamically optimize the data transmission path and reduce transmission delay.

[0060] Implement multi-hop routing function to ensure that data can bypass faulty nodes in the event of transmission path interruption or node failure, thus ensuring the continuity and stability of data transmission.

[0061] Monitor the status of the transmission path in real time and load balance the path to avoid data congestion.

[0062] 3. The central processing module consists of a data processing unit, a fire risk assessment and prediction unit, and a decision and command output unit. It completes fire risk assessment and trend prediction based on the random forest algorithm, and generates early warning or emergency instructions.

[0063] Data processing unit:

[0064] Responsible for receiving environmental parameters from the data transmission module and decoding and preprocessing the data.

[0065] The Kalman filter algorithm is used to denoise the data and remove interference signals.

[0066] Data standardization and normalization: Convert data of different dimensions such as temperature, gas concentration, smoke concentration, etc. into standardized values ​​between 0 and 1 to provide high-quality input for the evaluation algorithm.

[0067] Fire Risk Assessment and Prediction Unit:

[0068] The random forest algorithm is used to conduct fire risk assessment based on data collected by multiple sensors and output the fire risk level.

[0069] The construction of the random forest model includes:

[0070] The training is performed using a historical mine fire dataset. Each piece of data includes environmental parameters collected by sensors and the corresponding fire risk level (label). The division of historical data labels is determined by expert evaluation or simulation experiments.

[0071] A "forest" is formed by randomly dividing the dataset into multiple subsets and using these subsets to train multiple decision trees.

[0072] Each tree makes its own judgment and finally reaches the final result through a voting mechanism.

[0073] The training data includes: temperature, CO concentration, CH4 concentration, O2 concentration, smoke concentration, gas change rate, etc. The output label is the fire risk level, which is usually divided into three levels: low risk, medium risk and high risk.

[0074] Model building steps:

[0075] Step 1: The data set is randomly divided into multiple training subsets (Bootstrap sampling), and a decision tree is trained for each subset.

[0076] Step 2: For each tree, use a different subset of features to split at each node (feature selection). This can increase the diversity of the model and prevent overfitting.

[0077] Step 3: Each tree determines the fire risk based on the training data set and outputs the prediction result. Each decision tree makes a classification judgment (low risk, medium risk, high risk).

[0078] Step 4: The final prediction result is determined by the voting mechanism of all decision trees, and the risk level is selected by "majority voting".

[0079] Important hyperparameters:

[0080] Number of trees: The number of trees in a random forest (i.e., the size of the forest) has an important impact on the performance of the model. Generally, the more trees there are, the more stable the model is, but it also increases the computational overhead. The optimal number of trees can be determined through cross-validation.

[0081] Maximum depth: Controls the depth of each tree to avoid overfitting caused by excessive tree depth.

[0082] Number of features per tree: The number of features randomly selected by each tree when splitting a node to prevent a single feature from overly influencing the decision result.

[0083] After training with the random forest model, the central processing module uses the model to assess the fire risk of the mine environment data collected in real time. The process of fire risk assessment is divided into the following steps:

[0084] Input real-time data: Get the latest mine environment data from the data acquisition module and input it into the random forest model.

[0085] For example, temperature sensor, CO, CH4, O2, and smoke sensor data.

[0086] The random forest model classifies and outputs fire risks based on real-time data to determine whether the current mine environment is at a low, medium or high risk level.

[0087] Low risk: When the probability of fire is less than 10%, the model output is low risk. At this time, there is no need to trigger an alarm, and the system only records the current status.

[0088] Medium risk: When the probability of fire is between 10% and 50%, the model output is medium risk. At this time, the system will trigger an early warning signal, such as an audible and visual alarm or a text message to notify the mine management personnel.

[0089] High risk: When the probability of fire is greater than 50%, the model output is high risk, the system immediately activates the alarm system and triggers emergency response measures (such as fire extinguishing system, personnel evacuation, etc.).

[0090] Fire trend forecast:

[0091] In addition to assessing current fire risks, the model also uses trend analysis to predict future fire risk trends based on historical data. This trend prediction can help mine managers make emergency decisions in advance.

[0092] The forecast output may include a distribution of fire probabilities over the next few hours or days, helping mine managers take more proactive preventive measures.

[0093] Decision-making and instruction output unit:

[0094] Based on the output of the fire risk assessment and prediction unit, alarm instructions, emergency response instructions and related operational recommendations are generated.

[0095] Send emergency instructions to relevant execution equipment (such as fire extinguishing system, ventilation equipment) through the data transmission module

[0096] 4. The monitoring and alarm module displays environmental parameters in real time, triggers alarm information, and executes corresponding emergency measures based on the output of the central processing module.

[0097] Real-time monitoring unit

[0098] Parameter visualization: The well monitoring system can display temperature, CO concentration, CH4 concentration and other data curves in real time.

[0099] Monitoring threshold setting: allows users to customize alarm thresholds, such as setting the alarm threshold for CH4 concentration to 0.8%.

[0100] Fire warning and alarm unit

[0101] Multi-level alarm: Output different alarm signals according to the fire risk level:

[0102] Low risk: Status logging only.

[0103] Medium risk: Triggers a yellow warning light and sends a text message notification.

[0104] High risk: Triggers the red alarm light and buzzer, and sends an emergency command.

[0105] Alarm mode selection: Users can enable single or combined alarm modes in system settings.

[0106] Emergency Response and Execution Unit

[0107] Automatic linkage: connecting the mine ventilation system and the fire extinguishing system, which can automatically close the ventilation duct or spray fire extinguishing agent according to the risk level.

[0108] Remote control: Emergency instructions are issued through the monitoring center to quickly respond to mine fires.

[0109] Power supply module: Provides stable power supply for the system. Even in the case of power outage in the mine, the system can still work continuously for 48 hours.

[0110] Main power supply: Mine-specific power supply, providing 220V AC.

[0111] Backup power supply: A 50Ah lithium battery pack is used to support the system to operate for 48 hours in the event of a power outage.

[0112] Intelligent switching: When the main power is restored, the system automatically switches to charging mode while ensuring continuous power supply to the system.

[0113] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A mine fire data acquisition and monitoring system, characterized in that: It includes data acquisition module, data transmission module, central processing module and monitoring and alarm module. The data acquisition module is used to collect mine environment parameters; The data transmission module is responsible for the transmission of mine environment parameters; The central processing module performs fire risk assessment and trend prediction based on the random forest algorithm; The monitoring and alarm module is used to monitor the mine environmental parameters in real time and issue an alarm in time when the fire risk assessment results trigger an early warning.

2. The mine fire data acquisition and monitoring system according to claim 1, characterized in that: The data acquisition module includes a sensor layer, an abnormality detection unit and a data calibration unit. The sensor layer includes a temperature sensor, a CO sensor, a CH4 sensor, an O2 sensor and a smoke sensor. The anomaly detection unit is used to identify unnatural and violent fluctuations and signal interruptions in a short period of time, and to remove abnormal data; The data calibration unit periodically calibrates the sensor based on preset reference environmental parameters.

3. The mine fire data acquisition and monitoring system according to claim 1, characterized in that: The data transmission module includes a communication unit and a routing management unit. The communication unit is used to transmit the mine environment parameters in the data acquisition module to the central processing module through the LoRa technology; The routing management unit is used to optimize and manage the data transmission path.

4. The mine fire data acquisition and monitoring system according to claim 1, characterized in that: The central processing module includes a data processing unit, a fire risk assessment and prediction unit, and a decision and instruction output unit. The data processing unit receives, decodes and pre-processes the mine environment data from the data acquisition module; The fire risk assessment unit is used to perform fire risk analysis on the mine environment data, assess the fire risk level, and predict future fire trends; The decision and instruction output unit is used to generate corresponding emergency instructions or warning information according to the risk assessment results and prediction information.

5. The mine fire data acquisition and monitoring system according to claim 4, characterized in that: The fire risk levels include: When the fire probability output by the fire risk assessment and prediction unit is less than 10%, it is a low risk. At this time, no alarm is required, only the status is recorded. When the fire probability output by the fire risk assessment and prediction unit is 10%-50%, it is a medium risk and triggers an early warning signal. When the fire probability output by the fire risk assessment and prediction unit is >50%, it is a high risk, and an alarm is immediately issued and emergency measures are implemented.

6. The mine fire data acquisition and monitoring system according to claim 1, characterized in that: The monitoring and alarm module includes a real-time monitoring unit, a fire warning and alarm unit, and an emergency response and execution unit. The real-time monitoring unit continuously monitors various environmental parameters of the mine; The fire warning and alarm unit is used to trigger alarm information according to the fire risk assessment result; The emergency response and execution unit is used to automatically trigger relevant emergency response measures and control the emergency equipment of the mine according to the fire risk assessment results and alarm information.

7. The mine fire data acquisition and monitoring system according to claim 6, characterized in that: The methods of triggering the alarm information include sound and light alarm, SMS notification and APP push.

8. The mine fire data acquisition and monitoring system according to claim 1, characterized in that: The system also includes a power supply module. The power supply module includes a main power supply and a backup power supply. The main power supply is a mine-specific power supply, and the backup power supply is a 50Ah lithium battery pack, which supports the system to run continuously for 48 hours in the event of a power outage and automatically switches to charging mode after the main power supply is restored.

Images