Coal mine gas concentration monitoring and early warning system

Abstract

The invention discloses a coal mine gas concentration monitoring and early warning system, relates to the technical field of safety early warning, and aims to solve the technical problems that a traditional coal mine gas concentration monitoring and early warning system is insufficient in monitoring precision and not timely in early warning, and the coal mine gas concentration monitoring and early warning system comprises a monitoring module used for monitoring the temperature, humidity, oxygen concentration and gas concentration value of the environment in real time, the temperature, the humidity, the oxygen concentration, the gas concentration value and the position data of the monitoring nodes are transmitted to a data processing module; the self-adaptive compensation module is used for receiving the temperature, humidity, oxygen concentration and gas concentration values which are transmitted by the data processing module and monitored by the monitoring nodes in real time and correcting the temperature, humidity, oxygen concentration and gas concentration values to obtain temperature, humidity, oxygen concentration and gas concentration standard values; and a data processing module. The coal mine underground gas concentration monitoring system has the advantages of accurately and comprehensively performing real-time monitoring and effective early warning on the coal mine underground gas concentration.

Description

Technical Field

[0001] The present invention relates to the technical field of safety early warning, and more particularly to a coal mine gas concentration monitoring and early warning system. Background Art

[0002] Accurate monitoring of gas concentration and timely early warning are crucial in the coal mining process. Currently, traditional coal mine gas concentration monitoring and early warning systems face difficulties in real-time, accurate, and comprehensive monitoring of gas concentration in the complex environment of underground coal mines and effective early warning, making them unable to fully guarantee the safety and reliability of coal mining. Specifically, traditional systems are unable to monitor parameters such as gas concentration with sufficient accuracy when faced with fluctuations in environmental parameters such as temperature and humidity in coal mines, as well as the influence of the sensor's own performance. Furthermore, they lack effective multi-parameter collaborative analysis and dynamic early warning mechanisms, making it difficult to accurately determine the dangerous state of gas concentration and issue corresponding early warning signals in a timely manner. In view of this, we propose a coal mine gas concentration monitoring and early warning system. Summary of the Invention

[0003] The purpose of the present invention is to provide a coal mine gas concentration monitoring and early warning system to solve the technical problems of insufficient monitoring accuracy and untimely early warning in traditional coal mine gas concentration monitoring and early warning systems.

[0004] To solve the above technical problems, the present invention provides the following technical solutions: a coal mine gas concentration monitoring and early warning system, comprising: A monitoring module is used to monitor the temperature, humidity, oxygen concentration and gas concentration of the environment in real time, and transmit the temperature, humidity, oxygen concentration and gas concentration values ​​and the location data of the monitoring node to the data processing module; The adaptive compensation module is used to receive the temperature, humidity, oxygen concentration and gas concentration values ​​monitored by the monitoring node in real time and transmitted by the data processing module, and correct them to obtain the standard values ​​of temperature, humidity, oxygen concentration and gas concentration; The data processing module is used to determine whether there are abnormal data for at least two parameters among the standard values ​​of temperature, humidity, oxygen concentration and gas concentration. If so, it controls the alarm communication module to issue a warning signal and send the warning signal to the receiving end; if not, it further determines whether there is a dangerous gas concentration signal; Alarm communication module, used to transmit early warning signals, danger level signals, abnormal concentration signals or normal gas concentration signals when corresponding conditions are met; The receiving end is used to receive the signal sent by the alarm communication module and take corresponding measures according to the content of the signal.

[0005] Preferably, when the data processing module determines whether there is a gas concentration danger signal, it compares the real-time monitored temperature, humidity, oxygen concentration and gas concentration values ​​with normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values; The method by which the system automatically determines the normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values ​​is as follows: the data processing module uses a big data analysis algorithm that combines time series decomposition with machine learning based on historical monitoring data to decompose the monitoring data into time dimensions of hours, days and months, and extract the change characteristics of temperature, humidity, oxygen concentration and gas concentration at different time scales.

[0006] Preferably, the specific method for the data processing module to judge the gas concentration danger signal is: Determine whether the temperature, humidity, oxygen concentration and gas concentration exceed normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values; Use 3 when judging In principle, when the real-time monitoring value exceeds the range of "mean ± 3 times the standard deviation of the normal range of the corresponding parameter", it is determined that the monitoring value exceeds the normal range; If the standard temperature value, humidity value, oxygen concentration value or gas concentration value exceeds the normal temperature, normal humidity, normal oxygen concentration or normal gas concentration value, the concentration at the time of detection will be a gas concentration danger signal; Otherwise, the product of temperature and humidity is further calculated and compared with the preset value to determine whether there is a sensor data anomaly caused by the combined effect of temperature and humidity; The gas concentration danger signal indicates that the corresponding detection location at the detection moment is a danger zone.

[0007] Preferably, the data processing module counts the number of sensor output data anomalies caused by the combined effects of temperature and humidity, which are gas concentration danger signals; if the counted number is greater than or equal to 2, the concentration at the time of detection is determined to be a gas concentration danger signal; The gas concentration danger signals are divided into three levels: slight danger, moderate danger and severe danger, among which: Slight danger means that only one parameter exceeds the normal range and no temperature and humidity combined abnormality is triggered; Moderate risk refers to the presence of 2-3 parameters exceeding the normal range or triggering a combined abnormality of temperature and humidity; Major danger means that all four parameters exceed the normal range or two moderate danger signals appear consecutively in the same area within a short period of time; Different danger levels correspond to graded warning response strategies, with sound and light warnings used for minor dangers; Moderate danger: activate local ventilation enhancement; Major dangers trigger emergency evacuation orders for the entire area, and at the same time send evacuation route information to all personnel positioning devices underground. The evacuation routes are dynamically generated based on the real-time monitoring of the dangerous conditions in each area.

[0008] Preferably, the adaptive compensation module corrects the temperature, humidity, oxygen concentration and gas concentration values ​​monitored in real time by the monitoring node and received by the data processing module in the following manner: Set up the calibration procedure according to the sensor type. For temperature sensors, use the Steinhart-Hart equation based on platinum resistance for nonlinear compensation. For humidity sensors, a polynomial fitting algorithm is used to establish a compensation model based on the temperature-humidity cross-sensitivity characteristics; For the oxygen concentration sensor, an electrochemical sensor zero drift compensation algorithm is used, and dynamic correction is performed through the reference gas calibration coefficient; For the gas concentration sensor, a composite algorithm of temperature compensation and aging correction of the catalytic combustion sensor is adopted. The temperature compensation is linearly compensated according to the built-in temperature sensor data of the sensor. The calibration procedure is used to correct the original data of temperature, humidity, oxygen concentration and gas concentration to obtain corresponding standard values.

[0009] Preferably, the monitoring module uses an edge computing node to implement local preprocessing of data. The edge computing node has a built-in FPGA chip and an ARM processor collaborative processing architecture. The FPGA chip is responsible for high-speed data acquisition and preliminary processing, and the ARM processor performs complex algorithm calculations to reduce noise and normalize the collected raw data and compress the data volume.

[0010] Preferably, the data processing module adopts a distributed computing architecture, including multiple data processing sub-nodes and a management node. The management node is based on Kubernetes container orchestration technology and dynamically allocates tasks to each sub-node according to the data type and computing task amount.

[0011] Preferably, a multi-parameter time series association model is established, wherein the multi-parameter time series association model has a built-in environment perception module, and the environment perception module is connected to a geological radar, a vibration sensor, and an equipment status monitor via an RS485 bus, and collects in real time geological structure data of the monitoring area, operating parameters of mining equipment, and environmental information of the ventilation system status as model input variables; The collected data is first formatted and verified to ensure its integrity and accuracy before being input into the model; Set up an online learning mechanism to automatically trigger the model update process when the model prediction results deviate from the actual monitoring data; Utilizing the latest environmental information and monitoring data, the stochastic gradient descent algorithm is used to dynamically adjust the model parameters, and the adjustment step size is adaptively adjusted according to the data change rate.

[0012] Preferably, the online learning mechanism adopts a transfer learning algorithm, and when a new scene or new working condition appears, the training model parameters based on existing similar scenes are quickly adjusted; The feature distributions of the source domain and the target domain are aligned through the domain adaptation algorithm, reducing the model training time and data requirements.

[0013] Preferably, the data processing module uses a long short-term memory network combined with an attention mechanism to construct a prediction model to predict the abnormal development trend of gas concentration.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention realizes real-time monitoring, precise correction, multi-parameter collaborative analysis and graded early warning of parameters such as temperature, humidity, oxygen concentration and gas concentration in coal mines by designing a monitoring module, an adaptive compensation module, a data processing module, an alarm communication module and a receiving end. It effectively solves the problems of insufficient monitoring accuracy and untimely early warning in traditional systems, and can more accurately and comprehensively monitor the gas concentration in coal mines in real time and provide effective early warning, greatly improving the accuracy and reliability of gas concentration monitoring and early warning during coal mining, and providing a strong guarantee for coal mine safety production.

[0015] 2. The present invention uses a big data analysis algorithm based on the time series decomposition of historical monitoring data combined with machine learning through the data processing module to automatically determine the normal temperature, humidity, oxygen concentration and gas concentration values, and use 3 The method of comparing the product of temperature and humidity to judge the dangerous signal of gas concentration is adopted, which realizes the accurate judgment and graded early warning of the dangerous state of gas concentration, further solves the problem of inaccurate judgment of the dangerous state of gas concentration by the traditional system, makes the early warning more scientific and reasonable, and can adopt corresponding early warning response strategies according to different danger levels, thereby improving the ability of coal mines to deal with gas hazards.

[0016] 3. The present invention uses an adaptive compensation module to set different correction programs according to the type of sensor, and corrects the original data of temperature, humidity, oxygen concentration and gas concentration, thereby improving the accuracy and reliability of sensor data, further solving the problem that the performance of the sensor itself affects the monitoring accuracy, and providing a more reliable data basis for subsequent data processing and early warning, ensuring that the monitoring and early warning functions of the entire system are more stable and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0018] To facilitate those skilled in the art to understand the technical solution of the present invention, the technical solution of the present invention is further described with reference to the accompanying drawings.

[0019] Example 1, as Figure 1 As shown, the present invention provides a coal mine gas concentration monitoring and early warning system, comprising: The monitoring module is used to monitor the temperature, humidity, oxygen concentration and gas concentration of the environment in real time, and transmit the temperature, humidity, oxygen concentration and gas concentration values ​​and the location data of the monitoring node to the data processing module; The adaptive compensation module is used to receive the temperature, humidity, oxygen concentration and gas concentration values ​​monitored by the monitoring node in real time and transmitted by the data processing module, and correct them to obtain the standard values ​​of temperature, humidity, oxygen concentration and gas concentration; The data processing module is used to determine whether there are abnormal data for at least two parameters among the standard values ​​of temperature, humidity, oxygen concentration and gas concentration. If so, it controls the alarm communication module to issue a warning signal and send the warning signal to the receiving end; if not, it further determines whether there is a dangerous gas concentration signal; Alarm communication module, used to transmit early warning signals, danger level signals, abnormal concentration signals or normal gas concentration signals when corresponding conditions are met; The receiving end is used to receive the signal sent by the alarm communication module and take corresponding measures according to the content of the signal.

[0020] In an embodiment of the present invention, when the data processing module determines whether a gas concentration danger signal exists, it compares the real-time monitored temperature, humidity, oxygen concentration, and gas concentration values ​​with normal temperature, normal humidity, normal oxygen concentration, and normal gas concentration values; The system automatically determines normal temperature, humidity, oxygen concentration, and gas concentration values ​​as follows: The data processing module uses a big data analysis algorithm that combines time series decomposition with machine learning based on the coal mine's historical monitoring data from the past 3-5 years. This algorithm decomposes the monitoring data into hourly, daily, and monthly time dimensions, extracting the temperature, humidity, oxygen concentration, and gas concentration variation characteristics at different time scales. Specifically: For hourly data, analyze its fluctuation range and change rate; For daily data, calculate the daily mean and standard deviation; For monthly data, calculate the monthly trend change slope; Combined with the geological structure data of the monitoring site (including coal seam thickness, rock permeability, etc.), mining progress data (such as mining face advancement distance, mining intensity, etc.), and ventilation system status data (such as wind speed, wind pressure, ventilation volume, etc.), a random forest regression model is trained; The output of each decision tree in the random forest regression model Determined by the following formula: ; in, is the number of decision tree nodes, For nodes The weight of For nodes Input features of The random forest model contains 500 decision trees, and the maximum depth of each decision tree is set to 10. The normal temperature of the corresponding area in the next 72 hours is calculated by averaging the output results of all decision trees. , normal humidity , normal oxygen concentration and normal gas concentration value: ; ; ; ; in, 、 、 and Respectively represent The predicted values ​​of temperature, humidity, oxygen concentration and oxygen concentration output by the decision tree The above values ​​are updated every 6 hours using an incremental learning approach, where only newly collected data is processed and incorporated into the model.

[0021] In an embodiment of the present invention, the specific method for the data processing module to determine the gas concentration danger signal is: Determine whether the temperature, humidity, oxygen concentration and gas concentration exceed normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values; The 3σ principle is used in judgment. When the real-time monitoring value exceeds the range of "mean ± 3 times the standard deviation of the normal range of the corresponding parameter", it is judged that the monitoring value exceeds the normal range and may be abnormal or dangerous; Among them, the standard deviation The calculation is based on the rolling calculation of historical data over the past 7 days, reflecting the degree of data dispersion. The 3σ principle defines a normal fluctuation range in a statistical sense through "mean ± 3 times the standard deviation". Monitoring values ​​outside this range are considered to be low-probability events and are therefore identified as abnormal. Taking temperature data as an example, the calculation formula is: ; Where, Represents the temperature standard deviation, which is used to quantify the degree of dispersion of temperature data. is the number of samples of temperature data in the past 7 days, For the Temperature sample values, is the average temperature data of the past 7 days, that is ; The standard deviation calculated by the above formula can reflect the data fluctuation in real time; If the standard temperature value, humidity value, oxygen concentration value or gas concentration value exceeds the normal temperature, normal humidity, normal oxygen concentration or normal gas concentration value, the concentration at the time of detection will be a gas concentration danger signal; Otherwise, the product of temperature and humidity is further calculated and compared with the preset value to determine whether there is a sensor data anomaly caused by the combined effect of temperature and humidity; The gas concentration danger signal indicates that the corresponding detection location at the detection time is a dangerous area; The preset value is determined based on the sensor type and historical data statistics using the 95% confidence interval method; The specific determination process is to collect 1000 sets of historical data of this type of sensor under different temperature and humidity combinations. , calculate the product data ,calculate The mean and standard deviation : ; ; in, It represents the standard deviation of the product of temperature and humidity, and is used to quantify the degree of dispersion of data under the combined effect of temperature and humidity. Indicates the The product of temperature and humidity in the group historical data is used to analyze the impact of the combined effect of temperature and humidity on sensor data. Indicates the Temperature values ​​in group history data, Indicates the Humidity values ​​in group history data; The default value for: ; in, Indicates the threshold for determining whether the combined effect of temperature and humidity causes sensor data anomalies. If this value is exceeded, an anomaly is considered present. Where 1.645 is the quantile of the 95% confidence interval of the standard normal distribution; If the current calculated value Greater than the preset value , it is determined that there is an abnormality in the sensor data, and then it is determined whether it is a dangerous signal or an abnormal signal of gas concentration.

[0022] In an embodiment of the present invention, the data processing module counts the number of sensor output data anomalies caused by the combined effects of temperature and humidity, which are gas concentration danger signals. If the counted number is greater than or equal to 2, the concentration at the time of detection is determined to be a gas concentration danger signal. Gas concentration danger signals are divided into three levels: slight danger, moderate danger and severe danger, among which: Slight danger means that only one parameter exceeds the normal range and no temperature and humidity combined abnormality is triggered; Moderate risk refers to the presence of 2-3 parameters exceeding the normal range or triggering a combined abnormality of temperature and humidity; Major danger means that all four parameters exceed the normal range or two moderate danger signals appear consecutively in the same area within a short period of time; Different danger levels correspond to graded warning response strategies. Minor dangers use sound and light warnings. The sound intensity of the sound and light alarm is set to 80-90 decibels, and the flash frequency is 2-3 times per second. Moderate risk: Local ventilation enhancement is activated, and the ventilation volume of ventilation equipment is increased to 1.5 times the normal volume; Major dangers trigger emergency evacuation orders for the entire area, and at the same time send evacuation route information to all personnel positioning devices underground. The evacuation routes are dynamically generated based on the real-time monitoring of the dangerous conditions in each area.

[0023] In an embodiment of the present invention, the adaptive compensation module corrects the temperature, humidity, oxygen concentration, and gas concentration values ​​monitored in real time by the monitoring node and received by the data processing module in the following manner: The calibration procedure is set according to the sensor type. For temperature sensors, the Steinhart-Hart equation based on platinum resistance is used for nonlinear compensation. The equation is: ; in, is the temperature value, is the resistance of the platinum resistor, 、 、 are equation parameters, which are obtained by three calibrations at standard temperature environments (0°C, 25°C, and 100°C); And combined with the environmental pressure data, a secondary correction is performed, and the correction formula is: ; in, is the corrected temperature value, To measure the temperature, is the pressure correction coefficient, which is determined to be 0.0001 / Pa through experiments. is the pressure deviation value; For the humidity sensor, a polynomial fitting algorithm is used to establish a compensation model based on the temperature-humidity cross-sensitivity characteristics. The model is a quadratic polynomial: ; in, is the corrected humidity value, is the temperature value, To measure humidity, 、 、 It is obtained by collecting humidity data at different temperatures and performing least square fitting; When fitting by the least squares method, the objective function for: ; in, Indicates the The actual humidity value in the group data, Indicates the The measured value of the humidity sensor in the group data; By minimizing the objective function Solution 、 、 ; For the oxygen concentration sensor, the electrochemical sensor zero drift compensation algorithm is used, and dynamic correction is performed through the reference gas calibration coefficient. Zero point calibration is performed once every 24 hours using standard oxygen concentration gas (20.9%), and the calibration coefficient is updated after calibration. , the correction formula is: ; in, is the corrected oxygen concentration value, To measure the oxygen concentration, is the zero drift value; For the gas concentration sensor, a composite algorithm of temperature compensation and aging correction of the catalytic combustion sensor is adopted. The temperature compensation is linearly compensated according to the built-in temperature sensor data of the sensor. The formula is: ; in, is the gas concentration value after temperature compensation, To measure the gas concentration, is the temperature compensation coefficient, is the reference temperature value; Aging correction by recording the sensor usage time and cumulative number of detections , establish the aging curve model for correction, and the aging correction formula is: ; in, is the gas concentration value after aging correction, is the aging correction factor, Design the usage time for the sensor, Design the cumulative number of detections for the sensor; Use the calibration program to correct the original data of temperature, humidity, oxygen concentration and gas concentration to obtain the corresponding standard values; The calibration procedure is carried out by the staff according to the dust concentration, electromagnetic interference intensity and other factors in the gas monitoring environment to fine-tune the parameters. When the dust concentration exceeds 50mg / m 3 When , the humidity sensor compensation model parameter a increases by 10%.

[0024] In an embodiment of the present invention, the monitoring module uses edge computing nodes to implement local preprocessing of data. The edge computing nodes have a built-in FPGA chip and an ARM processor collaborative processing architecture. The FPGA chip is responsible for high-speed data acquisition and preliminary processing, and the ARM processor performs complex algorithm calculations. Perform noise reduction, normalization and data compression on the collected raw data; The denoising algorithm uses the wavelet threshold denoising algorithm, the db4 wavelet basis function is selected, the number of decomposition layers is set to 3, and the threshold is processed using the soft threshold function; For noisy signals , after wavelet transform, we get the wavelet coefficients , the soft threshold processing formula is: ; in, is the wavelet coefficient after threshold processing, is a symbolic function, is the threshold, and the general threshold formula calculate, is the noise standard deviation, which can be estimated by the median method , is the signal length; Normalization uses the maximum-minimum normalization method to normalize the data to the [0,1] interval. The formula is: ; in, is the normalized data, is the original data, and are the minimum and maximum values ​​in the original data, respectively.

[0025] The data is compressed using the LZ77 lossless compression algorithm. During the compression process, the sliding window size is set to 4096 bytes and the forward search buffer size is set to 256 bytes. This compresses the data volume to less than 30% of the original data volume, and only transmits key data features to the data processing module. The system uses 5G-U or industrial-grade wireless mesh networks to build underground data transmission channels. It allocates dedicated transmission channels for monitoring data through network slicing technology, and sets the bandwidth to no less than 100Mbps. And set up three layers of redundant transmission paths, including the main path, backup path and emergency path; Real-time monitoring of network transmission status. When network delay exceeds 50ms or packet loss rate exceeds 5%, it automatically switches to redundant paths and starts a fast retransmission mechanism based on the TCP protocol. When switching paths, a weighted round-robin algorithm is used to select the best backup path. The calculation formula is: ; in, For path The historical delay For path Available bandwidth, is the total number of backup paths.

[0026] In an embodiment of the present invention, the data processing module adopts a distributed computing architecture, including multiple data processing sub-nodes and a management node. The management node is based on Kubernetes container orchestration technology and dynamically allocates tasks to each sub-node according to the data type and computing task amount; The specific allocation strategy is to prioritize the allocation of computing tasks with high real-time requirements to sub-nodes with lower load and higher performance; For batch processing tasks, task sharding is used to evenly distribute them to multiple child nodes; Each sub-node uses the Spark Streaming framework for parallel processing. Data processing sub-nodes interact with each other through the ZeroMQ message queue, and the results are fed back to the management node for integrated analysis. Using stream computing technology and based on the Flink real-time computing engine, it performs duration series correlation calculations on real-time data streams. The data is segmented using a sliding window algorithm and the similarity of time series with different parameters is calculated using a dynamic time warping algorithm. When calculating similarity, Euclidean distance is used as the metric, and in order to reduce the amount of calculation, the time series is downsampled, and the sampling interval is dynamically adjusted according to the frequency of data changes; Suppose there are two time series and , after downsampling, and ,DTW algorithm seeks the best time series alignment path so that the sum of ,the aligned Euclidean distances is minimized; Define the local distance matrix ,in , the cumulative distance matrix The calculation is as follows: ; in, , ,and , ; final That is, the DTW distance between two downsampled time series. The smaller the distance, the higher the similarity between the two time series. At the same time, a watermark mechanism is set to process disordered data to ensure the accuracy of data processing, and the watermark timestamp Maximum delay according to event time in the data stream OK, the formula is ,in The current processing time.

[0027] In an embodiment of the present invention, a multi-parameter time series association model is established, and the multi-parameter time series association model has a built-in environmental perception module. The environmental perception module is connected to a geological radar, a vibration sensor, and an equipment status monitor via an RS485 bus, and collects in real time geological structure data of the monitoring area, operating parameters of mining equipment, and environmental information of the ventilation system status as model input variables; The collected data is first formatted and verified to ensure its integrity and accuracy before being input into the model; Set up an online learning mechanism to automatically trigger the model update process when the model prediction results deviate from the actual monitoring data; Using the latest environmental information and monitoring data, the stochastic gradient descent algorithm is used to dynamically adjust the model parameters, and the adjustment step size is adaptively adjusted according to the data change rate; The specific adjustment method is to set the current data change rate to , the step size is , then the step size adjustment formula is: ; in, is the adjusted step length; In the stochastic gradient descent algorithm, for the model parameters , and its update formula is ,in is the loss function About parameters The gradient, For input data, is the true label.

[0028] In an embodiment of the present invention, the online learning mechanism uses a transfer learning algorithm. When a new scenario or new working condition appears, the parameters of the training model based on existing similar scenarios are quickly adjusted; The domain adaptation algorithm aligns the feature distributions of the source domain (existing scenarios) and the target domain (new scenarios), reducing model training time and data requirements. When aligning feature distributions, the maximum mean difference (MMD) is used as a metric to achieve distribution alignment by minimizing the MMD values ​​of source and target domain features. Assume that the source domain feature set is , the target domain feature set is , the kernel function is , then the MMD calculation formula is: ; By optimizing the objective function To adjust the model parameters ,in is the loss function on the source domain, is the balance coefficient; Even when the amount of target domain data is less than 20% of the original training data, the model prediction accuracy can still be guaranteed to be no less than 85%, improving the model's generalization ability in different environments.

[0029] In an embodiment of the present invention, the data processing module uses a long short-term memory network combined with an attention mechanism to construct a prediction model to predict the abnormal development trend of gas concentration; The LSTM network contains three hidden layers, with 128 neurons in each layer.

[0030] In the LSTM unit, the forget gate , input gate , output gate and cell status The update formula is as follows: ; ; ; ; ; ; in, is the Sigmoid function, is the hyperbolic tangent function, , , , is the weight matrix, , , , is the bias vector, is element-wise multiplication, Indicates hiding the previous state and the current input splicing; The attention mechanism highlights the impact of key time steps on the prediction results by calculating the attention weights of different time steps; Set the attention weight , calculated as: ; ; in, is the weight matrix, is the bias vector; The final output y is: ; If it is predicted that the gas concentration will exceed 80% of the upper limit of the normal range, the advance control alarm module will send a pre-warning signal to the communication module. The pre-warning signal contains information such as the predicted concentration value, expected time of exceeding the standard, potential danger level, etc., to remind staff to make preventive preparations in advance. At the same time, the system will automatically generate preventive suggestions, such as increasing the ventilation frequency of the area, arranging personnel to conduct on-site inspections, etc. The preventive suggestions are generated based on effective handling measures for similar situations in historical data.

[0031] The embodiments disclosed in the present invention are preferred embodiments, but are not limited to them. Ordinary technicians in this field can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. As long as they do not deviate from the spirit of the present invention, they are all within the scope of protection of the present invention.

Claims

1. A coal mine gas concentration monitoring and early warning system, characterized in that: include: A monitoring module is used to monitor the temperature, humidity, oxygen concentration and gas concentration of the environment in real time, and transmit the temperature, humidity, oxygen concentration and gas concentration values ​​and the location data of the monitoring node to the data processing module; The adaptive compensation module is used to receive the temperature, humidity, oxygen concentration and gas concentration values ​​monitored by the monitoring node in real time and transmitted by the data processing module, and correct them to obtain the standard values ​​of temperature, humidity, oxygen concentration and gas concentration; The data processing module is used to determine whether there are abnormal data for at least two parameters among the standard values ​​of temperature, humidity, oxygen concentration and gas concentration. If so, it controls the alarm communication module to issue a warning signal and send the warning signal to the receiving end; if not, it further determines whether there is a dangerous gas concentration signal; Alarm communication module, used to transmit early warning signals, danger level signals, abnormal concentration signals or normal gas concentration signals when corresponding conditions are met; The receiving end is used to receive the signal sent by the alarm communication module and take corresponding measures according to the content of the signal.

2. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: When the data processing module determines whether there is a gas concentration danger signal, it compares the real-time monitored temperature, humidity, oxygen concentration and gas concentration values ​​with normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values; The method by which the system automatically determines the normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values ​​is as follows: the data processing module uses a big data analysis algorithm that combines time series decomposition with machine learning based on historical monitoring data to decompose the monitoring data into time dimensions of hours, days and months, and extract the change characteristics of temperature, humidity, oxygen concentration and gas concentration at different time scales.

3. A coal mine gas concentration monitoring and early warning system according to claim 2, characterized in that: The specific method for the data processing module to judge the gas concentration danger signal is: Determine whether the temperature, humidity, oxygen concentration and gas concentration exceed normal temperature, normal humidity, normal oxygen concentration and normal gas concentration values; The 3σ principle is used for judgment. When the real-time monitoring value exceeds the range of "mean ± 3 times the standard deviation of the normal range of the corresponding parameter", it is determined that the monitoring value exceeds the normal range; If the standard temperature value, humidity value, oxygen concentration value or gas concentration value exceeds the normal temperature, normal humidity, normal oxygen concentration or normal gas concentration value, the concentration at the time of detection will be a gas concentration danger signal; Otherwise, the product of temperature and humidity is further calculated and compared with the preset value to determine whether there is a sensor data anomaly caused by the combined effect of temperature and humidity; The gas concentration danger signal indicates that the corresponding detection location at the detection moment is a danger zone.

4. A coal mine gas concentration monitoring and early warning system according to claim 3, characterized in that: The data processing module counts the number of sensor output data anomalies caused by the combined effects of temperature and humidity, which are gas concentration danger signals. If the counted number is greater than or equal to 2, the concentration at the time of detection is determined to be a gas concentration danger signal. The gas concentration danger signals are divided into three levels: slight danger, moderate danger and severe danger, among which: Slight danger means that only one parameter exceeds the normal range and no temperature and humidity combined abnormality is triggered; Moderate risk refers to the presence of 2-3 parameters exceeding the normal range or triggering a combined abnormality of temperature and humidity; Major danger means that all four parameters exceed the normal range or two moderate danger signals appear consecutively in the same area within a short period of time; Different danger levels correspond to graded warning response strategies, with sound and light warnings used for minor dangers; Moderate danger: activate local ventilation enhancement; Major dangers trigger emergency evacuation orders for the entire area, and at the same time send evacuation route information to all personnel positioning devices underground. The evacuation routes are dynamically generated based on the real-time monitoring of the dangerous conditions in each area.

5. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: The method by which the adaptive compensation module corrects the temperature, humidity, oxygen concentration and gas concentration values ​​monitored in real time by the monitoring node received by the data processing module is as follows: Set up the calibration procedure according to the sensor type. For temperature sensors, use the Steinhart-Hart equation based on platinum resistance for nonlinear compensation. For humidity sensors, a polynomial fitting algorithm is used to establish a compensation model based on the temperature-humidity cross-sensitivity characteristics; For the oxygen concentration sensor, an electrochemical sensor zero drift compensation algorithm is used, and dynamic correction is performed through the reference gas calibration coefficient; For the gas concentration sensor, a composite algorithm of temperature compensation and aging correction of the catalytic combustion sensor is adopted. The temperature compensation is linearly compensated according to the built-in temperature sensor data of the sensor. The calibration procedure is used to correct the original data of temperature, humidity, oxygen concentration and gas concentration to obtain corresponding standard values.

6. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: The monitoring module uses edge computing nodes to implement local preprocessing of data. The edge computing nodes have a built-in FPGA chip and ARM processor collaborative processing architecture. The FPGA chip is responsible for high-speed data acquisition and preliminary processing, and the ARM processor performs complex algorithm calculations to reduce noise and normalize the collected raw data and compress the data volume.

7. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: The data processing module adopts a distributed computing architecture, including multiple data processing sub-nodes and a management node. The management node is based on Kubernetes container orchestration technology and dynamically allocates tasks to each sub-node according to data type and computing task volume.

8. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: Establish a multi-parameter time series association model, which has a built-in environmental perception module. The environmental perception module is connected to a geological radar, a vibration sensor, and an equipment status monitor via an RS485 bus, and collects real-time geological structure data of the monitoring area, mining equipment operating parameters, and environmental information of the ventilation system status as model input variables; The collected data is first formatted and verified to ensure its integrity and accuracy before being input into the model; Set up an online learning mechanism to automatically trigger the model update process when the model prediction results deviate from the actual monitoring data; Utilizing the latest environmental information and monitoring data, the stochastic gradient descent algorithm is used to dynamically adjust the model parameters, and the adjustment step size is adaptively adjusted according to the data change rate.

9. A coal mine gas concentration monitoring and early warning system according to claim 8, characterized in that: The online learning mechanism uses a transfer learning algorithm to quickly adjust the parameters of the training model based on existing similar scenarios when new scenarios or new working conditions appear; The feature distributions of the source domain and the target domain are aligned through the domain adaptation algorithm, reducing the model training time and data requirements.

10. A coal mine gas concentration monitoring and early warning system according to claim 1, characterized in that: The data processing module uses a long short-term memory network combined with an attention mechanism to build a prediction model to predict the abnormal development trend of gas concentration.

Images