CN116311751A

CN116311751A - Underground coal mine use fire prevention and extinguishment electric control system

Info

Publication number: CN116311751A
Application number: CN202310547989.7A
Authority: CN
Inventors: 王克华; 李寅; 康凯; 杨振川
Original assignee: Shaanxi Kailai Electromechanical Equipment Manufacturing Co ltd
Current assignee: Shaanxi Kailai Electromechanical Equipment Manufacturing Co ltd
Priority date: 2023-05-16
Filing date: 2023-05-16
Publication date: 2023-06-23
Anticipated expiration: 2043-05-16
Also published as: CN116311751B

Abstract

The utility model relates to an intelligent control system technical field, it specifically discloses a colliery is in pit uses fire prevention electrical system, adopts the artificial intelligence monitoring technology based on degree of depth study, gathers temperature value, smog concentration value and the carbon monoxide concentration value of fire monitoring point in the colliery in pit through many sensors, further judges whether fire monitoring point takes place the conflagration through the time sequence change characteristic that fuses between the three to after the judgement result is the conflagration, cuts off the power supply of fire monitoring point, has improved the accuracy of fire judgement.

Description

Underground coal mine use fire prevention and extinguishment electric control system

Technical Field

The application relates to the technical field of intelligent control systems, and more particularly, to an underground coal mine fire prevention and extinguishing electric control system.

Background

Fire in the coal mine is a serious potential safety hazard, and not only can cause casualties, but also can cause equipment damage and resource waste. In order to effectively prevent and control underground fires, it is necessary to provide fire monitoring points underground, collect data of the fire monitoring points through sensors to monitor the fires, and start a fire control scheme after the fires are found.

The China patent with application number 201610919731.5 discloses a mine explosion monitoring alarm and control system. The system mainly comprises an information processing server, an alarm device, a communication network, explosion suppression and fire extinguishing equipment, a gas concentration monitoring device and various environment monitoring devices; the system can monitor the change of various data such as smog, temperature and the like caused by gas explosion, monitor the concentration of the marking gas through the gas concentration monitoring device, alarm the mine explosion according to the data obtained by monitoring, automatically suppress explosion and extinguish the fire, reduce casualties and reduce the loss caused by gas explosion. The system overcomes the defects of slow response, high false alarm rate, high missing report rate and the like of the methods of gas monitoring and the like adopted by the traditional explosion monitoring, greatly improves the alarm accuracy and provides important guarantee for the coal mine safety production.

The Chinese patent with the application number of 202120664451.0 discloses an intelligent fire extinguishing and monitoring system based on single chip microcomputer information linkage, which comprises a central control and processor, a system monitoring server, a fire extinguisher terminal, an alarm module, an information inquiry module, an instruction sending module, a positioning module, a sensor module, a GPRS wireless communication module and an electromagnetic valve, wherein the electromagnetic valve, the positioning module and the sensor module are arranged on the fire extinguisher terminal, and the fire extinguisher terminal is arranged at the indoor top; the system monitoring server is respectively connected with the central control and processing module and the electromagnetic valve through the GPRS wireless communication module, the alarm module, the information inquiry module and the instruction sending module are connected with the central control and processing module, and the positioning module and the sensor module are connected with the central control and processing module through wireless signals. The system can monitor in real time and timely feed back dangerous points to timely and intelligently extinguish fire, and finally realizes four-fold linkage of real-time monitoring, early warning, timely warning and intelligent fire extinguishment.

However, in the existing fire monitoring scheme, only one type of sensor (for example, a temperature sensor) or a few types of sensors are generally used for fire monitoring, but the downhole environment is a complex and variable working condition environment, and the fire monitoring performed by only using the collected single-dimensional data may cause missed judgment and erroneous judgment.

Thus, an optimized coal mine downhole use fire protection scheme is desired.

Disclosure of Invention

The present application has been made in order to solve the above technical problems. The embodiment of the application provides a colliery is in pit and is used fire prevention and extinguishing electrical system, and it is through adopting the artificial intelligence monitoring technology based on degree of depth study, gathers temperature value, smog concentration value and the carbon monoxide concentration value of colliery fire monitoring point in the pit through many sensors, further judges whether fire disaster is happened to the fire monitoring point through the time sequence change characteristic that fuses between the three to after the judgement result is the conflagration, cut off the power supply of fire monitoring point.

According to one aspect of the present application, there is provided a fire prevention and extinguishing electrical control system for use downhole in a coal mine, comprising:

the system comprises a multi-sensor data acquisition module, a sensor data acquisition module and a data processing module, wherein the multi-sensor data acquisition module is used for acquiring sensor data of a plurality of preset time points in a preset time period acquired by sensors of fire monitoring points deployed in underground environment, and the sensor data comprises a temperature value, a smoke concentration value and a carbon monoxide concentration value;

The vectorization module is used for respectively arranging the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of a plurality of preset time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to the time dimension;

the time sequence feature extraction module is used for respectively passing the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through the multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector;

the characteristic fusion module is used for fusing the temperature time sequence characteristic vector, the smoke concentration value time sequence characteristic vector and the carbon monoxide concentration value time sequence characteristic vector to obtain a classification characteristic vector;

the regularization module is used for regularizing the characteristic distribution of the classification characteristic vector to obtain an optimized classification characteristic vector;

and the control result generation module is used for enabling the optimized classification feature vector to pass through the classifier to obtain a classification result, and the classification result is used for indicating whether the power supply of the fire monitoring point is cut off or not.

In the underground coal mine fire prevention and extinguishing electric control system, the multi-scale neighborhood feature extraction module comprises a first convolution layer and a second convolution layer which are parallel, and a multi-scale fusion layer which is connected with the first convolution layer and the second convolution layer at the same time, wherein the first convolution layer and the second convolution layer respectively use one-dimensional convolution kernels with different scales.

In the above-mentioned colliery is in pit used fire prevention electrical system, the time sequence characteristic draws the module, includes: the first neighborhood scale feature extraction unit is used for respectively inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the first convolution layer is provided with a first one-dimensional convolution kernel with a first length; the second neighborhood scale feature extraction unit is used for respectively inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the second convolution layer is provided with a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and the multi-scale cascading unit is used for cascading the first neighborhood scale temperature time sequence feature vector, the first neighborhood scale smoke concentration value time sequence feature vector and the first neighborhood scale carbon monoxide concentration value time sequence feature vector with the second neighborhood scale temperature time sequence feature vector, the second neighborhood scale smoke concentration value time sequence feature vector and the second neighborhood scale carbon monoxide concentration value time sequence feature vector respectively to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector. The first neighborhood scale feature extraction unit is used for respectively carrying out one-dimensional convolution coding on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula so as to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

；

wherein ,ais the first convolution kernelxWidth in the direction,

For the first convolution kernel parameter vector, +.>

For a local vector matrix that operates with a convolution kernel,wfor the size of the first convolution kernel,Xrepresenting a temperature timing input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector, +.>

The method comprises the steps of respectively carrying out one-dimensional convolution coding on a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector;

the second neighborhood scale feature extraction unit is used for respectively carrying out one-dimensional convolution coding on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector by using a second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula so as to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

；

wherein b is the second convolution kernelxWidth in the direction,

For a second convolution kernel parameter vector, +.>

For the local vector matrix to operate with the convolution kernel function, m is the size of the second convolution kernel, XRepresenting a temperature timing input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector, +.>

The one-dimensional convolution coding is respectively carried out on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector.

In the underground coal mine fire prevention and extinguishing electric control system, the characteristic fusion module is used for fusing the temperature time sequence characteristic vector, the smoke concentration value time sequence characteristic vector and the carbon monoxide concentration value time sequence characteristic vector according to the following formula to obtain the classification characteristic vector; wherein, the formula is:

；

wherein

Representing a temperature timing feature vector, ">

Time sequence characteristic vector representing smoke concentration value, +.>

Time sequence characteristic vector representing concentration value of carbon monoxide, +.>

Representing classification feature vectors, ++>

、/>

and />

Weighting parameters respectively representing a temperature time series characteristic vector, a smoke concentration value time series characteristic vector and a carbon monoxide concentration value time series characteristic vector, +.>

Representing the sum by location.

In the underground coal mine fire prevention and extinguishing electric control system, the regularization module is used for regularizing the characteristic distribution of the classification characteristic vector according to the following optimization formula to obtain an optimized classification characteristic vector; wherein, the formula is:

；

wherein

and />

Is the mean and standard deviation of the feature values of the respective positions of the classification feature vector, and +.>

Is the +.o of the optimized classification feature vector>

Characteristic value of individual position->

Is the +.>

Characteristic values of the individual positions.

In the above-mentioned colliery is in pit to use fire prevention and extinguishing electrical system, control result generation module includes: the full-connection coding unit is used for carrying out full-connection coding on the optimized classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and the classification result generation unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain a classification result.

According to another aspect of the present application, there is provided an electrical control method for fire prevention and extinguishment for use downhole in a coal mine, comprising:

acquiring sensor data of a plurality of preset time points in a preset time period acquired by sensors of fire monitoring points deployed in an underground environment, wherein the sensor data comprise a temperature value, a smoke concentration value and a carbon monoxide concentration value;

arranging the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of a plurality of preset time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to a time dimension respectively;

Respectively passing the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through a multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector;

fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector to obtain a classification feature vector;

regularizing the feature distribution of the classified feature vectors to obtain optimized classified feature vectors;

and the optimized classification feature vector passes through a classifier to obtain a classification result, and the classification result is used for indicating whether the power supply of the fire monitoring point is cut off.

The application provides an electronic device, comprising: a processor; and a memory in which computer program instructions are stored which, when executed by the processor, cause the processor to perform the fire protection and extinguishing electrical control method for use downhole in a coal mine as described above.

The application provides a computer readable medium having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform a coal mine downhole use fire protection and suppression electrical control method as above.

Compared with the prior art, the underground coal mine fire prevention and extinguishing electric control system provided by the application adopts an artificial intelligent monitoring technology based on deep learning, the temperature value, the smoke concentration value and the carbon monoxide concentration value of the underground coal mine fire monitoring point are collected through multiple sensors, whether the fire monitoring point is in fire disaster or not is further judged by fusing time sequence change characteristics among the three, so that after the fire disaster occurs as a judgment result, the power supply of the fire monitoring point is cut off, and the fire disaster judgment accuracy is improved.

Drawings

The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

FIG. 1 is a schematic diagram of a scenario in which a fire prevention and extinguishing electronic control system is used downhole in a coal mine according to an embodiment of the present application;

FIG. 2 is a block diagram of a fire prevention and extinguishing electrical control system for use downhole in a coal mine according to an embodiment of the present application;

FIG. 3 is a system architecture diagram of a coal mine downhole use fire prevention and suppression electrical control system in accordance with an embodiment of the present application;

FIG. 4 is a block diagram of a timing feature extraction module in a coal mine downhole use fire prevention and suppression electrical control system in accordance with an embodiment of the present application;

FIG. 5 is a block diagram of a control result generation module in a coal mine downhole use fire prevention and extinguishing electrical control system according to an embodiment of the present application;

FIG. 6 is a flow chart of an electrical control method for fire prevention and extinguishing for use downhole in a coal mine according to an embodiment of the present application;

fig. 7 is a block diagram of an electronic device according to an embodiment of the present application.

Detailed Description

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Summary of the application

Aiming at the technical problems, the technical conception of the application is that the multi-sensor data of the fire monitoring points under the coal mine are collected through the multi-sensor, the multi-sensor data comprise temperature data, smoke concentration data, carbon monoxide concentration data and the like, and whether the fire monitoring points are in fire or not is judged through the fusion result of the multi-sensor data, so that after the judgment result is that the fire occurs, the power supply of the fire monitoring points is cut off.

Specifically, sensor data at a plurality of predetermined time points within a predetermined time period acquired by sensors deployed at fire monitoring points of a downhole environment is first acquired, wherein the sensor data includes a temperature value, a smoke concentration value, and a carbon monoxide concentration value. That is, a multi-sensor combination is deployed at a fire monitoring point to collect data of the fire monitoring point at a plurality of predetermined time points within a predetermined period of time, including temperature data, smoke concentration data, and carbon monoxide data, by the multi-sensor combination to judge whether a fire occurs or to early warn of the occurrence of the fire, etc., by a discrete distribution of the above data in time series.

Next, the temperature value, the smoke concentration value, and the carbon monoxide concentration value in the sensor data at a plurality of predetermined time points are arranged in a time dimension into a temperature time sequence input vector, a smoke concentration value input vector, and a carbon monoxide concentration value input vector, respectively. Here, the sensor data at a plurality of predetermined time points are arranged into different time-series input vectors in the time dimension in order to enable the machine learning model to process and classify the data. When a fire disaster occurs in a coal mine underground, parameters such as temperature, smoke concentration and carbon monoxide concentration change along with time, and the parameters are arranged into different vectors according to time axes, so that characteristic information related to time can be conveniently extracted, and the fire disaster can be better identified and monitored. For example, by analyzing the change trend of the temperature value, the carbon monoxide concentration value and the smoke concentration value, whether a fire disaster occurs or not can be judged in advance, and measures can be taken in time to treat the fire disaster, so that the occurrence of accidents is avoided.

And the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector are respectively passed through a multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector. Here, in the technical solution of the present application, in underground coal mines, the occurrence of fire is often caused by complex actions of various factors, such as temperature rise, smoke concentration increase, and carbon monoxide concentration exceeding, so that effective feature extraction needs to be performed on these factors from different levels and different scales.

The multi-scale neighborhood feature extraction module can simultaneously consider features in the time dimension and the space dimension, and extract important features related to fire, such as temperature fluctuation conditions, smoke concentration change trend, carbon monoxide concentration peak value and the like, by analyzing the relation between sensor data of different time points. Specifically, the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector are respectively processed through a multi-scale neighborhood feature extraction module to obtain three different time sequence feature vectors, wherein each feature vector reflects the change features of corresponding sensor data in time and space.

In a specific example of the present application, the multi-scale neighborhood feature extraction module includes a first convolution layer and a second convolution layer in parallel, and a multi-scale fusion layer connected to the first convolution layer and the second convolution layer simultaneously, where the first convolution layer and the second convolution layer use one-dimensional convolution kernels having different scales, respectively.

Then, the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector are fused to obtain a classification feature vector. In a coal mine underground fire prevention scheme, a plurality of sensors are used for collecting different data, such as temperature values, smoke concentration, carbon monoxide concentration and the like. It should be appreciated that the above sensor data is fire-related and may reflect the occurrence and risk of fire. However, considering the value of each parameter alone does not determine well whether a fire has occurred, because the occurrence of a fire is often caused by the complex actions of a plurality of factors, and comprehensive determination of these factors is required.

Based on this, by fusing the temperature time series feature vector, the smoke concentration value time series feature vector, and the carbon monoxide concentration value time series feature vector, a classification feature vector containing more comprehensive information can be obtained. Therefore, whether fire occurs or not can be accurately judged, and measures can be timely taken for treatment. Meanwhile, according to different fire conditions, the weight of each feature vector needs to be adjusted so as to better adapt to early warning and prevention under different scenes. In a specific example of the present application, the classification feature vector is obtained by fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector in a point-to-point adding manner.

And then, the classification feature vector passes through a classifier to obtain a classification result, and the classification result is used for indicating whether the power supply of the fire monitoring point is cut off. That is, a classifier is used to determine the class boundary to which the classification feature vector belongs to obtain a class probability tag, wherein the classification result is to cut off the power supply of the fire monitoring point (first tag) and not cut off the power supply of the fire monitoring point (second tag). An implicit logic determination is hidden, namely, when a fire is determined to occur, the power supply of the fire monitoring point (the adopted coping strategy) is cut off, and when no fire is determined to occur, the power supply of the fire monitoring point (the adopted coping strategy) is not cut off.

Particularly, in the technical scheme of the application, when the classification feature vector is obtained by fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector in a point-by-point adding manner, the time sequence distribution of each sensor at a plurality of time points has different distribution characteristics due to the fact that noise or abnormal values exist in the data of each sensor, so that when the multi-scale neighborhood feature extraction module extracts the multi-scale time sequence neighborhood associated features, the respective feature distributions have mismatch in the arrangement direction of the time sequence features, the obtained classification feature vector has the problem of irregularity of the feature distribution, and the classification accuracy of the classification feature vector is affected.

Based on this, the applicant of the present application classifies feature vectors

Performing Gaussian probability density parameter secondary regularization on the manifold curved surface, wherein the method is specifically expressed as follows:

；

wherein

and />

Is a feature value set +.>

Mean and standard deviation of (2), and->

Is an optimized classification feature vector +.>

Is>

Characteristic value of individual position->

Is the +.>

Characteristic values of the individual positions.

Specifically, to solve the classification feature vector

The problem of irregular distribution of high-dimensional feature distribution of feature set in high-dimensional feature space by classifying feature vector +.>

Performing secondary regularization of the overall distribution characteristic of the feature values with respect to the likelihood of the gaussian probability density parameter of the class probability to perform smooth constraint of the feature values based on equidistant distribution in the parameter space of the gaussian probability density parameter to obtain regularized reformation of the original probability density likelihood function of the manifold surface expression of the high-dimensional feature in the parameter space, thereby improving the optimized classification feature vector->

And further improves the accuracy of the classification result of the classifier.

Based on this, this application provides a colliery underground use fire prevention electrical system, and it includes: the system comprises a multi-sensor data acquisition module, a sensor data acquisition module and a data processing module, wherein the multi-sensor data acquisition module is used for acquiring sensor data of a plurality of preset time points in a preset time period acquired by sensors of fire monitoring points deployed in underground environment, and the sensor data comprises a temperature value, a smoke concentration value and a carbon monoxide concentration value; the vectorization module is used for respectively arranging the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of a plurality of preset time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to the time dimension; the time sequence feature extraction module is used for respectively passing the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through the multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector; the characteristic fusion module is used for fusing the temperature time sequence characteristic vector, the smoke concentration value time sequence characteristic vector and the carbon monoxide concentration value time sequence characteristic vector to obtain a classification characteristic vector; the regularization module is used for regularizing the characteristic distribution of the classification characteristic vector to obtain an optimized classification characteristic vector; and the control result generation module is used for enabling the optimized classification feature vector to pass through the classifier to obtain a classification result, and the classification result is used for indicating whether the power supply of the fire monitoring point is cut off or not.

Fig. 1 is a schematic view of a scenario of a coal mine downhole use fire prevention and extinguishing electrical control system according to an embodiment of the present application. As shown in fig. 1, in this application scenario, temperature values at a plurality of predetermined time points within a predetermined period are acquired by a temperature sensor (e.g., V1 as illustrated in fig. 1), smoke concentration values at a plurality of predetermined time points within a predetermined period are acquired by a smoke concentration sensor (e.g., V2 as illustrated in fig. 1), and carbon monoxide concentration values at a plurality of predetermined time points within a predetermined period are acquired by a carbon monoxide concentration sensor (e.g., V3 as illustrated in fig. 1). The data is then input to a server (e.g., S in fig. 1) deployed with an electrical control algorithm for fire protection downhole in the coal mine, wherein the server is capable of processing the input data with the electrical control algorithm for fire protection downhole in the coal mine to generate a classification result indicative of whether to cut off power to the fire monitoring point.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

Exemplary System

Fig. 2 is a block diagram of a fire prevention and extinguishing electrical control system for use downhole in a coal mine according to an embodiment of the present application. As shown in fig. 2, a downhole-use fire prevention and extinguishing electronic control system 300 for a coal mine according to an embodiment of the present application includes: a multi-sensor data acquisition module 310; a vectorization module 320; a timing feature extraction module 330; a feature fusion module 340; a regularization module 350; and a control result generation module 360.

The multi-sensor data acquisition module 310 is configured to acquire sensor data of a plurality of predetermined time points within a predetermined time period acquired by sensors deployed at fire monitoring points in a downhole environment, where the sensor data includes a temperature value, a smoke concentration value, and a carbon monoxide concentration value; a vectorization module 320, configured to arrange the temperature value, the smoke concentration value, and the carbon monoxide concentration value in the sensor data at a plurality of predetermined time points into a temperature time sequence input vector, a smoke concentration value input vector, and a carbon monoxide concentration value input vector according to a time dimension, respectively; the time sequence feature extraction module 330 is configured to pass the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through the multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector; the feature fusion module 340 is configured to fuse the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector to obtain a classification feature vector; a regularization module 350, configured to regularize the feature distribution of the classification feature vector to obtain an optimized classification feature vector; and a control result generation module 360, configured to pass the optimized classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether to cut off the power supply of the fire monitoring point.

Fig. 3 is a system architecture diagram of a fire prevention and extinguishing electrical control system used downhole in a coal mine according to an embodiment of the present application. As shown in fig. 3, in the network architecture, sensor data of a plurality of predetermined time points within a predetermined time period acquired by sensors deployed at fire monitoring points of a downhole environment are first acquired by a multi-sensor data acquisition module 310, wherein the sensor data includes a temperature value, a smoke concentration value and a carbon monoxide concentration value; next, the vectorizing module 320 arranges the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of a plurality of predetermined time points acquired by the multi-sensor data acquisition module 310 into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to a time dimension; the time sequence feature extraction module 330 respectively passes the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector obtained by the vectorization module 320 through the multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector; the feature fusion module 340 fuses the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector obtained by the time sequence feature extraction module 330 to obtain a classification feature vector; then, the regularization module 350 regularizes the feature distribution of the classification feature vector obtained by the feature fusion module 340 to obtain an optimized classification feature vector; further, the control result generation module 360 passes the optimized classification feature vector through the classifier to obtain a classification result, which is used to indicate whether to cut off the power supply of the fire monitoring point.

Specifically, during operation of the fire prevention and extinguishing electrical control system 300 in use downhole in a coal mine, the multi-sensor data acquisition module 310 is configured to acquire sensor data at a plurality of predetermined time points within a predetermined time period acquired by sensors deployed at fire monitoring points in the downhole environment, wherein the sensor data includes a temperature value, a smoke concentration value, and a carbon monoxide concentration value. It should be understood that in the technical solution of the present application, whether a fire disaster occurs at a fire disaster monitoring point may be determined by the concept of multi-sensor data fusion, that is, a multi-sensor is deployed at the fire disaster monitoring point to collect data of a plurality of predetermined time points of the fire disaster monitoring point within a predetermined period of time through the multi-sensor, including temperature data, smoke concentration data and carbon monoxide data, so as to determine whether the fire disaster occurs or early warning the occurrence of the fire disaster through discrete distribution of the above data in time sequence. In one example, temperature values at a plurality of predetermined time points within a predetermined time period may be acquired by a temperature sensor, smoke concentration values at a plurality of predetermined time points within a predetermined time period may be acquired by a smoke concentration sensor, and carbon monoxide concentration values at a plurality of predetermined time points within a predetermined time period may be acquired by a carbon monoxide concentration sensor.

Specifically, during the operation of the fire prevention and extinguishing electrical control system 300 used in a coal mine, the vectorization module 320 is configured to arrange the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of a plurality of predetermined time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to a time dimension, respectively. It should be appreciated that the arrangement of the sensor data at a plurality of predetermined points in time into different time-series input vectors in the time dimension is to enable the machine learning model to process and classify the data. When a fire disaster occurs in a coal mine underground, parameters such as temperature, smoke concentration and carbon monoxide concentration change along with time, and the parameters are arranged into different vectors according to time axes, so that characteristic information related to time can be conveniently extracted, and the fire disaster can be better identified and monitored. For example, by analyzing the change trend of the temperature value, the carbon monoxide concentration value and the smoke concentration value, whether a fire disaster occurs or not can be judged in advance, and measures can be taken in time to treat the fire disaster, so that the occurrence of accidents is avoided.

Specifically, in the operation process of using the fire prevention and extinguishing electric control system 300 in the coal mine, the time sequence feature extraction module 330 is configured to pass the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through the multi-scale neighborhood feature extraction module to obtain the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector. That is, in the technical scheme of the application, the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector are respectively passed through the multi-scale neighborhood feature extraction module to obtain the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector. In the technical scheme of the application, under the coal mine, fire disasters often occur due to complex actions of various factors, such as temperature rise, smoke concentration increase, carbon monoxide concentration exceeding and the like, so that effective feature extraction needs to be carried out on the factors from different levels and different scales. The multi-scale neighborhood feature extraction module can simultaneously consider features in the time dimension and the space dimension, and extract important features related to fire, such as temperature fluctuation conditions, smoke concentration change trend, carbon monoxide concentration peak value and the like, by analyzing the relation between sensor data of different time points. Specifically, the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector are respectively processed through a multi-scale neighborhood feature extraction module to obtain three different time sequence feature vectors, wherein each feature vector reflects the change features of corresponding sensor data in time and space. In particular, the multi-scale neighborhood feature extraction module comprises a first convolution layer and a second convolution layer which are parallel, and a multi-scale fusion layer connected with the first convolution layer and the second convolution layer at the same time, wherein the first convolution layer and the second convolution layer respectively use one-dimensional convolution kernels with different scales.

Fig. 4 is a block diagram of a timing feature extraction module in a coal mine downhole use fire prevention and extinguishing electrical control system according to an embodiment of the application. As shown in fig. 4, the timing feature extraction module 330 includes: the first neighborhood scale feature extraction unit 331 is configured to input a temperature time sequence input vector, a smoke concentration value input vector, and a carbon monoxide concentration value input vector into a first convolution layer of the multi-scale neighborhood feature extraction module respectively to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector, and a first neighborhood scale carbon monoxide concentration value time sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second neighborhood scale feature extraction unit 332, configured to input the temperature timing sequence input vector, the smoke concentration value input vector, and the carbon monoxide concentration value input vector into a second convolution layer of the multi-scale neighborhood feature extraction module, respectively, to obtain a second neighborhood scale temperature timing sequence feature vector, a second neighborhood scale smoke concentration value timing sequence feature vector, and a second neighborhood scale carbon monoxide concentration value timing sequence feature vector, where the second convolution layer has a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and a multi-scale cascading unit 333, configured to cascade the first neighborhood scale temperature timing feature vector, the first neighborhood scale smoke concentration value timing feature vector, and the first neighborhood scale carbon monoxide concentration value timing feature vector with the second neighborhood scale temperature timing feature vector, the second neighborhood scale smoke concentration value timing feature vector, and the second neighborhood scale carbon monoxide concentration value timing feature vector, respectively, to obtain a temperature timing feature vector, a smoke concentration value timing feature vector, and a carbon monoxide concentration value timing feature vector. The first neighborhood scale feature extraction unit 331 is configured to perform one-dimensional convolution encoding on the temperature time sequence input vector, the smoke concentration value input vector, and the carbon monoxide concentration value input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula, so as to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector, and a first neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

；

wherein ,ais the first convolution kernelxWidth in the direction,

For the first convolution kernel parameter vector, +.>

The method comprises the steps of respectively carrying out one-dimensional convolution coding on a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector; second neighborhood scale feature extractionThe fetching unit 332 is configured to perform one-dimensional convolution encoding on the temperature time sequence input vector, the smoke concentration value input vector, and the carbon monoxide concentration value input vector by using the second convolution layer of the multi-scale neighborhood feature extraction module according to the following one-dimensional convolution formula, so as to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector, and a second neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

；

wherein b is the second convolution kernelxWidth in the direction,

For a second convolution kernel parameter vector, +.>

Specifically, during the operation of the fire prevention and extinguishing electrical control system 300 used in the underground coal mine, the feature fusion module 340 is configured to fuse the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector to obtain a classification feature vector containing more comprehensive information. That is, after the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector are obtained, the three are further subjected to feature fusion, and in the underground coal mine fire prevention scheme, a plurality of sensors are used for collecting different data such as temperature values, smoke concentrations, carbon monoxide concentrations and the like. It should be appreciated that the above-described sensingThe device data are all related to fire and can reflect the occurrence and the dangerous degree of the fire. However, considering the value of each parameter alone does not determine well whether a fire has occurred, because the occurrence of a fire is often caused by the complex actions of a plurality of factors, and comprehensive determination of these factors is required. In a specific example of the application, the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector are integrated in a point-to-point adding mode to obtain the classification feature vector. More specifically, the temperature time series feature vector, the smoke concentration value time series feature vector and the carbon monoxide concentration value time series feature vector are fused in the following formula to obtain a classification feature vector; wherein, the formula is:

, wherein />

Representing a temperature timing feature vector, ">

Time sequence characteristic vector representing smoke concentration value, +.>

Representing classification feature vectors, ++>

And

Representing the sum by location.

Specifically, operation of fire suppression and electric control system 300 in a coal mine downholeIn the process, the regularization module 350 is configured to regularize the feature distribution of the classification feature vector to obtain an optimized classification feature vector. In the technical scheme of the application, when the classification feature vector is obtained by fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector in a point-by-point adding mode, the problem that the obtained classification feature vector has irregular feature distribution is caused by considering that noise or abnormal values exist in data of each sensor, so that time sequence distribution of each sensor has different distribution characteristics at a plurality of time points, and when the multiscale neighborhood feature extraction module extracts multiscale time sequence neighborhood associated features, mismatch exists in the time sequence feature arrangement direction of each feature distribution, and classification accuracy of the classification feature vector is affected. Based on this, the applicant of the present application classifies feature vectors

concatenated vector

；

wherein

and />

Is the +.o of the optimized classification feature vector>

Characteristic value of individual position->

Is the +.>

Characteristic values of the individual positions. Specifically, to solve the classification feature vector +.>

Specifically, during the operation of the fire prevention and extinguishing electric control system 300 used in the underground coal mine, the control result generating module 360 is configured to pass the optimized classification feature vector through the classifier to obtain a classification result, where the classification result is used to indicate whether to cut off the power supply of the fire monitoring point. That is, the optimized classification feature vector is passed through a classifier as a classification feature vector to obtain a classification result, and specifically, the optimized classification feature vector is processed using the classifier in the following formula:

, wherein ,/>

To->

Is a weight matrix>

To->

For the bias vector +.>

To optimize the classification feature vector. Specifically, the classifier includes a plurality of fully connected layers and a Softmax layer cascaded with a last fully connected layer of the plurality of fully connected layers. In the classification processing of the classifier, a plurality of full-connection layers of the classifier are used for carrying out full-connection coding on the optimized classification feature vector for a plurality of times to obtain a coding classification feature vector; further, the encoded classification feature vector is input to the Softmax layer of the classifier, i.e., the encoded classification feature vector is classified using a Softmax classification function to obtain a classification label. That is, a classifier is used to determine the class boundary to which the classification feature vector belongs to obtain a class probability tag, wherein the classification result is to cut off the power supply of the fire monitoring point (first tag) and not cut off the power supply of the fire monitoring point (second tag). Here, an implicit logic determination is hidden, that is, when it is determined that a fire is occurring, the power supply of the fire monitoring point (the adopted coping strategy) is shut off, and when it is determined that a fire is not occurring, the power supply of the fire monitoring point (the adopted coping strategy) is not shut off.

Fig. 5 is a block diagram of a control result generation module in a coal mine underground use fire prevention and extinguishing electric control system according to an embodiment of the application. As shown in fig. 5, the control result generation module 360 includes: a full-connection encoding unit 361, configured to perform full-connection encoding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain an encoded classification feature vector; and a classification result generating unit 362, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain a classification result.

In summary, according to the underground coal mine use fire prevention and extinguishing electric control system 300 of the embodiment of the application, an artificial intelligent monitoring technology based on deep learning is adopted, a temperature value, a smoke concentration value and a carbon monoxide concentration value of a fire monitoring point in the underground coal mine are collected through multiple sensors, whether the fire monitoring point is in fire or not is further judged by fusing time sequence change characteristics among the three, so that after the judgment result is that the fire occurs, a power supply of the fire monitoring point is cut off, and the fire judgment accuracy is improved.

As above, the underground coal mine fire prevention and extinguishing electric control system according to the embodiment of the application can be implemented in various terminal devices. In one example, the downhole use fire suppression electrical control system 300 for coal mines according to embodiments of the present application may be integrated into the terminal equipment as a software module and/or hardware module. For example, the downhole use fire prevention and extinguishing electronic control system 300 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the fire suppression electrical control system 300 for use downhole in a coal mine may also be one of many hardware modules of the terminal equipment.

Alternatively, in another example, the downhole-use fire prevention and extinguishing electronic control system 300 may be a separate device from the terminal device, and the downhole-use fire prevention and extinguishing electronic control system 300 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.

Exemplary method

Fig. 6 is a flow chart of an electrical control method for fire prevention and extinguishing in a coal mine according to an embodiment of the application. As shown in fig. 6, the electric control method for preventing and extinguishing fire for underground coal mine according to the embodiment of the application comprises the following steps: s110, acquiring sensor data of a plurality of preset time points in a preset time period acquired by sensors of fire monitoring points deployed in an underground environment, wherein the sensor data comprise a temperature value, a smoke concentration value and a carbon monoxide concentration value; s120, arranging temperature values, smoke concentration values and carbon monoxide concentration values in sensor data of a plurality of preset time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to a time dimension respectively; s130, respectively passing the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector through a multi-scale neighborhood feature extraction module to obtain a temperature time sequence feature vector, a smoke concentration value time sequence feature vector and a carbon monoxide concentration value time sequence feature vector; s140, fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector to obtain a classification feature vector; s150, regularizing the feature distribution of the classified feature vectors to obtain optimized classified feature vectors; and S160, enabling the optimized classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the power supply of the fire monitoring point is cut off.

In one example, in the above-mentioned method for controlling electric fire prevention and extinguishing in coal mine underground, step S130 includes: respectively inputting a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector into a first convolution layer of a multi-scale neighborhood feature extraction module to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the first convolution layer is provided with a first one-dimensional convolution kernel with a first length; respectively inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the second convolution layer is provided with a second one-dimensional convolution kernel with a second length, and the first length is different from the second length; and cascading the first neighborhood scale temperature timing characteristic vector, the first neighborhood scale smoke concentration value timing characteristic vector and the first neighborhood scale carbon monoxide concentration value timing characteristic vector with the second neighborhood scale temperature timing characteristic vector, the second neighborhood scale smoke concentration value timing characteristic vector and the second neighborhood scale carbon monoxide concentration value timing characteristic vector respectively to obtain a temperature timing characteristic vector, a smoke concentration value timing characteristic vector and a carbon monoxide concentration value timing characteristic vector. The multi-scale neighborhood feature extraction module comprises a first convolution layer and a second convolution layer which are parallel, and a multi-scale fusion layer which is connected with the first convolution layer and the second convolution layer at the same time, wherein the first convolution layer and the second convolution layer respectively use one-dimensional convolution kernels with different scales. More specifically, inputting a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector into a first convolution layer of a multi-scale neighborhood feature extraction module respectively to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector, including: using a first convolution layer of the multi-scale neighborhood feature extraction module to respectively perform one-dimensional convolution coding on a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to the following one-dimensional convolution formula so as to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

；

wherein ,ais the first convolution kernelxWidth in the direction,

For the first convolution kernel parameter vector, +.>

The method comprises the steps of respectively carrying out one-dimensional convolution coding on a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector; respectively inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale temperature time sequence feature vector,A second neighborhood scale smoke concentration value timing feature vector and a second neighborhood scale carbon monoxide concentration value timing feature vector, comprising: using a second convolution layer of the multi-scale neighborhood feature extraction module to respectively perform one-dimensional convolution coding on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector according to the following one-dimensional convolution formula so as to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector; wherein, the formula is:

； />

Wherein b is the second convolution kernelxWidth in the direction,

For a second convolution kernel parameter vector, +.>

For the local vector matrix to operate with the convolution kernel function, m is the size of the second convolution kernel,Xrepresenting a temperature timing input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector, +.>

In one example, in the above-mentioned method for controlling electric fire prevention and extinguishing in coal mine underground, step S140 includes: fusing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector to obtain a classification feature vector by the following formula; wherein, the formula is:

, wherein />

Representing a temperature timing feature vector, ">

Time sequence characteristic vector representing smoke concentration value, +.>

Representing classification feature vectors, ++>

and />

Representing the sum by location.

In one example, in the above-mentioned method for controlling electric fire prevention and extinguishing in coal mine underground, step S150 includes: regularizing the feature distribution of the classified feature vectors by using the following optimization formula to obtain optimized classified feature vectors; wherein, the formula is:

；

wherein

and />

Is the +.o of the optimized classification feature vector>

Characteristic value of individual position->

Is the +.>

Characteristic values of the individual positions.

In one example, in the above-mentioned method for controlling electric fire prevention and extinguishing in coal mine underground, step S160 includes: performing full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier to obtain a coded classification feature vector; and passing the encoded classification feature vector through a Softmax classification function of the classifier to obtain a classification result.

In summary, according to the underground coal mine use fire prevention and extinguishing electric control method of the embodiment of the application, an artificial intelligent monitoring technology based on deep learning is adopted, a temperature value, a smoke concentration value and a carbon monoxide concentration value of a fire monitoring point underground the coal mine are collected through multiple sensors, whether the fire monitoring point is in fire or not is further judged by fusing time sequence change characteristics among the three, so that after a judgment result is that the fire occurs, a power supply of the fire monitoring point is cut off, and the accuracy of fire judgment is improved.

Exemplary electronic device

Next, an electronic device according to an embodiment of the present application is described with reference to fig. 7.

Fig. 7 illustrates a block diagram of an electronic device according to an embodiment of the present application.

As shown in fig. 7, the electronic device 10 includes one or more processors 11 and a memory 12.

The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. Volatile memory can include, for example, random Access Memory (RAM) and/or cache memory (cache) and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on a computer readable storage medium and the processor 11 may execute the program instructions to implement the functions in the coal mine downhole use fire protection and suppression electrical control system of the various embodiments of the present application above and/or other desired functions. Various contents such as classification feature vectors may also be stored in the computer-readable storage medium.

In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).

The input means 13 may comprise, for example, a keyboard, a mouse, etc.

The output device 14 may output various information including the classification result and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.

Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 7 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer readable storage Medium

In addition to the methods and apparatus described above, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the coal mine downhole use fire protection and control method according to various embodiments of the present application described in the "exemplary systems" section of this specification.

The computer program product may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer-readable storage medium, having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in the functions of the coal mine downhole use fire prevention and suppression electrical control method according to various embodiments of the present application described in the "exemplary systems" section of the present specification.

A computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.

The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims

1. An electrical control system for preventing and extinguishing fire used underground in a coal mine, which is characterized by comprising:

The vectorization module is used for arranging the temperature value, the smoke concentration value and the carbon monoxide concentration value in the sensor data of the plurality of preset time points into a temperature time sequence input vector, a smoke concentration value input vector and a carbon monoxide concentration value input vector according to a time dimension respectively;

and the control result generation module is used for enabling the optimized classification feature vector to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether the power supply of the fire monitoring point is cut off or not.

2. The downhole use fire prevention and extinguishing electrical control system of claim 1, wherein the multi-scale neighborhood feature extraction module comprises a first convolution layer and a second convolution layer in parallel, and a multi-scale fusion layer simultaneously connected to the first convolution layer and the second convolution layer; wherein the first convolution layer and the second convolution layer each use one-dimensional convolution kernels having different dimensions.

3. The downhole use fire prevention and extinguishing electrical control system of claim 2, wherein the timing feature extraction module comprises:

the first neighborhood scale feature extraction unit is used for respectively inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into a first convolution layer of the multi-scale neighborhood feature extraction module to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the first convolution layer is provided with a first one-dimensional convolution kernel with a first length;

a second neighborhood scale feature extraction unit for inputting the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector into the multi-scale neighborhood respectively

The second convolution layer of the feature extraction module is used for obtaining a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector, wherein the second convolution layer is provided with a second one-dimensional convolution kernel with a second length, and the first length is different from the second length;

the multi-scale cascading unit is used for cascading the first neighborhood scale temperature time sequence feature vector, the first neighborhood scale smoke concentration value time sequence feature vector and the first neighborhood scale carbon monoxide concentration value time sequence feature vector with the second neighborhood scale temperature time sequence feature vector, the second neighborhood scale smoke concentration value time sequence feature vector and the second neighborhood scale carbon monoxide concentration value time sequence feature vector respectively to obtain the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector;

the first neighborhood scale feature extraction unit is configured to perform one-dimensional convolution encoding on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector by using a first convolution layer of the multi-scale neighborhood feature extraction module according to a one-dimensional convolution formula to obtain a first neighborhood scale temperature time sequence feature vector, a first neighborhood scale smoke concentration value time sequence feature vector and a first neighborhood scale carbon monoxide concentration value time sequence feature vector;

Wherein, the formula is:

;

wherein ,ais the first convolution kernelxWidth in the direction,

For the first convolution kernel parameter vector, +.>

For a local vector matrix that operates with a convolution kernel,wfor the size of the first convolution kernel,Xrepresenting the temperature timing input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector,/v>

Representing that the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector are respectively subjected to one-dimensional convolution coding;

the second neighborhood scale feature extraction unit is configured to perform one-dimensional convolution encoding on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector according to a one-dimensional convolution formula by using a second convolution layer of the multi-scale neighborhood feature extraction module to obtain a second neighborhood scale temperature time sequence feature vector, a second neighborhood scale smoke concentration value time sequence feature vector and a second neighborhood scale carbon monoxide concentration value time sequence feature vector;

wherein, the formula is:

;

wherein b is the second convolution kernelxWidth in the direction,

For a second convolution kernel parameter vector, +. >

For the local vector matrix to operate with the convolution kernel function, m is the size of the second convolution kernel,Xrepresenting the temperature timing input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector,/v>

And the one-dimensional convolution coding is respectively carried out on the temperature time sequence input vector, the smoke concentration value input vector and the carbon monoxide concentration value input vector.

4. A downhole-use fire prevention and extinguishing electrical control system according to claim 3, wherein the feature fusion module is configured to fuse the temperature time series feature vector, the smoke concentration value time series feature vector, and the carbon monoxide concentration value time series feature vector to obtain a classification feature vector according to the following formula;

wherein, the formula is:

;

wherein

Representing the temperature timing feature vector, +_>

Time sequence characteristic vector representing the smoke concentration value, < >>

Time sequence characteristic direction representing the concentration value of the carbon monoxideQuantity (S)>

Representing the classification feature vector,/->

and />

Weighting parameters respectively representing the temperature time sequence feature vector, the smoke concentration value time sequence feature vector and the carbon monoxide concentration value time sequence feature vector,/a>

Representing the sum by location.

5. The underground coal mine use fire prevention and extinguishing electric control system according to claim 4, wherein the regularization module is configured to regularize the feature distribution of the classification feature vector according to the following optimization formula to obtain an optimized classification feature vector;

wherein, the formula is:

;

wherein

and />

Is the +.o of the optimized classification feature vector>

Characteristic value of individual position->

Is the +.>

Characteristic values of the individual positions.

6. The downhole use fire prevention and extinguishing electrical control system of claim 5, wherein the control result generation module comprises:

the full-connection coding unit is used for carrying out full-connection coding on the optimized classification feature vector by using a plurality of full-connection layers of the classifier so as to obtain a coding classification feature vector;

and the classification result generating unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.