Mine fire monitoring system

Abstract

The invention discloses a mine fire monitoring system which integrates a plurality of functional modules, a main controller is connected with various sensors and functional modules such as an alarm module, a fire extinguishing module, a display module and a communication module, and sensor data are divided into two groups to be input into different systems by adopting a mode of combining a fuzzy logic theory and a BP (Back Propagation) neural network; a fire judgment result is obtained through fusion judgment, and the accuracy of fire judgment is improved. The sensor collects underground environment data and transmits the underground environment data to the main controller, and the main controller controls the alarm module to send out different types of alarm signals according to the result and controls the fire extinguishing module to execute corresponding operation at the same time. The display module displays environment data in real time, and the communication module realizes data transmission and interaction between external devices. And a comprehensive and efficient solution is provided for monitoring, early warning and fire extinguishing of underground fire disasters.

Description

Technical Field

[0001] The present application relates to the field of mine safety technology, and in particular to a mine fire monitoring system. Background Art

[0002] Coal resources are one of the most important energy sources for human development. They are widely used in people's daily life and industrial production, especially in power generation, life and metallurgy. They still play an irreplaceable role in industry and life, and will continue to play a stable role in the future development of society. However, in the process of coal mining, safety issues have always been one of the difficult problems to solve. Fire is also one of the most common accidents in coal mines, which seriously threatens the safety of workers' lives and property. Therefore, the design and research of coal mine fire monitoring systems is particularly important.

[0003] With the continuous development of science and technology, the development of fire monitoring systems is generally fast. From the initial multi-line fire monitoring system, to the later bus-type fire monitoring system, to the currently studied intelligent fire monitoring system, people have continuously broken through technical difficulties, improved the reliability and stability of the system, and also made sufficient preparations and foundations for the design of coal mine fire monitoring systems.

[0004] In addition to monitoring the environment and transmitting fire information to the main controller, foreign fire monitoring systems can send alarm signals to the alarm after receiving fire signals, and can also link other equipment, such as: automatic opening of fire extinguishing modules, automatic closing of fire shutter doors, automatic smoke exhaust, automatic power switching, etc. In addition, some fire monitoring systems can also directly send alarms to the fire department.

[0005] Compared with Western countries, my country's fire protection industry started relatively late, with only a short history of more than 30 years since the 1980s. Although it has not been long, my country regards the fire protection industry as an emerging technology-intensive scientific and technological product to be developed, and has continuously developed and upgraded fire protection products by learning from foreign advanced achievements and high technology, and its development speed is extremely rapid. In recent years, due to the accelerated economic construction in my country, higher requirements have been put forward for fire monitoring systems. Domestic fire monitoring systems have also introduced mature and advanced foreign technologies, taking their essence and removing their dross, adding their own research and development on the basis of foreign fire monitoring systems for upgrading, and also accelerating the improvement of the level of domestic fire protection industry.

[0006] The current fire monitoring system generally has the problem of poor reliability, and false alarms and missed alarms occur frequently. This is mainly due to the limitations of sensor technology. Some sensors are susceptible to environmental interference (such as humidity, dust, electromagnetic interference, etc.), which leads to data deviation and then misjudgment. At the same time, the traditional method based on a single parameter or simple threshold judgment lacks a comprehensive analysis of fire characteristics and cannot effectively distinguish fire signals from normal environmental fluctuations or interference signals, resulting in an increased risk of missed alarms.

[0007] The wired fire monitoring system has obvious defects in setting flexibility, and the wiring process is greatly restricted by factors such as mine tunnel layout and geological conditions. During the mining process, tunnel extension and working face changes frequently, and rewiring is costly, time-consuming and labor-intensive. It is difficult to meet the needs of flexible deployment and rapid adjustment of the monitoring system in the dynamic mining environment of coal mines, limiting the effective coverage and application effect of the system.

[0008] The lack of intelligence is also a major shortcoming of the existing system. Although the intelligent fire monitoring system has made some progress, most of them are still at the stage of simple data collection and preliminary analysis, and fail to make full use of cutting-edge technologies such as artificial intelligence and big data analysis to achieve in-depth prediction and precise prevention and control of fire risks. For example, in terms of multi-sensor data fusion, there is a lack of efficient and intelligent algorithm models, which makes it impossible to deeply explore the inherent correlation and potential fire characteristics of the data, making it difficult to achieve accurate early warning and intelligent decision-making of fires, and unable to adapt to the complex and changing safety management needs of coal mines. Summary of the invention

[0009] The embodiment of the present application provides a mine fire monitoring system, which can improve the accuracy and efficiency of fire monitoring.

[0010] In order to achieve the above object, the technical solution of the embodiment of the present invention is:

[0011] In a first aspect, an embodiment of the present invention provides a mine fire monitoring system, comprising: a plurality of sensors for collecting underground environment data of a target mine, the sensors comprising at least a temperature sensor, a smoke sensor, a CO sensor and an infrared flame sensor;

[0012] A main controller is connected to the multiple sensors and is used to receive and process the data collected by the sensors; the main controller processes the fire signal based on a combination of fuzzy logic theory and BP neural network, including:

[0013] The temperature, smoke concentration and flame signal collected by the sensor are input into a preset fuzzy logic reasoning subsystem for processing to obtain a preliminary prediction result of the fire probability;

[0014] The temperature, smoke concentration, and CO concentration signals are input into the pre-trained BP neural network subsystem for processing, and the prediction results of the probability of open fire, the probability of smoldering, and the probability of no fire are obtained respectively;

[0015] The output results of the fuzzy logic reasoning subsystem and the BP neural network subsystem are then fused and judged through fuzzy logic to obtain the final fire judgment result;

[0016] An alarm module, connected to the main controller, for sending an alarm signal when a fire occurs according to the fire determination result of the main controller;

[0017] A fire extinguishing module, connected to the main controller, for performing a fire extinguishing operation when a fire occurs according to the fire determination result of the main controller;

[0018] A display module, connected to the main controller, for displaying the environmental data collected by the multiple sensors in real time;

[0019] The communication module is connected to the main controller and is used to transmit the data collected by the multiple sensors to an external device and receive control instructions from the external device.

[0020] In some possible implementations, in the fuzzy logic reasoning subsystem, the data of the temperature sensor and the smoke sensor are fuzzified using a Gaussian membership function, and the data of the infrared flame sensor are fuzzified using a triangular membership function;

[0021] The fuzzy logic reasoning subsystem performs reasoning operations based on the fuzzified data according to preset fuzzy rules, and obtains a preliminary prediction result of the fire probability after defuzzification.

[0022] In some possible implementations, in the preset BP neural network subsystem, a three-layer forward network structure and a preset BP learning algorithm are used, and the temperature, smoke concentration and CO concentration signals collected by the sensor are used as input to output the probability of open fire, the probability of smoldering and the probability of no fire.

[0023] In some possible implementations, the alarm module includes but is not limited to alarm modes of sound and light signals, and the alarm module has different alarm modes according to different fire determination results.

[0024] In some possible implementations, the fire extinguishing module is connected to a water pump via a relay, and is used to turn on or off the water pump to extinguish a fire upon receiving an instruction from the main controller.

[0025] In some possible implementations, the system also includes a cloud service platform, and the main controller sends the data collected by the sensor to the cloud service platform through the communication module for fire signal processing, and receives the fire determination result sent by the cloud service platform through the communication module.

[0026] In some possible implementations, the external device includes a remote control terminal, which can send a variety of control instructions for controlling the working status of the alarm module and / or the fire extinguishing module to the main controller by interacting with the communication module.

[0027] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0028] In the embodiment of the present invention, multi-sensor information fusion technology is used to collect and process environmental information, and a fuzzy logic reasoning system is used to make judgments at the decision-making level, which integrates the advantages of fuzzy reasoning and neural networks, and improves the accuracy of fire judgment. When the system detects that a fire has occurred, it will automatically turn on the alarm module to warn the workers underground, and at the same time turn on the fire extinguishing module to extinguish the fire. When the monitoring value returns to normal, the fire extinguishing module and the alarm module automatically return to the closed state. The temperature, smoke, CO and flame sensors are used to monitor the environmental data information under the mine in real time, and the monitored data and images are displayed, and transmitted to the control center through the communication module. On the one hand, the workers can correctly judge the current situation in the environment according to the information displayed on the on-site screen, and can react according to the alarm information when a fire occurs. On the other hand, the relevant person in charge of the ground control center can also obtain the data information of the mine environment in real time, and can obtain the fire scene data information and the fire scene location in time when the environment is abnormal, and control the alarm and arrange relevant personnel to carry out fire fighting and rescue in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention, the accompanying drawings required for use in the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0030] Figure 1 A schematic diagram of the structure of a mine fire monitoring system provided for the implementation of the present invention;

[0031] Figure 2 Schematic diagram of the fire signal processing structure based on fuzzy neural network in an embodiment of the present invention;

[0032] Figure 3is a structural schematic diagram of a fuzzy logic reasoning subsystem in an embodiment of the present invention;

[0033] Figure 4 is a curve diagram of a triangular membership function in an embodiment of the present invention;

[0034] Figure 5 A schematic diagram of the structure of a BP neural network subsystem in an embodiment of the present invention;

[0035] Figure 6 Schematic diagram of the correlation between actual samples and training results in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] In the relevant description of this embodiment, the terms "including, containing, having" and the like are open terms and are generally understood to include but not be limited to; the term "at least one" is generally understood to mean one or more, where "plurality" refers to two or more; the term "at least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items, for example, "at least one of a, b or c", or "at least one of a, b and c", can all represent: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, c can be single or multiple, respectively; the symbol "A / B" is used to describe the selection relationship of associated objects, generally indicating an "or" relationship before and after.

[0038] In the following description of the present embodiment, the terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0039] Those skilled in the art should understand that in the following description of the embodiments of the present application, the order of serial numbers does not mean the order of execution, some or all of the steps can be executed in parallel or sequentially, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0040] Those skilled in the art will appreciate that the numerical ranges in the embodiments of the present application are to be construed as also specifically disclosing each intermediate value between the upper and lower limits of the scope. Each smaller range between the intermediate value in any stated value or stated range and any other stated value or intermediate value in the range is also included in the present invention. The upper and lower limits of these smaller ranges may be independently included or excluded in the scope.

[0041] Unless otherwise specified, the technical / scientific terms used herein have the same meanings as those generally understood by those skilled in the art to which this application belongs. Although this application only describes preferred methods and materials, any methods and materials similar or equivalent to these may also be used in the implementation or testing of this application. All documents mentioned in this specification are incorporated by reference to disclose and describe methods and / or materials related to the documents. In the event of a conflict with any incorporated document, the content of this specification shall prevail.

[0042] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0043] Figure 1 A schematic diagram of a mine fire monitoring system provided for the implementation of the present invention is shown in FIG. Figure 1 As shown, the above-mentioned mine fire monitoring system may include:

[0044] Multiple sensors for collecting underground environment data of the target mine, the sensors at least including a temperature sensor, a smoke sensor, a CO sensor and an infrared flame sensor;

[0045] Among them, the temperature sensor, smoke sensor, CO sensor and infrared flame sensor are respectively connected to the main controller by wire or wireless means to form an efficient data transmission link, so that the corresponding data collected by each can be transmitted to the main controller in real time and stably, ensuring that the system can grasp the dynamic changes of the underground environment in real time. The selection of each sensor type and model can be based on the needs in the actual application process, and the embodiment of the present invention does not specifically limit this.

[0046] Exemplarily, each sensor is described below with a specific embodiment:

[0047] The temperature sensor is used to monitor the temperature data in the environment. For example, the temperature sensor can be the DS18B20 temperature sensor, which integrates advanced digital technology, single bus communication, high precision, fast conversion, low power consumption, and can sensitively capture temperature changes and transmit them to the main controller. The smoke sensor is used to monitor the smoke concentration. The smoke sensor can be the MQ-2 smoke sensor. Its gas sensitive resistor principle enables it to sense the resistance change caused by smoke, with a wide detection range, high sensitivity and anti-interference, and can transmit the smoke concentration data to the main controller in a complex mine environment. The CO sensor can be the MQ-7 CO sensor. It relies on electrochemical sensing technology, has high selectivity and high sensitivity to CO, low detection limit and fast response, can eliminate interference in a complex gas environment, accurately monitor the CO generated by the fire and transmit data to the main controller. The infrared flame sensor uses the principle of infrared radiation detection. Its internal detector can capture the flame infrared signal, filter the interference through the signal processing circuit, accurately judge the open fire situation and transmit the information to the main controller.

[0048] A main controller is connected to the multiple sensors and is used to receive and process the data collected by the sensors; the main controller processes the fire signal based on a combination of fuzzy logic theory and BP neural network, including:

[0049] The temperature, smoke concentration and flame signal collected by the sensor are input into a preset fuzzy logic reasoning subsystem for processing to obtain a preliminary prediction result of the fire probability;

[0050] The temperature, smoke concentration, and CO concentration signals are input into the pre-trained BP neural network subsystem for processing, and the prediction results of the probability of open fire, the probability of smoldering, and the probability of no fire are obtained respectively;

[0051] The output results of the fuzzy logic reasoning subsystem and the BP neural network subsystem are then fused and judged through fuzzy logic to obtain the final fire judgment result;

[0052] An alarm module, connected to the main controller, for sending an alarm signal when a fire occurs according to the fire determination result of the main controller;

[0053] A fire extinguishing module, connected to the main controller, for performing a fire extinguishing operation when a fire occurs according to the fire determination result of the main controller;

[0054] A display module, connected to the main controller, for displaying the environmental data collected by the multiple sensors in real time;

[0055] The communication module is connected to the main controller and is used to transmit the data collected by the multiple sensors to an external device and receive control instructions from the external device.

[0056] In some embodiments, the main controller can be a mature single-chip microcomputer product on the market. For example, an STM32 microprocessor can be used as the main controller.

[0057] The STM32 series processor is a 32-bit microcontroller based on ARM architecture produced by STMicroelectronics, which supports real-time simulation and tracing. It uses the latest and advanced architecture of ARM Cortex-M3 core, has excellent real-time performance, outstanding power consumption control, outstanding and innovative peripherals, and the maximum degree of integration, which greatly simplifies the development process and enables products to quickly enter the application market.

[0058] Considering the number of interfaces, cost and scalability required to complete the system, STM32 is selected as the main control chip, which has the advantages of high performance and low power consumption. Its GPIO, IIC, SPI, ADC, DAC and other interfaces can meet the needs of various embedded applications. The STM32F103C8T6 unit is responsible for collecting environmental data from modules such as temperature sensors and smoke sensors, processing the data and using it as a basis to control the alarm module and fire extinguishing module. The program of the main controller can be written in any existing programming language to write software control programs. For example, it can be written in C language in the Keil5-MDK environment.

[0059] In some embodiments, in the fuzzy logic reasoning subsystem, the data of the temperature sensor and the smoke sensor are fuzzified using a Gaussian membership function, and the data of the infrared flame sensor are fuzzified using a triangular membership function;

[0060] The fuzzy logic reasoning subsystem performs reasoning operations based on the fuzzified data according to preset fuzzy rules, and obtains a preliminary prediction result of the fire probability after defuzzification.

[0061] In some embodiments, in the preset BP neural network subsystem, a three-layer forward network structure and a preset BP learning algorithm are used, and the temperature, smoke concentration and CO concentration signals collected by the sensor are used as input to output the probability of open fire, the probability of smoldering and the probability of no fire.

[0062] It can be understood that the main control performs fire signal processing based on the combination of fuzzy logic theory and BP neural network. When fusing the information collected by the above sensors, it is considered that the actual monitored fire data is small and cannot represent all fire conditions. In addition, the generalization ability of the neural network is limited, and there is a lack of experience in formulating fuzzy logic reasoning rules. Therefore, there are certain limitations in using a single method for judgment. In the embodiment of the present invention, a method combining fuzzy logic theory with BP neural network is used for design. The specific processing logic can be found in Figure 2 As shown, Figure 2 Schematic diagram of the fire signal processing structure based on fuzzy neural network in an embodiment of the present invention.

[0063] It should be noted that fuzzy reasoning logic theory is a mathematical tool for dealing with fuzziness and uncertainty. It expands traditional logical reasoning methods and allows modeling and reasoning of imprecise concepts and information.

[0064] Figure 3 FIG. 1 is a schematic diagram of the structure of a fuzzy logic reasoning subsystem in an embodiment of the present invention. Figure 3 As shown in the figure, fuzzy logic reasoning mainly includes fuzzification, logical reasoning and defuzzification processes. Among them, fuzzification is used to convert precise input into membership of fuzzy sets. For example, the specific temperature and smoke concentration measurements are converted into their membership in the corresponding fuzzy sets. The reasoning process is based on fuzzy rules, and different reasoning methods can be used, such as Mamdani reasoning or Sugeno reasoning, to obtain the output fuzzy set membership based on the input membership. Defuzzification converts the fuzzy results obtained by reasoning into precise outputs, such as converting the fuzzy results into a precise value through the centroid method, maximum membership method or weighted average method, so as to make clear decisions.

[0065] Specifically, in order to fully extract the characteristic information of the fire signal, first, the key information such as temperature, smoke concentration, flame signal collected from the sensor is input into the preset fuzzy logic reasoning subsystem. In this subsystem, these data can be processed according to the preset algorithm and process. For temperature and smoke concentration, since their data characteristics have certain continuity and gradualness, the Gaussian membership function can be used for fuzzification operation. This function can convert the precise numerical value into the membership in the fuzzy set according to its physical characteristics and actual monitoring range, so as to better reflect its fuzzy concept under different fire degrees, such as dividing the temperature into fuzzy states such as low temperature, medium temperature, and high temperature, and dividing the smoke concentration into fuzzy states such as low concentration, medium concentration, and high concentration. For the data of the infrared flame sensor, considering the particularity of the flame signal, that is, the existence or non-existence of the flame is more discrete, the triangle membership function can be used for fuzzification. By setting different triangle membership functions, the flame signal can be converted into a fuzzy representation that is more in line with its characteristics, such as the existence of flame, the possible existence of flame, and the absence of flame.

[0066] Exemplarily, the expression of Gaussian membership function can be expressed by the following formula:

[0067]

[0068] Among them, a and c are constants, a determines the width of the function curve, c determines the center of the function curve, and x is the normalized fire parameter value or fire probability value such as temperature and smoke.

[0069] Figure 4 Schematic diagram of a triangular membership function in an embodiment of the present invention. In an embodiment of the present invention, the analytical expression of the triangular membership function can be expressed as:

[0070]

[0071] For example, according to the actual situation, a=b=c=0 and a=b=c=1 can be selected to correspond to the two fuzzy states of flame infrared signal (Y) and flame infrared signal (N) respectively; in the design of temperature data, a=0.2 can be selected, and the c values ​​corresponding to the fuzzy states of normal temperature (TS), slightly high temperature (TM), and too high temperature (TL) are 0, 0.5, and 1 respectively; in the design of smoke density data, a=0.2 can be selected, and the c values ​​corresponding to the fuzzy states of normal smoke density (SS), slightly high temperature (SM), and too high temperature (SL) are 0.2, 0.5, and 0.8 respectively; the fuzzy states of outputting fire conditions as no fire (PS), smoldering (PM), and open fire (PL) can be selected a=0.2, and the c values ​​in the corresponding Gaussian membership function can be 0.2, 0.7, and 1 respectively. Thus, the membership curves of flame data, temperature data, smoke density data, and fire probability can be obtained.

[0072] In some embodiments, the preset fuzzy rules may select as few rules as possible, thereby making the design and implementation of the fuzzy controller simpler.

[0073] For example, according to the establishment method and requirements of the fuzzy control rules in the above example, there are 18 fuzzy rules for the fuzzy system of fire signal processing, and the specific rules may include:

[0074] 1. If (temperature isTL) and (smoke concentration isSL) and (flame sensor data isY) then (fire probability isPL);

[0075] 2. If (temperature isTL) and (smoke concentration isSL) and (flame sensor data isN) then (fire probability isPM);

[0076] 3. If (temperature isTL) and (smoke concentration isSM) and (flame sensor data isY) then (fire probability isPL);

[0077] 4. If (temperature isTL) and (smoke concentration isSM) and (flame sensor data isN) then (fire probability isPM);

[0078] 5. If (temperature isTL) and (smoke concentration isSS) and (flame sensor data isY) then (fire probability isPM);

[0079] 6. If (temperature isTL) and (smoke concentration isSS) and (flame sensor data isN) then (fire probability isPS);

[0080] 7. If (temperature is TM) and (smoke concentration is SL) and (flame sensor data is Y) then (fire probability is PM);

[0081] 8. If (temperature is TM) and (smoke concentration is SL) and (flame sensor data is N) then (fire probability is PM);

[0082] 9. If (temperature is TM) and (smoke concentration is SM) and (flame sensor data is Y) then (fire probability is PM);

[0083] 10. If (temperature is TM) and (smoke concentration is SM) and (flame sensor data is N) then (fire probability is PM);

[0084] 11. If (temperature is TM) and (smoke concentration is SS) and (flame sensor data is Y) then (fire probability is PM);

[0085] 12. If (temperature is TM) and (smoke concentration is SS) and (flame sensor data is N) then (fire probability is PS);

[0086] 13. If (temperature isTS) and (smoke concentration isSL) and (flame sensor data isY) then (fire probability isPM);

[0087] 14. If (temperature isTS) and (smoke concentration isSL) and (flame sensor data isN) then (fire probability isPS);

[0088] 15. If (temperature isTS) and (smoke concentration isSM) and (flame sensor data isY) then (fire probability isPM);

[0089] 16. If (temperature isTS) and (smoke concentration isSM) and (flame sensor data isN) then (fire probability isPS);

[0090] 17. If (temperature isTS) and (smoke concentration isSS) and (flame sensor data isY) then (fire probability isPS);

[0091] 18. If (temperature isTS) and (smoke concentration isSS) and (flame sensor data isN) then (fire probability isPS).

[0092] In some embodiments, the fuzzy logic reasoning subsystem can obtain a preliminary prediction result of the fire probability for judging whether a fire occurs or not through defuzzification processing. In an embodiment of the present invention, a weighted average method can be used to defuzzify the output fuzzy subset.

[0093] For example, the expression of the weighted average method can be expressed as:

[0094]

[0095] Among them, h i is the element of the fuzzy subset; v i is the membership degree corresponding to the element, n is the number of fuzzy subset elements, and k is the output value after fuzzification.

[0096] In some embodiments, the BP neural network subsystem is a three-layer feed-forward network. Figure 5 FIG. 1 is a schematic diagram of the structure of a BP neural network subsystem in an embodiment of the present invention. Figure 5 As shown, the BP neural network subsystem uses the BP learning algorithm. The first layer is the input layer, the second layer is the hidden layer, and the third layer is the output layer. The collected data temperature, smoke concentration and CO concentration signals are used as input, and the output is the probability of open fire p2, the probability of smoldering p1 and the probability of no fire p0, and its value range is 0 to 1. Then the most appropriate number of hidden layer nodes is selected based on experience. The selection of the number of hidden layer neuron nodes is a complex issue, which usually needs to be determined based on the experience of the designer and multiple verifications. There is no perfect analytical expression to represent it. In the embodiment of the present invention, it can be expressed according to the formula Or n = 2b + 1, to select the best number of hidden nodes, where a is the number of output units, b is the number of input units, and n is the number of hidden units. According to the above formula, the range of the number of hidden units can be selected to be 7 to 13.

[0097] Exemplarily, by selecting existing sample data as training samples, part of the data and expected output may be shown in Table 1.

[0098] Table 1:

[0099]

[0100]

[0101] In some embodiments, by performing simulation experiments, such as simulation experiments in MATLAB, the hidden layer activation function can be selected to use the hyperbolic tangent S-type transfer function tansig, the output layer activation function is the S-type logarithmic function logsig, the training function is the traingd function, the output error and output range of the test data are calculated, and finally the number of hidden units with good convergence and small error is selected as 12. The weight between the input layer and the hidden layer is W1, and the weight between the hidden layer and the output layer is W2

[0102] In some embodiments, the correlation between the actual sample and the training result obtained after MATLAB simulation training is as follows: Figure 6 As shown, Figure 6 Schematic diagram of the correlation between the actual sample and the training result in the embodiment of the present invention. R is the correlation coefficient between the training result and the sample set, and Y=T means that the target and the output are the same. Figure 6 As shown, the prediction result curve after training has a high degree of fit with the actual sample result and a good linear correlation, which can meet the design requirements, proving that the neural network model established in the embodiment of the present invention is relatively accurate.

[0103] In some embodiments, the input data of the system is processed by the fuzzy reasoning subsystem and the BP neural network subsystem respectively, and then the fuzzy logic is used to make decisions at the information layer. The BP neural network subsystem outputs the probability of open fire p2, the probability of smoldering p1, and the probability of no fire p0, and the fuzzy logic reasoning subsystem outputs the probability of fire occurrence k. However, there are certain contradictions in the judgments made by these two subsystems.

[0104] It is understandable that when the fire probabilities output by the neural network subsystem and the fuzzy logic reasoning subsystem both determine that a fire has occurred, it can be determined that a fire has occurred; however, when the neural network system determines that there is no fire and the fuzzy logic outputs a higher probability of fire, it is difficult to make a judgment. Therefore, the embodiment of the present invention sends the results output by the two subsystems to the decision layer, and uses fuzzy logic reasoning to make a judgment again. If it cannot be judged, it delays for a few seconds before recollecting data and judging whether a fire has occurred again. The probability of fire p2, the probability of smoldering p1, and the probability of no fire p0 output by the neural network and the probability of fire occurrence k output by the fuzzy logic reasoning subsystem can be used as the input of the fuzzy logic reasoning subsystem of the decision layer to obtain the final fire judgment result.

[0105] Exemplarily, specific logic judgment rules may include:

[0106] If p2≥0.7, k≥0.7, it is judged that an open flame fire occurs;

[0107] If p2≥0.7, 0.3≤k≤0.7, the judgment will be delayed for a few seconds;

[0108] 0.3≤p2≤0.7, k≥0.7, then the judgment will be delayed for a few seconds;

[0109] If p1≥0.7 and k≥0.3, it is judged that a smoldering fire has occurred;

[0110] If p1≥0.7, k≤0.3, the judgment will be delayed for several seconds;

[0111] 0.3≤p1≤0.7, k≥0.3, then the judgment will be delayed for a few seconds;

[0112] If p0≥0.7 and k≤0.3, it is judged that no fire has occurred;

[0113] If p0≥0.7, k≥0.3, the judgment will be delayed for a few seconds.

[0114] In the embodiment of the present invention, multi-sensor information fusion technology is used to collect and process environmental information, and a fuzzy logic reasoning system is used to make judgments at the decision-making level, which integrates the advantages of fuzzy reasoning and neural networks, and improves the accuracy of fire judgment. When the system detects that a fire has occurred, it will automatically turn on the alarm module to warn the workers underground, and at the same time turn on the fire extinguishing module to extinguish the fire. When the monitoring value returns to normal, the fire extinguishing module and the alarm module automatically return to the closed state. The temperature, smoke, CO and flame sensors are used to monitor the environmental data information under the mine in real time, and the monitored data and images are displayed, and transmitted to the control center through the communication module. On the one hand, the workers can correctly judge the current situation in the environment according to the information displayed on the on-site screen, and can react according to the alarm information when a fire occurs. On the other hand, the relevant person in charge of the ground control center can also obtain the data information of the mine environment in real time, and can obtain the fire scene data information and the fire scene location in time when the environment is abnormal, and control the alarm and arrange relevant personnel to carry out fire fighting and rescue in time.

[0115] In some embodiments, the core task of the alarm module is to send out an alarm signal promptly and effectively at the moment of detecting a fire according to the fire determination result made by the main controller, so as to inform the underground workers and related personnel of the occurrence of the fire, thereby initiating corresponding emergency measures. The alarm module has a high degree of reliability and timeliness, and will respond quickly once it receives a fire signal from the main controller.

[0116] The alarm module includes but is not limited to the alarm methods of sound and light signals, and can also be expanded to other alarm methods according to actual needs, aiming to provide underground workers and ground monitoring personnel with comprehensive fire warning information. The alarm module can adopt a variety of alarm methods that match different fire determination results to achieve more targeted and differentiated early warning functions. The core is to subdivide the fire determination results into multiple situations. According to different judgment logics and results, such as the above-mentioned logical judgment rules to determine whether a fire has occurred and the specific type of fire that has occurred (such as open flame fire, smoldering fire, no fire, etc.), the system can preset different alarm methods to meet the early warning needs in different scenarios.

[0117] For example, when it is determined to be an open fire, a buzzer or other device can be used to send out a rapid high-decibel alarm, a red warning light can be flashed at a high frequency, or a voice broadcast can be used to play a warning message to guide people to evacuate. If it is a smoldering fire, the buzzer will continue to sound at a lower frequency, the warning light will flash at a low frequency (orange or yellow), and another warning message will be broadcasted by voice to remind people to pay attention to potential dangers. In the case of no fire, a short prompt tone or green light signal (green indicator light flashes briefly) can be used to indicate that the system is operating normally.

[0118] In some embodiments, when it is determined that there is a fire, the main controller can also plan an escape route based on the fire situation and the situation of people in the mine, and guide the evacuation of people in the mine through the alarm module.

[0119] In some embodiments, the alarm module may also include a manual alarm module. The manual alarm module is an additional safety measure that is directly connected to the main controller and provides a way for underground personnel to actively trigger alarms and fire extinguishing operations. When underground personnel discover abnormal conditions, but the system may not have made a fire determination, the fire extinguishing module and the alarm module can be directly controlled through the manual alarm module. When underground personnel detect fire hazards or other dangerous situations, they can immediately press the button of the manual alarm module to trigger the alarm module to send out an alarm sound and voice message, and at the same time start the fire extinguishing module to perform fire extinguishing operations, which complements the automatic alarm and fire extinguishing system, avoids the spread of fire due to system misjudgment or delay, and improves the safety and emergency response capabilities of the system.

[0120] In some embodiments, the fire extinguishing module can be connected to the water pump through a relay, which is used to turn on or off the water pump for fire extinguishing when receiving the instruction of the main controller. The relay module plays the role of signal conversion and power amplification. When the main controller issues a fire extinguishing instruction, it can convert the weak current signal into a strong current signal to drive the water pump to work. As the main fire extinguishing execution component, the water pump can transport water and other fire extinguishing media to the area where fire extinguishing is required underground. The main controller can accurately control the working state of the fire extinguishing module, and decide when to start or shut down the water pump according to the specific situation of the fire, so as to achieve precise fire extinguishing. For example, when the fire is in the early stage, the main controller can start the water pump to spray water at a small flow rate to suppress the fire; if the fire is large, the power of the water pump can be increased to improve the efficiency of fire extinguishing. At the same time, the module can also set two control modes, automatic and manual, which can not only perform fire extinguishing operations according to the automatic judgment of the main controller, but also accept manual operations under special circumstances, which improves the flexibility and reliability of fire extinguishing.

[0121] In some embodiments, the display module is an information display window of the system, which is intended to display the environmental data collected by multiple sensors in real time so that the staff can intuitively understand the underground environmental conditions. Through the display module, the operator can monitor the temperature, smoke concentration, carbon monoxide concentration and flame information in real time, so as to have a clear understanding of the safety conditions in the well.

[0122] In the embodiment of the present invention, an OLED screen can be selected as a display module, which has the advantages of high contrast, low power consumption, and wide viewing angle. The OLED screen can clearly display the data information collected by various sensors, and present this information in the form of intuitive charts, numbers or text. For example, it can display the current underground temperature, smoke concentration, carbon monoxide concentration, flame status and other information in real time, so that operators can quickly determine whether the underground environment is within the normal range. When the data is abnormal, it can assist in analyzing the possibility and development trend of the fire. The interface layout of the display module is concise and clear, which is convenient for operators to quickly view key information. At the same time, historical data can be stored and reviewed, providing data support for fire cause analysis and system optimization.

[0123] In some embodiments, the communication module is a bridge for the entire system to exchange information with the external environment. Its main function is to transmit data collected by multiple sensors to external devices, and receive control instructions from external devices to ensure that the information of the system can be transmitted in time and the system can be effectively controlled. It can be wired communication or wireless communication, or it can have wired communication and wireless communication functions at the same time. For example, in an embodiment of the present invention, an ESP8266WiFi module can be selected as a communication module to connect to the main controller.

[0124] In some embodiments, the mine fire monitoring system may further include a cloud service platform. The main controller may send the data collected by the sensor to the cloud service platform through the communication module for fire signal processing, and receive the fire determination results sent by the cloud service platform through the communication module.

[0125] As a centralized data processing and decision-making center, the cloud service platform is able to receive massive amounts of data from the main controller. Unlike local data processing, the cloud service platform has stronger computing resources and data processing capabilities, and can perform more rapid and accurate in-depth processing of fire signals. On the one hand, it can store this data and build a long-term data repository for subsequent data analysis and mining. For example, it can conduct comparative analysis of underground environmental data in different time periods and different areas to discover potential trends in fire hazards. On the other hand, the cloud service platform can use its complex algorithms to process fire signals, such as using machine learning algorithms to analyze data. Through training with a large amount of historical data, it can continuously optimize the judgment logic of fire signals and improve the ability to identify fires.

[0126] In addition, the cloud service platform can also provide remote monitoring and management functions for managers at the ground control center. Managers can log in to the cloud service platform through a web page or a dedicated application to view underground environmental data, fire determination results, historical data and other information at any time. In addition, they can send remote control instructions to the main controller through the cloud service platform, such as adjusting the monitoring parameters of the sensor, changing the alarm threshold, remotely starting or stopping the fire extinguishing module, etc., which greatly enhances the remote management and control capabilities of the mine fire monitoring system and improves the flexibility and adaptability of the entire system in responding to fires.

[0127] In some embodiments, the external device includes a remote control terminal, which can send a variety of control instructions for controlling the working status of the alarm module and / or the fire extinguishing module to the main controller by interacting with the communication module.

[0128] The remote control terminal can be in various forms, for example, it can be a computer terminal installed in the ground control center, or a mobile device application (APP) specially equipped for managers, or other special equipment with remote communication and control functions. Through these remote control terminals, managers or related operators can flexibly control the underground fire monitoring system in a safe area away from the underground operation site.

[0129] When operators use the remote control terminal, they can send various control instructions to the main controller as needed. These instructions are mainly used to control the working status of the alarm module and / or the fire extinguishing module.

[0130] For the alarm module, the remote control terminal can issue a variety of control commands. For example, the operator can issue a command to turn on or off the alarm sound through the remote control terminal. When some special circumstances occur underground and the alarm sound needs to be temporarily stopped to avoid unnecessary interference, or when the alarm function needs to be turned on again, precise operations can be achieved through the remote control terminal.

[0131] For the fire extinguishing module, the remote control terminal can also send a variety of control commands. It can control the start and shut down of the fire extinguishing module. When a fire is confirmed underground, the operator can immediately send a start command to the main controller through the remote control terminal. After receiving the command, the main controller will start the relevant equipment of the fire extinguishing module, such as turning on the water pump to perform fire extinguishing operations.

[0132] Through the interaction between the remote control terminal and the communication module, the working status of the alarm module and the fire extinguishing module is diversified and refined, which greatly improves the flexibility and operability of the coal mine fire monitoring system. At the same time, the remote control function also facilitates the maintenance and optimization of the system. Operators can remotely check, maintain and update the system without entering the mine, reducing the risk of personnel entering the mine and further improving the safety and reliability of the system.

[0133] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other. Each embodiment focuses on the differences from other embodiments.

[0134] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present application.

Claims

1. A mine fire monitoring system, characterized in that: include: Multiple sensors for collecting underground environment data of the target mine, the sensors at least including a temperature sensor, a smoke sensor, a CO sensor and an infrared flame sensor; A main controller, connected to the plurality of sensors, and configured to receive and process data collected by the sensors; The main controller processes fire signals based on a combination of fuzzy logic theory and BP neural network, including: The temperature, smoke concentration and flame signal collected by the sensor are input into a preset fuzzy logic reasoning subsystem for processing to obtain a preliminary prediction result of the fire probability; The temperature, smoke concentration, and CO concentration signals are input into the pre-trained BP neural network subsystem for processing, and the prediction results of the probability of open fire, the probability of smoldering, and the probability of no fire are obtained respectively; The output results of the fuzzy logic reasoning subsystem and the BP neural network subsystem are then fused and judged through fuzzy logic to obtain the final fire judgment result; An alarm module, connected to the main controller, for sending an alarm signal when a fire occurs according to the fire determination result of the main controller; A fire extinguishing module, connected to the main controller, for performing a fire extinguishing operation when a fire occurs according to the fire determination result of the main controller; A display module, connected to the main controller, for displaying the environmental data collected by the multiple sensors in real time; The communication module is connected to the main controller and is used to transmit the data collected by the multiple sensors to an external device and receive control instructions from the external device.

2. The mine fire monitoring system according to claim 1, characterized in that: In the fuzzy logic reasoning subsystem, the data of the temperature sensor and the smoke sensor are fuzzified using a Gaussian membership function, and the data of the infrared flame sensor are fuzzified using a triangular membership function; The fuzzy logic reasoning subsystem performs reasoning operations based on the fuzzified data according to preset fuzzy rules, and obtains a preliminary prediction result of the fire probability after defuzzification.

3. The mine fire monitoring system according to claim 2, characterized in that: In the preset BP neural network subsystem, a three-layer forward network structure and a preset BP learning algorithm are adopted, and the temperature, smoke concentration and CO concentration signals collected by the sensor are used as input to output the probability of open fire, smoldering and no fire.

4. The mine fire monitoring system according to claim 3, characterized in that: The alarm module includes but is not limited to the alarm modes of sound and light signals, and the alarm module has different alarm modes according to different fire determination results.

5. The mine fire monitoring system according to claim 4, characterized in that: The fire extinguishing module is connected to the water pump via a relay, and is used to open or close the water pump for fire extinguishing upon receiving an instruction from the main controller.

6. The mine fire monitoring system according to claim 5, characterized in that: The system also includes a cloud service platform. The main controller sends the data collected by the sensor to the cloud service platform through the communication module for fire signal processing, and receives the fire determination result sent by the cloud service platform through the communication module.

7. The mine fire monitoring system according to claim 6, characterized in that: The external device includes a remote control terminal, which can send a variety of control instructions for controlling the working status of the alarm module and / or the fire extinguishing module to the main controller by interacting with the communication module.

Images