Mining smoke intelligent early warning device, early warning system and early warning method

Abstract

The invention relates to a mine smoke intelligent early warning device, early warning system and early warning method. The device is characterized in that an aerosol measuring light chamber, a particle high-efficiency filter, a trace gas monitoring chamber and a piezoelectric micropump are sequentially mounted in an inner cavity of an outer shell from front to back; a 650nm laser source, a first detection assembly, a 450nm laser source, a second detection assembly, a 850nm laser source, a third detection assembly and a temperature sensor are respectively arranged in the aerosol measurement light chamber; a piezoelectric quartz crystal detection matrix, a quartz wafer oscillation module and a high-precision frequency measurement module are arranged in the trace gas monitoring chamber; according to the system, a central server is respectively connected with a visual client and a plurality of mining smoke intelligent alarm devices; the method comprises the following steps: monitoring a plurality of monitoring areas by using a plurality of mining smoke intelligent alarm devices; and the central server processes the monitoring data and outputs a monitoring result. The device, the system and the method can accurately monitor smoke in a working place in real time.

Description

Technical Field

[0001] The invention belongs to the technical field of smoke concentration detection, and in particular relates to an intelligent early warning device, an early warning system and an early warning method for mine smoke. Background Art

[0002] Mine fire is a severe challenge faced by the current coal mine safety field. The occurrence of mine fire will not only pose a serious threat to the safe production of coal mines, but also have a great impact on the safety of life and property of miners. Although various safety measures have been taken in recent years, the number of underground fire accidents has shown a significant downward trend, but mine fires have not been completely controlled. The causes of mine fires may come from a series of complex factors, such as friction of conveyor belts, overheating of cables, and spontaneous combustion of coal seams. Under the combined effect of these factors, whether the smoke sensing equipment can issue early warning actions in a timely and accurate manner is crucial to controlling fires and reducing losses. An effective early warning system can provide valuable time for underground workers to take effective countermeasures in a timely manner, thereby avoiding catastrophic consequences to the greatest extent possible.

[0003] At present, mining smoke sensors have been widely used in coal mines, playing a certain role in coal mine fire warning. Common sensor types include ion type, ordinary photoelectric type, gas sensitive type, etc. However, the mine environment has its own particularity, such as high coal dust concentration, significant geothermal temperature, high humidity, insufficient light, and dense electrical equipment. These factors have posed great challenges to the performance of the sensor. In addition, the invisible smoke produced during the fire process will be accompanied by many gases, such as CO, CO2 and other volatile organic compounds, and may also be accompanied by particles of unburned substances. These gases may be mixed with unburned coal dust particles to form a complex aerosol system. Therefore, the accuracy of smoke sensors using a single detection principle is bound to be affected by non-fire smoke, and it is easy to produce radioactive pollution. In actual use, there are often false alarms or missed alarms. In the process of fire detection at the mine site, the false alarm phenomenon of smoke sensors with a single detection principle is more serious when working near the working surface in a high dust environment, which brings considerable trouble to normal production work.

[0004] In view of the defects of the mine smoke detection method using a single measurement principle in the coal mine environment, there is an urgent need to provide a smoke warning device, warning system and warning method that can adapt to the complex environment of the mine. Summary of the invention

[0005] In view of the problems existing in the above-mentioned prior art, the present invention provides an intelligent early warning device, early warning system and early warning method for mine smoke. The device has a simple structure, low manufacturing cost, high intelligence and strong versatility. It can adapt to the early warning needs in the complex environment of the mine, and can ensure the accuracy and efficiency of smoke detection, which is conducive to enabling relevant personnel to adopt countermeasures in a timely and effective manner through early warning, and can effectively ensure the safe production operation of the coal mine; the system has a high degree of intelligence, and can realize real-time monitoring of smoke in multiple different monitoring areas in the workplace, and can help reduce the probability of underground fire accidents; the method has a high degree of intelligence, which can not only realize real-time monitoring and early warning of fire, but also predict the diffusion trend of smoke in advance, which is conducive to relevant personnel to take effective countermeasures in time.

[0006] In order to achieve the above-mentioned object, the present invention provides an intelligent early warning device for mining smoke, comprising an outer shell, an aerosol measurement optical chamber, a particle high efficiency filter, a pressure difference detection component, a trace gas monitoring chamber, a piezoelectric micro pump, a data acquisition module, a communication module and a microprocessor; the front end and the rear end of the outer shell are respectively provided with a main air inlet and a main air outlet connected to the inner cavity thereof; The aerosol measurement optical chamber, the particle high efficiency filter, the trace gas monitoring chamber and the piezoelectric micro pump are sequentially installed in the inner cavity of the housing from front to back; The aerosol measurement optical chamber comprises an optical chamber shell, a 650nm laser source, a first detection component, a 450nm laser source, a second detection component, an 850nm laser source, a third detection component and a temperature sensor; the optical chamber shell is made of a matte black or black light-absorbing material, and an optical chamber detection cavity is formed inside the optical chamber shell; the front end and the rear end of the optical chamber shell are respectively provided with an optical chamber inlet and an optical chamber outlet at the part corresponding to the center of the optical chamber detection cavity, and the optical chamber inlet is connected to the main air inlet on the outer shell through a main air inlet pipeline; the 650nm laser source is installed in the middle section of the inner wall on the left side of the optical chamber shell, and its laser emission direction is perpendicular to the axial direction of the optical chamber detection cavity and passes through the center of the optical chamber detection cavity; the first detection component is installed in the middle section of the inner wall on the upper side of the optical chamber shell, and the axial center of its detection surface is perpendicular to the laser emission direction of the 650nm laser source and passes through the center of the optical chamber detection cavity; the first detection component comprises a first photoelectric sensing unit and a first light trap distributed coaxially, and the first photoelectric sensing unit is located on the inner side of the first light trap; the 45 The 450nm laser source is installed at the front of the left inner wall of the optical chamber housing, and its laser emission direction passes through the center of the optical chamber detection cavity; the second detection component is installed at the rear of the right inner wall of the optical chamber housing, and its detection surface corresponds to the laser emission direction of the 450nm laser source; the second detection component includes a 450nm filter, a second photoelectric sensing unit and a second light trap coaxially distributed, and the 450nm filter, the second photoelectric sensing unit and the second light trap are distributed in sequence from the inside to the outside; the 850nm laser source is installed at the rear of the left inner wall of the optical chamber housing, and its laser emission direction passes through the center of the optical chamber detection cavity; the third detection component is installed at the front of the right inner wall of the optical chamber housing, and its detection surface corresponds to the laser emission direction of the 850nm laser source; the third detection component includes a 850nm filter, a third photoelectric sensing unit and a third light trap coaxially distributed, and the 850nm filter, the third photoelectric sensing unit and the third light trap are distributed in sequence from the inside to the outside; the temperature sensor is installed on the lower inner wall of the optical chamber housing; The front end and the rear end of the particle high efficiency filter are respectively provided with a filter inlet and a filter outlet, and the filter inlet is connected to the optical chamber outlet of the aerosol measurement optical chamber through a first connecting pipeline, and the filter outlet is connected to the inlet end of the second connecting pipeline; The differential pressure detection assembly is arranged in the inner cavity of the outer shell and is located on one side outside the particle high efficiency filter. The differential pressure detection assembly includes a pre-filter connecting pipeline, a differential pressure sensor and a post-filter connecting pipeline; the inlet end of the pre-filter connecting pipeline is connected to the first connecting pipeline, one end of the differential pressure sensor is connected to the outlet end of the pre-filter connecting pipeline, the inlet end of the post-filter connecting pipeline is connected to the other end of the differential pressure sensor, and the outlet end is connected to the second connecting pipeline; The trace gas monitoring chamber comprises a monitoring chamber shell, a piezoelectric quartz crystal detection matrix, a quartz wafer oscillation module and a high-precision frequency measurement module; a monitoring chamber is formed inside the monitoring chamber shell; a monitoring inlet and a monitoring outlet connected to the monitoring chamber are respectively provided at the front and rear ends of the monitoring chamber shell, and the monitoring inlet is connected to the outlet end of the second connecting pipeline; the piezoelectric quartz crystal detection matrix is ​​installed in the central area of ​​the monitoring chamber, and the piezoelectric quartz crystal detection matrix comprises a sensing array acquisition module, a first piezoelectric quartz crystal sensor, a second piezoelectric quartz crystal sensor and a third piezoelectric quartz crystal sensor; three sensing interfaces are arranged on the same side of the sensing array acquisition module; the first piezoelectric quartz crystal sensor, the second piezoelectric quartz crystal sensor and the third piezoelectric quartz crystal sensor are respectively attached with CO characteristic adsorption film materials, CO2 characteristic adsorption film materials and high hydrophilicity characteristic adsorption film materials, and the three piezoelectric quartz crystal sensors are respectively connected to the sensing array acquisition module through three sensing interfaces; the quartz wafer oscillation module is connected to the piezoelectric quartz crystal detection matrix; the high-precision frequency measurement module is connected to the quartz wafer oscillation module; The inlet end of the piezoelectric micro pump is connected to the monitoring outlet of the trace gas monitoring chamber through a third connecting pipeline, and the outlet end thereof is connected to the main gas outlet of the main housing through a main gas outlet pipeline; The data acquisition module, the communication module 1 and the microprocessor are all installed in the inner cavity of the outer shell; the data acquisition module is respectively connected to the first photoelectric sensing unit, the third photoelectric sensing unit, the second photoelectric sensing unit, the temperature sensor, the pressure difference sensor and the high-precision frequency measurement module; the microprocessor is respectively connected to the data acquisition module, the 650nm laser source, the 450nm laser source, the 850nm laser source, the piezoelectric micropump and the communication module 1.

[0007] In the present invention, the setting of the aerosol measurement optical chamber can facilitate the completion of the measurement of aerosols in dusty airflow. By sequentially installing a 450nm laser source, a 650nm laser source and an 850nm laser source at the front, middle and rear of the left side wall of the optical chamber, and making the laser beams emitted by the three laser sources pass through the center of the optical chamber detection cavity, at the same time, installing a first detection component on the upper side wall of the optical chamber, and installing a third detection component and a second detection component at the front and rear of the right side wall of the optical chamber, respectively, in this way, the first photoelectric sensing unit in the first detection component can be used to receive the light signals emitted by the three laser sources at different angles and scattered by the particles in the dusty airflow, the second photoelectric sensing unit in the second detection component can be used to receive the light signals emitted by the 450nm laser source and scattered by the particles in the dusty airflow, and the second photoelectric sensing unit in the third detection component can be used to receive the light signals emitted by the 850nm laser source and scattered by the particles in the dusty airflow, thereby, the reception of scattered light signals at different angles can be achieved in an all-round manner, which is helpful to fully and accurately obtain the wet particle size distribution characteristics of aerosols in the dusty airflow. A 450nm filter and an 850nm filter are respectively arranged on the inner side of the second and third photoelectric sensing units, which can only allow light with wavelengths of 450nm and 850nm to pass through, thereby ensuring the detection accuracy of the second and third photoelectric sensing units. The first, second and third light traps are respectively arranged on the outer side of the first, second and third photoelectric sensing units, which can effectively absorb the light beams passing through the corresponding photoelectric sensing units, avoid the influence of reflected light on the detection accuracy, and significantly improve the detection accuracy. By arranging a temperature sensor in the aerosol measurement optical chamber, it is convenient to collect the temperature signal in the aerosol measurement optical chamber in real time, which is conducive to the subsequent calibration of the aerosol wet particle size distribution characteristics in combination with the temperature data. A particle high efficiency filter is arranged between the aerosol measurement optical chamber and the trace gas monitoring chamber, which can effectively filter out the particulate matter in the airflow after the aerosol measurement, thereby ensuring that the airflow entering the trace gas monitoring chamber does not contain particulate matter. By installing the first, second and third piezoelectric quartz crystal sensors on the sensor array acquisition module in the trace gas monitoring room, and attaching CO characteristic adsorption film materials, CO2 characteristic adsorption film materials and high hydrophilic characteristic adsorption film materials on the three piezoelectric quartz crystal sensors, and then connecting the quartz wafer oscillation module with the sensor array acquisition module, the CO characteristic signal, CO2 characteristic signal and relative humidity signal can be synchronously collected, which is conducive to obtaining CO concentration data, CO2 concentration data and relative humidity data. By obtaining relative humidity data, the wet particle size distribution characteristics of aerosols can be further calibrated in combination with humidity information, so that the dry particle size distribution characteristic information of aerosols can be obtained, and accurate aerosol particle concentration data can be further calculated.At the same time, the CO concentration data and CO2 concentration data can also be calibrated in combination with the temperature data and the relative humidity data, so as to obtain the CO concentration correction data and the CO2 concentration correction data. A differential pressure sensor is arranged outside the particle high efficiency filter, and the two ends of the differential pressure sensor are respectively connected to the first and second connecting pipes through the connecting pipes before and after the filter, so that the differential pressure sensor can be used to collect the pressure difference data before and after the filter in real time, and then the filter element in the particle high efficiency filter can be judged according to the pressure difference data whether it needs to be replaced, thereby further ensuring the accuracy of the smoke warning. A piezoelectric micro pump is connected to the downstream side of the trace gas monitoring chamber, so that the negative pressure can be accurately provided, so that the airflow in the environment can smoothly enter the intelligent alarm device for mine smoke, further improving the accuracy of the smoke warning. Through the setting of the communication module one, it is convenient for the intelligent alarm device for mine smoke to communicate with external equipment in real time. The device has a simple structure, low manufacturing cost, high intelligence and strong versatility. It can adapt to the early warning needs in complex mine environments and ensure the accuracy and efficiency of smoke detection. It can detect mine fires in a timely and accurate manner, which is conducive to enabling relevant personnel to adopt response measures in a timely and effective manner through early warning, significantly improving the safety factor in coal mine production operations.

[0008] Furthermore, in order to promptly and effectively perform early warning actions when abnormal situations occur, an alarm component is also included, which is installed on the outside of the outer shell and includes a buzzer and a flash device; the buzzer and the flash device are both connected to the microprocessor. By providing a buzzer and a flash device at the same time, different warning combination actions can be controlled according to different warning levels, thereby realizing diversified alarm actions.

[0009] Furthermore, a pair of electrodes are mounted on the quartz crystals of the first piezoelectric quartz crystal sensor, the second piezoelectric quartz crystal sensor and the third piezoelectric quartz crystal sensor at the outer edge of the upper end surface, and nano-gold is coated on the portion other than the pair of electrodes. By setting the nano-gold, the piezoelectric effect of the quartz crystal and the special physical and chemical properties of the nano-gold can be fully utilized, thereby fully improving the sensitivity and accuracy of the detection.

[0010] Furthermore, in order to improve detection accuracy, the main air inlet of the outer shell and the light chamber inlet of the aerosol measurement light chamber are located at the same horizontal plane, and the main air outlet of the outer shell and the outlet end of the piezoelectric micropump are located at the same horizontal plane.

[0011] Furthermore, in order to prevent the inner surface of the optical chamber housing from being adsorbed with solid particles during the detection process, the inner surface of the optical chamber housing is coated with a non-adsorbent material with anti-adhesion properties.

[0012] Furthermore, in order to fully save energy consumption, the temperature sensor is a low-power consumption sensor.

[0013] Furthermore, in order to facilitate the real-time display of alarm information and other reminder information, a display screen is also included, which is installed on the surface of the outer shell and connected to the microprocessor.

[0014] The present invention also provides a mine smoke intelligent early warning system, including a smoke alarm real-time monitoring network, the smoke alarm real-time monitoring network includes a plurality of mine smoke intelligent early warning devices, and also includes a central server and a visualization client; A plurality of intelligent smoke alarm devices for mining are respectively arranged in different monitoring areas of the workplace to be monitored, and are connected to the central server via a communication link between the communication module 1 and the communication module 2; The central server includes a smoke alarm monitoring database, a data processing unit and a communication module 2, wherein the data processing unit is connected to the smoke alarm monitoring database and the communication module 2 respectively, and the data processing unit is connected to the smoke alarm real-time monitoring network respectively through the communication module 2; The visualization client is connected to the data processing unit via the communication module 2.

[0015] In the present invention, by arranging a plurality of intelligent smoke alarm devices for mining in different monitoring areas of the workplace to be monitored, it is convenient to simultaneously perform synchronous monitoring operations on different monitoring areas in the workplace. By connecting a central server with a plurality of intelligent smoke alarm devices for mining, it is convenient for the central server to perform real-time interactive communication with the plurality of intelligent smoke alarm devices for mining. By setting a smoke alarm monitoring database in the central server, it is convenient to use the smoke alarm monitoring database to store the three-dimensional information of the workplace to be monitored, the position information of a plurality of intelligent smoke alarm devices for mining, and the characteristic information of the dry particle size distribution of the aerosol obtained by monitoring, the concentration of aerosol particles, the CO concentration correction data and the CO2 concentration correction data, so as to facilitate the data processing unit to read the information stored in the smoke alarm monitoring database in real time. In this way, the data processing unit can not only integrate various monitoring data to form a comprehensive data view, but also extract characteristic data that is helpful for fire identification from various monitoring data, so as to quickly and accurately obtain fire warning result data, possible causes of fire and preliminary judgment information of the type of combustion materials. Furthermore, the data processing unit can also extract characteristic data that is helpful for smoke diffusion identification from various monitoring data, so as to quickly and accurately obtain smoke diffusion trend data and potential high-risk area warning information. By connecting the visualization client to the central database, it is possible to display the location information of multiple intelligent smoke alarm devices for mines, the three-dimensional information of the workplace to be monitored, the fire warning result data, the possible cause of the fire, the type of burning material, the trend data of smoke diffusion and the potential high-risk area warning information in real time through the visualization client, so that the managers of the monitoring center can timely understand the warning information of each monitoring area. The system has a high degree of intelligence, which can realize the real-time monitoring of smoke in multiple different monitoring areas in the workplace, and can help reduce the probability of underground fire accidents.

[0016] The present invention also provides a method for an intelligent early warning system for mining smoke, which adopts an intelligent early warning system for mining smoke, comprising the following steps: Step 1: Arrange multiple intelligent smoke alarm devices for mines in different monitoring areas of the workplace to be monitored, and record the location information of the multiple intelligent smoke alarm devices for mines. At the same time, store the location information of the multiple intelligent smoke alarm devices for mines and the three-dimensional information of the workplace to be monitored in the smoke alarm monitoring database of the central server; Step 2: Use multiple intelligent smoke alarm devices for mining to monitor smoke concentration in multiple monitoring areas respectively, and send the monitoring data to the central server in real time; S21: Control the piezoelectric micro pump to start working, and use the piezoelectric micro pump to provide negative pressure so that air in the environment enters the intelligent early warning device for mining smoke; S22: Performing a continuous three-stage detection operation on the dust-containing gas passing through the aerosol measurement light chamber. At the same time, the temperature signal in the aerosol measurement light chamber is collected in real time through the temperature sensor and sent to the data acquisition module; In the first detection stage, the 650nm laser source is controlled to start working, and the 650nm laser source is used to emit a first laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit is controlled to start working, and the first photoelectric sensing unit is used to receive the first laser light signal A scattered by the dust-containing gas, and the first laser light signal A is converted into a first laser electrical signal A, and then sent to the data acquisition module; in the second detection stage, the 650nm laser source is controlled to stop working, the 450nm laser source is controlled to start working, and the 450nm laser source is used to emit a second laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit is controlled to start working, and the second photoelectric sensing unit is controlled to start working, and the second photoelectric sensing unit is controlled to start working, and the second laser light signal B scattered by the dust-containing gas is received by the second photoelectric sensing unit, and the second laser light signal B is converted into a second laser electrical signal B, and then sent to the data acquisition module, and the first The photoelectric sensing unit receives the second laser light signal C scattered by the dust-containing gas, and converts the second laser light signal C into a second laser electrical signal C, and then sends it to the data acquisition module; in the third detection stage, the 650nm laser source is controlled to stop working, the 450nm laser source is controlled to stop working, and the 850nm laser source is controlled to start working, and the 850nm laser source is used to emit a third laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit is controlled to start working, and the third photoelectric sensing unit is controlled to start working, and the third photoelectric sensing unit is used to receive the third laser light signal D scattered by the dust-containing gas, and the third laser light signal D is converted into a third laser electrical signal D, and then sent to the data acquisition module, and the first photoelectric sensing unit is used to receive the third laser light signal E scattered by the dust-containing gas, and the third laser light signal E is converted into a third laser electrical signal E, and then sent to the data acquisition module; S23: The airflow entering the trace gas monitoring chamber is filtered through a particle high efficiency filter to filter out particulate matter in the airflow. At the same time, the CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics in the airflow are respectively collected by a quartz chip oscillation module through a piezoelectric quartz crystal detection matrix, and sent to a high-precision frequency measurement module. The high-precision frequency measurement module obtains a CO concentration signal, a CO2 concentration signal and a relative humidity signal according to the CO frequency characteristics, the CO2 frequency characteristics and the relative humidity frequency characteristics, and sends them to a data acquisition module; At the same time, the pressure difference signal before and after the airflow passes through the particle high efficiency filter is detected by the pressure difference sensor and sent to the data acquisition module; S24: The data acquisition module sends the received first laser electrical signal A, second laser electrical signal B, second laser electrical signal C, third laser electrical signal D, third laser electrical signal E, temperature signal, CO concentration signal, CO2 concentration signal, relative humidity signal and pressure difference signal to the microprocessor; The microprocessor uses the built-in light scattering sensor signal correction model, and combines the first laser electrical signal A and the receiving angle of the first photoelectric sensing unit to the 650nm laser source to obtain the 650nm laser electrical signal, combines the second laser electrical signal B, the second laser electrical signal C and the receiving angle of the first photoelectric sensing unit to the 450nm laser source to obtain the 450nm laser electrical signal, combines the third laser electrical signal D, the third laser electrical signal E and the receiving angle of the first photoelectric sensing unit to the 850nm laser source to obtain the 850nm laser electrical signal, and then based on the threshold range calibrated by the standard material and the particle size classification algorithm, the 650nm laser electrical signal is respectively The voltage pulse frequency and amplitude signals of the 650nm laser electrical signal, the 450nm laser electrical signal and the 850nm laser electrical signal are analyzed and processed to obtain the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition, and further cross-check and supplement the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition to obtain the aerosol wet particle size distribution characteristic information; At the same time, the microprocessor obtains the temperature data of the dust-laden airflow according to the temperature signal, obtains the CO concentration data and the CO2 concentration data according to the CO concentration signal and the CO2 concentration signal, respectively, and obtains the relative humidity data of the airflow according to the relative humidity signal; the microprocessor uses the built-in trace gas quantitative detection and correction model, and calibrates the aerosol wet particle size distribution characteristic information in combination with the temperature data and the relative humidity data to obtain the aerosol dry particle size distribution characteristic information, and then calculates the aerosol particle concentration, CO concentration correction data and CO2 concentration correction data based on the aerosol dry particle size distribution characteristic information, and then performs time series processing on the aerosol dry particle size distribution characteristic information, the aerosol particle concentration, the CO concentration correction data and the CO2 concentration correction data, and then sends them to the central server; At the same time, the microprocessor obtains pressure difference data according to the pressure difference signal, and compares the pressure difference data with the set pressure difference threshold. When the pressure difference data is greater than or equal to the set pressure difference threshold, the flash device in the alarm component is controlled to alarm, and at the same time, the display screen is controlled to display a reminder message for replacing the particle high efficiency filter element; Step 3: After receiving the alarm signal, the data processing unit sends the location information of the mining smoke intelligent alarm device that issued the alarm to the smoke alarm monitoring database for storage. At the same time, after receiving the aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, it sends them to the smoke alarm monitoring database for storage; At the same time, the data processing unit uses the built-in multivariate fusion smoke alarm recognition model to denoise the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and integrates the denoised data to form a comprehensive data view, and then sends it to the visualization client for real-time display; at the same time, the denoised data is used as the input data of the multivariate fusion smoke alarm recognition model, and the multivariate fusion smoke alarm recognition model extracts characteristic data from the input data that is helpful for fire identification, and then analyzes and processes the characteristic data and outputs fire warning result data. When the fire warning result data contains fire conditions in one or several monitoring areas, it outputs the possible cause of the fire and the preliminary judgment information of the type of burning material, and sends a fire warning signal, the possible cause of the fire and the preliminary judgment information of the type of burning material to the mining smoke intelligent alarm device in the corresponding monitoring area; At the same time, after receiving the fire warning signal, the microprocessor in the intelligent smoke alarm device for mining controls the buzzer and flash device in the alarm component to perform sound and light alarm actions at the same time, and controls the display screen to display the possible cause of the fire and the preliminary judgment information of the type of burning material; At the same time, the data processing unit uses the built-in smoke distribution spatiotemporal evolution model to preprocess the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and then uses the preprocessed data as the input data of the smoke distribution spatiotemporal evolution model. The smoke distribution spatiotemporal evolution model extracts characteristic data from the input data that is helpful for smoke diffusion identification, and then analyzes and processes the characteristic data and outputs smoke diffusion trend data and potential high-risk area warning information; Step 4: The data processing unit sends the location information of multiple mine smoke intelligent alarm devices, the three-dimensional information of the workplace to be monitored, the fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information to the visualization client, and the visualization client realizes the real-time display of the three-dimensional information of the workplace to be monitored and the location information of multiple mine smoke intelligent alarm devices. At the same time, based on the fire warning result data, the alarm position is displayed in the three-dimensional information of the workplace to be monitored. At the same time, the received fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information are displayed in real time.

[0017] As a preferred embodiment, in step one, after the smoke alarm monitoring database receives the location information of multiple mine smoke intelligent alarm devices and the three-dimensional information of the workplace, it is stored in the mine smoke intelligent alarm device location information storage area and the workplace three-dimensional information storage area respectively; in step three, after the smoke alarm monitoring database receives the CO concentration correction data, CO2 concentration correction data, aerosol dry particle size distribution characteristic information and aerosol particle concentration, it is stored in the CO concentration correction data storage area, CO2 concentration correction data storage area, aerosol dry particle size distribution characteristic information storage area and aerosol particle concentration storage area respectively.

[0018] Compared with the prior art, the present invention has the following advantages: First, the aerosol measurement part uses three different wavelengths of laser sources: 650nm, 450nm and 850nm. Together with the corresponding three photoelectric sensing units, different types of aerosol particles can be captured more comprehensively. This multi-wavelength laser scattering technology can provide more detailed information about the size, distribution and concentration of aerosol particles, thereby significantly improving the ability to identify smoke types, accurately and efficiently identifying aerosol particle characteristics, helping to reduce false alarms and missed alarms, and providing reliable technical support for accurate and efficient smoke warnings.

[0019] Secondly, in the trace gas monitoring part, by using a specific adsorption film material to cover the piezoelectric quartz crystal sensor, the detection specificity and sensitivity of specific gases are significantly enhanced, and the CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics can be accurately obtained, so that the CO concentration, CO2 concentration and relative humidity in the gas can be accurately determined. The microprocessor is equipped with a light scattering sensor signal correction model and a trace gas quantitative detection correction model. The light scattering sensor signal correction model can be used to cross-check and supplement the aerosol particle size characteristic distribution information determined by the three laser signals, and the aerosol wet particle size distribution characteristic information can be accurately determined. At the same time, the trace gas quantitative detection correction model can be used to correct the aerosol wet particle size distribution characteristic information under different environmental conditions to obtain accurate aerosol dry particle size distribution characteristic information, which can significantly improve the accuracy of the measurement results and provide reliable data support for the early identification of fire sources.

[0020] Furthermore, the central server is used as the core processing part of the data, which can receive the monitoring data of multiple monitoring areas in real time, and can quickly integrate and efficiently analyze and process the monitoring data. The central server uses a multi-fusion smoke alarm recognition model to extract feature data that is helpful for fire identification from the monitoring data, and then can accurately identify complex fire scenes through data fusion technology, and can obtain fire warning result data, possible causes of fire and preliminary judgment information on the type of burning materials in real time. At the same time, the smoke distribution spatiotemporal evolution model is used to extract feature data that is helpful for smoke diffusion identification from the monitoring data, and the trend data of smoke diffusion and potential high-risk area warning information can be obtained. In this way, the use of a multi-fusion smoke alarm recognition model and a smoke distribution spatiotemporal evolution model can not only monitor and alarm fires in real time, but also predict the diffusion path of smoke and potential high-risk areas. This prediction capability is not available in traditional smoke alarm systems, and is of great significance for taking countermeasures in advance and planning evacuation routes.

[0021] Based on the in-depth analysis of the aerosol composition of mine smoke, the present invention fully combines technical means such as multi-wavelength laser scattering and trace gas monitoring, and integrates a variety of sensor data through data fusion technology. The data is analyzed through a deep learning algorithm, and the fire scene can be more accurately identified in the complex environment of the mine, and the trend of smoke diffusion can be predicted. The method has a high degree of intelligence, which can not only realize real-time monitoring and early warning of fires, but also predict the diffusion trend of smoke in advance, which is conducive to relevant personnel to take effective countermeasures in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural schematic diagram of the intelligent early warning device for mining smoke in the present invention; Figure 2 It is a structural schematic diagram of the aerosol measurement optical chamber in the present invention; Figure 3 This is a circuit principle block diagram of the intelligent early warning device for mining smoke in the present invention; Figure 4 It is a circuit principle block diagram of the intelligent early warning system for mine smoke in the present invention.

[0023] In the figure: 1. outer shell, 2. aerosol measurement optical chamber, 3. trace gas monitoring chamber, 4. particle high efficiency filter, 5. piezoelectric micro pump, 6. data acquisition module, 7. microprocessor, 8. communication module 1, 9. optical chamber shell, 10. 650nm laser source, 11. 450nm laser source, 12. 850nm laser source, 13. first photoelectric sensing unit, 14. first optical trap, 15. third photoelectric sensing unit, 16. third optical trap, 17. second photoelectric sensing unit, 18. second optical trap, 19. temperature sensor, 20 , 450nm filter, 21, 850nm filter, 22, monitoring room housing, 23, first piezoelectric quartz crystal sensor, 24, second piezoelectric quartz crystal sensor, 25, third piezoelectric quartz crystal sensor, 26, sensor array acquisition module, 27, quartz chip oscillation module, 28, high-precision frequency measurement module, 29, buzzer, 30, flash device, 31, differential pressure sensor, 32, pre-filter connecting pipeline, 33, post-filter connecting pipeline, 34, first connecting pipeline, 35, second connecting pipeline, 36, third connecting pipeline. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings.

[0025] like Figures 1 to 4 As shown, the present invention provides an intelligent early warning device for mining smoke, comprising an outer shell 1, an aerosol measurement optical chamber 2, a particle high efficiency filter 4, a pressure difference detection component, a trace gas monitoring chamber 3, a piezoelectric micro pump 5, a data acquisition module 6, a communication module 8 and a microprocessor 7; the front end and the rear end of the outer shell 1 are respectively provided with a main air inlet and a main air outlet connected to the inner cavity thereof; The aerosol measurement optical chamber 2, the particle high efficiency filter 4, the trace gas monitoring chamber 3 and the piezoelectric micro pump 5 are sequentially installed in the inner cavity of the outer shell 1 from front to back; The aerosol measurement optical chamber 2 comprises an optical chamber housing 9, a 650nm laser source 10, a first detection component, a 450nm laser source 11, a second detection component, an 850nm laser source 12, a third detection component and a temperature sensor 19; the optical chamber housing 9 is made of a matte black or black light-absorbing material, and an optical chamber detection cavity is formed inside the optical chamber housing 9; the front end and the rear end of the optical chamber housing 9 are respectively provided with an optical chamber inlet and an optical chamber outlet at the part corresponding to the center of the optical chamber detection cavity, and the optical chamber inlet is connected to the main air inlet on the outer shell 1 through the main air inlet pipeline; the 650nm The 650nm laser source 10 is installed in the middle section of the left inner wall of the optical chamber housing 9, and its laser emission direction is perpendicular to the axial direction of the optical chamber detection cavity and passes through the center of the optical chamber detection cavity; the first detection component is installed in the middle section of the upper inner wall of the optical chamber housing 9, and the axis of its detection surface is perpendicular to the laser emission direction of the 650nm laser source 10 and passes through the center of the optical chamber detection cavity; the first detection component includes a first photoelectric sensing unit 13 and a first light trap 14 distributed coaxially, and the first photoelectric sensing unit 13 is located on the inner side of the first light trap 14; the 450nm laser source 11 is installed at the front of the left inner wall of the optical chamber housing 9, and its laser emission direction passes through the center of the optical chamber detection cavity; the second detection component is installed at the rear of the right inner wall of the optical chamber housing 9, and its detection surface corresponds to the laser emission direction of the 450nm laser source 11; the second detection component includes a 450nm filter 20, a second photoelectric sensing unit 17 and a second light trap 18 coaxially distributed, and the 450nm filter 20, the second photoelectric sensing unit 17 and the second light trap 18 are distributed in sequence from the inside to the outside; the 850nm laser source 12 is installed in the optical chamber The rear part of the left inner wall of the housing 9, the laser emission direction of which passes through the center of the optical chamber detection cavity; the third detection component is installed on the front part of the right inner wall of the optical chamber housing 9, and its detection surface corresponds to the laser emission direction of the 850nm laser source 12; the third detection component includes a 850nm filter 21, a third photoelectric sensing unit 15 and a third light trap 16 coaxially distributed, and the 850nm filter 21, the third photoelectric sensing unit 15 and the third light trap 16 are distributed in sequence from the inside to the outside; the temperature sensor 19 is installed on the lower inner wall of the optical chamber housing 9; The front end and the rear end of the particle high efficiency filter 4 are respectively provided with a filter inlet and a filter outlet, and the filter inlet is connected to the optical chamber outlet of the aerosol measurement optical chamber 2 through a first connecting pipe 34, and the filter outlet is connected to the inlet end of the second connecting pipe 35; The differential pressure detection assembly is arranged in the inner cavity of the outer shell 1 and is located on one side outside the particle high efficiency filter 4. The differential pressure detection assembly includes a pre-filter connecting pipeline 32, a differential pressure sensor 31 and a post-filter connecting pipeline 33; the inlet end of the pre-filter connecting pipeline 32 is connected to the first connecting pipeline 34, one end of the differential pressure sensor 31 is connected to the outlet end of the pre-filter connecting pipeline 32, the inlet end of the post-filter connecting pipeline 33 is connected to the other end of the differential pressure sensor 31, and the outlet end thereof is connected to the second connecting pipeline 35; The trace gas monitoring chamber 3 includes a monitoring chamber housing 22, a piezoelectric quartz crystal detection matrix, a quartz crystal oscillation module 27 and a high-precision frequency measurement module 28; a monitoring chamber is formed inside the monitoring chamber housing 22; a monitoring inlet and a monitoring outlet connected to the monitoring chamber are respectively provided at the front and rear ends of the monitoring chamber housing 22, and the monitoring inlet is connected to the outlet end of the second connecting pipe 35; the piezoelectric quartz crystal detection matrix is ​​installed in the central area of ​​the monitoring chamber, and the piezoelectric quartz crystal detection matrix includes a sensor array acquisition module 26, a first piezoelectric quartz crystal sensor 23, a second piezoelectric quartz crystal sensor 24 and a third piezoelectric quartz crystal sensor 25; three sensor array acquisition modules are arranged on the same side of the sensor array acquisition module 26. sensing interface; the first piezoelectric quartz crystal sensor 23, the second piezoelectric quartz crystal sensor 24 and the third piezoelectric quartz crystal sensor 25 are respectively attached with CO characteristic adsorption film materials, CO2 characteristic adsorption film materials and high hydrophilicity characteristic adsorption film materials, and the three piezoelectric quartz crystal sensors are respectively connected to the sensing array acquisition module 26 through three sensing interfaces; the quartz chip oscillation module 27 is connected to the piezoelectric quartz crystal detection matrix; the high-precision frequency measurement module 28 is connected to the quartz chip oscillation module 27; by attaching gas efficient adsorption film materials with different characteristics on the three piezoelectric quartz crystal sensors, it is convenient to collect multiple characteristic frequencies at the same time, and then the monitoring data of multiple objects can be obtained at the same time.

[0026] The inlet end of the piezoelectric micro pump 5 is connected to the monitoring outlet of the trace gas monitoring chamber 3 through the third connecting pipeline 36, and the outlet end thereof is connected to the main gas outlet of the main housing 1 through the main gas outlet pipeline; The data acquisition module 6, the communication module 8 and the microprocessor 7 are all installed in the inner cavity of the outer shell 1; the data acquisition module 6 is respectively connected to the first photoelectric sensing unit 13, the third photoelectric sensing unit 15, the second photoelectric sensing unit 17, the temperature sensor 19, the pressure difference sensor 31 and the high-precision frequency measurement module 28; the microprocessor 7 is respectively connected to the data acquisition module 6, the 650nm laser source 10, the 450nm laser source 11, the 850nm laser source 12, the piezoelectric micro pump 5 and the communication module 8. As a further preferred embodiment, a power module may also be included, which is used to supply electricity to various electrical components of the intelligent early warning device for mining smoke.

[0027] In the present invention, the setting of the aerosol measurement optical chamber can facilitate the completion of the measurement of aerosols in dusty airflow. By sequentially installing a 450nm laser source, a 650nm laser source and an 850nm laser source at the front, middle and rear of the left side wall of the optical chamber, and making the laser beams emitted by the three laser sources pass through the center of the optical chamber detection cavity, at the same time, installing a first detection component on the upper side wall of the optical chamber, and installing a third detection component and a second detection component at the front and rear of the right side wall of the optical chamber, respectively, in this way, the first photoelectric sensing unit in the first detection component can be used to receive the light signals emitted by the three laser sources at different angles and scattered by the particles in the dusty airflow, the second photoelectric sensing unit in the second detection component can be used to receive the light signals emitted by the 450nm laser source and scattered by the particles in the dusty airflow, and the second photoelectric sensing unit in the third detection component can be used to receive the light signals emitted by the 850nm laser source and scattered by the particles in the dusty airflow, thereby, the reception of scattered light signals at different angles can be achieved in an all-round manner, which is helpful to fully and accurately obtain the wet particle size distribution characteristics of aerosols in the dusty airflow. A 450nm filter and an 850nm filter are respectively arranged on the inner side of the second and third photoelectric sensing units, which can only allow light with wavelengths of 450nm and 850nm to pass through, thereby ensuring the detection accuracy of the second and third photoelectric sensing units. The first, second and third light traps are respectively arranged on the outer side of the first, second and third photoelectric sensing units, which can effectively absorb the light beams passing through the corresponding photoelectric sensing units, avoid the influence of reflected light on the detection accuracy, and significantly improve the detection accuracy. By arranging a temperature sensor in the aerosol measurement optical chamber, it is convenient to collect the temperature signal in the aerosol measurement optical chamber in real time, which is conducive to the subsequent calibration of the aerosol wet particle size distribution characteristics in combination with the temperature data. A particle high efficiency filter is arranged between the aerosol measurement optical chamber and the trace gas monitoring chamber, which can effectively filter out the particulate matter in the airflow after the aerosol measurement, thereby ensuring that the airflow entering the trace gas monitoring chamber does not contain particulate matter. By installing the first, second and third piezoelectric quartz crystal sensors on the sensor array acquisition module in the trace gas monitoring room, and attaching CO characteristic adsorption film materials, CO2 characteristic adsorption film materials and high hydrophilic characteristic adsorption film materials on the three piezoelectric quartz crystal sensors, and then connecting the quartz wafer oscillation module with the sensor array acquisition module, the CO characteristic signal, CO2 characteristic signal and relative humidity signal can be synchronously collected, which is conducive to obtaining CO concentration data, CO2 concentration data and relative humidity data. By obtaining relative humidity data, the wet particle size distribution characteristics of aerosols can be further calibrated in combination with humidity information, so that the dry particle size distribution characteristic information of aerosols can be obtained, and accurate aerosol particle concentration data can be further calculated.At the same time, the CO concentration data and CO2 concentration data can also be calibrated in combination with the temperature data and the relative humidity data, so as to obtain the CO concentration correction data and the CO2 concentration correction data. A differential pressure sensor is arranged outside the particle high efficiency filter, and the two ends of the differential pressure sensor are respectively connected to the first and second connecting pipes through the connecting pipes before and after the filter, so that the differential pressure sensor can be used to collect the pressure difference data before and after the filter in real time, and then the filter element in the particle high efficiency filter can be judged according to the pressure difference data whether it needs to be replaced, thereby further ensuring the accuracy of the smoke warning. A piezoelectric micro pump is connected to the downstream side of the trace gas monitoring chamber, so that the negative pressure can be accurately provided, so that the airflow in the environment can smoothly enter the intelligent alarm device for mine smoke, further improving the accuracy of the smoke warning. Through the setting of the communication module one, it is convenient for the intelligent alarm device for mine smoke to communicate with external equipment in real time. The device has a simple structure, low manufacturing cost, high intelligence and strong versatility. It can adapt to the early warning needs in complex mine environments and ensure the accuracy and efficiency of smoke detection. It can detect mine fires in a timely and accurate manner, which is conducive to enabling relevant personnel to adopt response measures in a timely and effective manner through early warning, significantly improving the safety factor in coal mine production operations.

[0028] In order to promptly and effectively perform early warning actions when abnormal situations occur, an alarm component is also included, which is installed on the outside of the outer shell 1 and includes a buzzer 29 and a flash device 30; the buzzer 29 and the flash device 30 are both connected to the microprocessor 7. By providing a buzzer and a flash device at the same time, different warning combination actions can be controlled according to different warning levels, thereby realizing diversified alarm actions.

[0029] A pair of electrodes are relatively mounted on the outer edge of the upper end surface of the quartz wafer of the first piezoelectric quartz crystal sensor 23, the second piezoelectric quartz crystal sensor 24 and the third piezoelectric quartz crystal sensor 25, and nano-gold is coated on the portion other than the pair of electrodes. Preferably, each pair of electrodes is symmetrical about the center of the corresponding quartz wafer. By setting the nano-gold, the piezoelectric effect of the quartz crystal and the special physical and chemical properties of the nano-gold can be fully utilized, thereby fully improving the sensitivity and accuracy of the detection.

[0030] In order to improve the detection accuracy, the main air inlet of the outer shell 1 and the optical chamber inlet of the aerosol measurement optical chamber 2 are located at the same horizontal plane, and the main air outlet of the outer shell 1 and the outlet end of the piezoelectric micro pump 5 are located at the same horizontal plane.

[0031] In order to prevent the inner surface of the optical chamber housing from being adsorbed with solid particles during the detection process, the inner surface of the optical chamber housing 9 is coated with a non-adsorbent material with anti-adhesion properties.

[0032] In order to fully save energy consumption, the temperature sensor 19 is a low power consumption sensor.

[0033] In order to facilitate the real-time display of alarm information and other reminder information, a display screen is also included. The display screen is installed on the surface of the outer shell 1 and is connected to the microprocessor 7.

[0034] The present invention also provides a mine smoke intelligent early warning system, including a smoke alarm real-time monitoring network, the smoke alarm real-time monitoring network includes a plurality of mine smoke intelligent early warning devices, and also includes a central server and a visualization client; A plurality of intelligent smoke alarm devices for mining are respectively arranged in different monitoring areas of the workplace to be monitored, and are connected to the central server via a communication link between the communication module 1 and the communication module 2; The central server includes a smoke alarm monitoring database, a data processing unit and a communication module 2, wherein the data processing unit is connected to the smoke alarm monitoring database and the communication module 2 respectively, and the data processing unit is connected to the smoke alarm real-time monitoring network respectively through the communication module 2; The visualization client is connected to the data processing unit via the communication module 2.

[0035] As a preferred embodiment, the microprocessor 7 in the intelligent early warning device for mining smoke has a built-in light scattering sensor signal correction model and a trace gas quantitative detection correction model, wherein the light scattering sensor signal correction model is established based on a large number of experiments, and the voltage pulse frequency and amplitude signals obtained by the photoelectric sensing unit measuring the 650nm laser source 10, the 450nm laser source 11, and the 850nm laser source 12, and the first photoelectric sensing unit 13 receiving the three laser signals at different angles, and the threshold range calibrated by the standard material is combined with the particle size classification algorithm to obtain the pulse signals generated by the 650nm laser, 450nm laser source 11, and 850nm laser source 12. The pulse signal generated by the 50nm laser and the pulse signal generated by the 850nm laser are processed, and the characteristic distribution of the aerosol particle size determined by the three signals is cross-checked and supplemented to determine the characteristics of the aerosol wet particle size distribution; wherein, the trace gas quantitative detection correction model is established based on a large number of experiments, and the gas concentration correction is performed to obtain high-precision gas concentration data by measuring the characteristics of the first piezoelectric quartz crystal sensor 23 attached with CO characteristic adsorption film material and the second piezoelectric quartz crystal sensor 24 attached with CO2 characteristic adsorption film material responding to the trace gas concentration under different temperature and humidity working conditions; As a preferred embodiment, a multivariate fusion smoke alarm recognition model and a smoke distribution spatiotemporal evolution model are built into the data processing unit of the central server, wherein the multivariate fusion smoke alarm recognition model is established based on a large number of experiments, and the multivariate fusion smoke alarm recognition model uses data fusion technology to extract data features that are helpful for fire recognition from input data, and uses deep learning technology to build a classifier. The classifier learns how to implement time series monitoring data analysis and how to use historical data to learn how to distinguish between fire and non-fire situations based on the collected feature information during the training phase, and identifies the possible causes of the fire and the preliminary judgment of the type of burning material; the trained multivariate fusion smoke alarm recognition model can perform real-time analysis based on real-time input data, and output a prediction of whether there is a fire. As a preferred embodiment, the multivariate fusion smoke alarm recognition model can collect feedback data in the recognition process in real time, including false alarms and missed alarms, so as to continuously adjust and optimize model parameters to ensure the accuracy of recognition. This continuous learning and adaptation mechanism ensures that the model can continue to improve over time, and thus can adapt to changing environmental conditions and needs. Among them, the spatiotemporal evolution model of smoke distribution is established based on massive experiments. The spatiotemporal evolution model of smoke distribution mainly uses the partial differential equation based on physics to describe the dynamic changes of smoke in space and time. The equation takes into account the convection and diffusion behavior of smoke and combines environmental factors such as wind direction and wind speed. The numerical solution of the equation is obtained by finite element method or finite difference method to ensure that the smoke distribution at each time point can be accurately simulated. At the same time, data assimilation technology is used to integrate the model output with real-time monitoring data, optimize model parameters and state variables, and improve the accuracy of model prediction. Finally, the spatiotemporal evolution model of smoke distribution integrates machine learning algorithms to analyze the propagation pattern of smoke. The model obtained through training continuously updates and predicts the future state of smoke, so that the diffusion trend of smoke can be monitored in real time and early warning can be provided for potential high-risk areas.

[0036] In the present invention, by arranging a plurality of intelligent smoke alarm devices for mining in different monitoring areas of the workplace to be monitored, it is convenient to simultaneously perform synchronous monitoring operations on different monitoring areas in the workplace. By connecting a central server with a plurality of intelligent smoke alarm devices for mining, it is convenient for the central server to perform real-time interactive communication with the plurality of intelligent smoke alarm devices for mining. By setting a smoke alarm monitoring database in the central server, it is convenient to use the smoke alarm monitoring database to store the three-dimensional information of the workplace to be monitored, the position information of a plurality of intelligent smoke alarm devices for mining, and the characteristic information of the dry particle size distribution of the aerosol obtained by monitoring, the concentration of aerosol particles, the CO concentration correction data and the CO2 concentration correction data, so as to facilitate the data processing unit to read the information stored in the smoke alarm monitoring database in real time. In this way, the data processing unit can not only integrate various monitoring data to form a comprehensive data view, but also extract characteristic data that is helpful for fire identification from various monitoring data, so as to quickly and accurately obtain fire warning result data, possible causes of fire and preliminary judgment information of the type of combustion materials. Furthermore, the data processing unit can also extract characteristic data that is helpful for smoke diffusion identification from various monitoring data, so as to quickly and accurately obtain smoke diffusion trend data and potential high-risk area warning information. By connecting the visualization client to the central database, it is possible to display the location information of multiple intelligent smoke alarm devices for mines, the three-dimensional information of the workplace to be monitored, the fire warning result data, the possible cause of the fire, the type of burning material, the trend data of smoke diffusion and the potential high-risk area warning information in real time through the visualization client, so that the managers of the monitoring center can timely understand the warning information of each monitoring area. The system has a high degree of intelligence, which can realize the real-time monitoring of smoke in multiple different monitoring areas in the workplace, and can help reduce the probability of underground fire accidents.

[0037] The present invention also provides a method for an intelligent early warning system for mining smoke, which adopts an intelligent early warning system for mining smoke, comprising the following steps: Step 1: Arrange multiple intelligent smoke alarm devices for mines in different monitoring areas of the workplace to be monitored, and record the location information of the multiple intelligent smoke alarm devices for mines. At the same time, store the location information of the multiple intelligent smoke alarm devices for mines and the three-dimensional information of the workplace to be monitored in the smoke alarm monitoring database of the central server; Step 2: Use multiple intelligent smoke alarm devices for mining to monitor smoke concentration in multiple monitoring areas respectively, and send the monitoring data to the central server in real time; S21: Control the piezoelectric micro pump 5 to start working, and use the piezoelectric micro pump 5 to provide negative pressure so that air in the environment enters the intelligent early warning device for mining smoke; S22: Performing a continuous three-stage detection operation on the dust-containing gas passing through the aerosol measurement optical chamber 2. At the same time, the temperature signal in the aerosol measurement optical chamber 2 is collected in real time by the temperature sensor 19 and sent to the data acquisition module 6; In the first detection stage, the 650nm laser source 10 is controlled to start working, and the 650nm laser source 10 is used to emit a first laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit 13 is controlled to start working, and the first photoelectric sensing unit 13 is used to receive the first laser light signal A scattered by the dust-containing gas, and the first laser light signal A is converted into a first laser electrical signal A, and then sent to the data acquisition module 6; in the second detection stage, the 650nm laser source 10 is controlled to stop working, and the 450nm laser source 11 is controlled to start working, and the 450nm laser source 11 is used to emit a second laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit 13 is controlled to start working, and the second photoelectric sensing unit 17 is controlled to start working, and the second photoelectric sensing unit 17 is used to receive the second laser light signal B scattered by the dust-containing gas, and the second laser light signal B is converted into a second laser electrical signal B, and then sent to the data acquisition module 6, and the first The photoelectric sensing unit 13 receives the second laser light signal C scattered by the dust-containing gas, and converts the second laser light signal C into a second laser electrical signal C, and then sends it to the data acquisition module 6; in the third detection stage, the 650nm laser source 10 is controlled to stop working, the 450nm laser source 11 is controlled to stop working, and the 850nm laser source 12 is controlled to start working, and the 850nm laser source 12 is used to emit a third laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit 13 is controlled to start working, and the third photoelectric sensing unit 15 is controlled to start working, and the third photoelectric sensing unit 15 is used to receive the third laser light signal D scattered by the dust-containing gas, and the third laser light signal D is converted into a third laser electrical signal D, and then sent to the data acquisition module 6, and the first photoelectric sensing unit 13 is used to receive the third laser light signal E scattered by the dust-containing gas, and the third laser light signal E is converted into a third laser electrical signal E, and then sent to the data acquisition module 6; S23: The airflow entering the trace gas monitoring chamber 3 is filtered by the particle high efficiency filter 4 to filter out particulate matter in the airflow. At the same time, the CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics in the airflow are respectively collected by the quartz chip oscillation module 27 through the piezoelectric quartz crystal detection matrix, and sent to the high-precision frequency measurement module 28. The high-precision frequency measurement module 28 obtains the CO concentration signal, CO2 concentration signal and relative humidity signal according to the CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics, and sends them to the data acquisition module 6; At the same time, the pressure difference signal before and after the airflow passes through the particle high efficiency filter 4 is detected by the pressure difference sensor 31 and sent to the data acquisition module 6; S24: The data acquisition module 6 sends the received first laser electrical signal A, second laser electrical signal B, second laser electrical signal C, third laser electrical signal D, third laser electrical signal E, temperature signal, CO concentration signal, CO2 concentration signal, relative humidity signal and pressure difference signal to the microprocessor 7; The microprocessor 7 uses the built-in light scattering sensor signal correction model, and combines the first laser electrical signal A and the receiving angle of the first photoelectric sensing unit 13 to the 650nm laser source 10 to obtain a 650nm laser electrical signal, combines the second laser electrical signal B, the second laser electrical signal C and the first photoelectric sensing unit 13 to the receiving angle of the 450nm laser source 11 to obtain a 450nm laser electrical signal, combines the third laser electrical signal D, the third laser electrical signal E and the first photoelectric sensing unit 13 to the receiving angle of the 850nm laser source 12 to obtain an 850nm laser electrical signal, and then classifies the particles based on the threshold range calibrated by the standard material and the particle size classification algorithm. The voltage pulse frequency and amplitude signals of the 650nm laser electrical signal, the 450nm laser electrical signal and the 850nm laser electrical signal are analyzed and processed respectively, and the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition are obtained respectively, and the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition are further cross-checked and supplemented to obtain the aerosol wet particle size distribution characteristic information; At the same time, the microprocessor 7 obtains the temperature data of the dust-laden airflow according to the temperature signal, obtains the CO concentration data and the CO2 concentration data according to the CO concentration signal and the CO2 concentration signal, respectively, and obtains the relative humidity data of the airflow according to the relative humidity signal; the microprocessor 7 uses the built-in trace gas quantitative detection and correction model, and calibrates the aerosol wet particle size distribution characteristic information in combination with the temperature data and the relative humidity data to obtain the aerosol dry particle size distribution characteristic information, and then calculates the aerosol particle concentration, CO concentration correction data and CO2 concentration correction data based on the aerosol dry particle size distribution characteristic information, and then performs time series processing on the aerosol dry particle size distribution characteristic information, the aerosol particle concentration, the CO concentration correction data and the CO2 concentration correction data, and then sends them to the central server; At the same time, the microprocessor 7 obtains the pressure difference data according to the pressure difference signal, and compares the pressure difference data with the set pressure difference threshold value. When the pressure difference data is greater than or equal to the set pressure difference threshold value, the flash device 30 in the alarm component is controlled to alarm, and at the same time, the display screen is controlled to display the reminder information of replacing the filter element of the particle high efficiency filter 4; Step 3: After receiving the alarm signal, the data processing unit sends the location information of the mining smoke intelligent alarm device that issued the alarm to the smoke alarm monitoring database for storage. At the same time, after receiving the aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, it sends them to the smoke alarm monitoring database for storage; At the same time, the data processing unit uses the built-in multivariate fusion smoke alarm recognition model to denoise the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and integrates the denoised data to form a comprehensive data view, and then sends it to the visualization client for real-time display; at the same time, the denoised data is used as the input data of the multivariate fusion smoke alarm recognition model, and the multivariate fusion smoke alarm recognition model extracts characteristic data from the input data that is helpful for fire identification, and then analyzes and processes the characteristic data and outputs fire warning result data. When the fire warning result data contains fire conditions in one or several monitoring areas, it outputs the possible cause of the fire and the preliminary judgment information of the type of burning material, and sends a fire warning signal, the possible cause of the fire and the preliminary judgment information of the type of burning material to the mining smoke intelligent alarm device in the corresponding monitoring area; At the same time, after receiving the fire warning signal, the microprocessor 7 in the intelligent smoke alarm device for mining controls the buzzer 29 and the flash device 30 in the alarm component to perform sound and light alarm actions at the same time, and controls the display screen to display the possible cause of the fire and the preliminary judgment information of the type of burning material, so as to effectively remind the personnel in the relevant monitoring area to take timely and effective response measures; At the same time, the data processing unit uses the built-in smoke distribution spatiotemporal evolution model to preprocess the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and then uses the preprocessed data as input data for the smoke distribution spatiotemporal evolution model. The smoke distribution spatiotemporal evolution model extracts characteristic data from the input data that is helpful for identifying smoke diffusion, and then analyzes and processes the characteristic data to output trend data of smoke diffusion and potential high-risk area warning information; as a preferred embodiment, the preprocessing includes data cleaning, standardization and noise removal to ensure the quality of the input data.

[0038] Step 4: The data processing unit sends the location information of multiple mine smoke intelligent alarm devices, the three-dimensional information of the workplace to be monitored, the fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information to the visualization client, and the visualization client realizes the real-time display of the three-dimensional information of the workplace to be monitored and the location information of multiple mine smoke intelligent alarm devices. At the same time, based on the fire warning result data, the alarm position is displayed in the three-dimensional information of the workplace to be monitored. At the same time, the received fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information are displayed in real time.

[0039] As a preferred embodiment, in step one, after the smoke alarm monitoring database receives the location information of multiple mine smoke intelligent alarm devices and the three-dimensional information of the workplace, it is stored in the mine smoke intelligent alarm device location information storage area and the workplace three-dimensional information storage area respectively; in step three, after the smoke alarm monitoring database receives the CO concentration correction data, CO2 concentration correction data, aerosol dry particle size distribution characteristic information and aerosol particle concentration, it is stored in the CO concentration correction data storage area, CO2 concentration correction data storage area, aerosol dry particle size distribution characteristic information storage area and aerosol particle concentration storage area respectively.

[0040] Compared with the prior art, the present invention has the following advantages: First, the aerosol measurement part uses three different wavelengths of laser sources: 650nm, 450nm and 850nm. Together with the corresponding three photoelectric sensing units, different types of aerosol particles can be captured more comprehensively. This multi-wavelength laser scattering technology can provide more detailed information about the size, distribution and concentration of aerosol particles, thereby significantly improving the ability to identify smoke types, accurately and efficiently identifying aerosol particle characteristics, helping to reduce false alarms and missed alarms, and providing reliable technical support for accurate and efficient smoke warnings.

[0041] Secondly, in the trace gas monitoring part, by using a specific adsorption film material to cover the piezoelectric quartz crystal sensor, the detection specificity and sensitivity of specific gases are significantly enhanced, and the CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics can be accurately obtained, so that the CO concentration, CO2 concentration and relative humidity in the gas can be accurately determined. The microprocessor is equipped with a light scattering sensor signal correction model and a trace gas quantitative detection correction model. The light scattering sensor signal correction model can be used to cross-check and supplement the aerosol particle size characteristic distribution information determined by the three laser signals, and the aerosol wet particle size distribution characteristic information can be accurately determined. At the same time, the trace gas quantitative detection correction model can be used to correct the aerosol wet particle size distribution characteristic information under different environmental conditions to obtain accurate aerosol dry particle size distribution characteristic information, which can significantly improve the accuracy of the measurement results and provide reliable data support for the early identification of fire sources.

[0042] Furthermore, the central server is used as the core processing part of the data, which can receive the monitoring data of multiple monitoring areas in real time, and can quickly integrate and efficiently analyze and process the monitoring data. The central server uses a multi-fusion smoke alarm recognition model to extract feature data that is helpful for fire identification from the monitoring data, and then can accurately identify complex fire scenes through data fusion technology, and can obtain fire warning result data, possible causes of fire and preliminary judgment information on the type of burning materials in real time. At the same time, the smoke distribution spatiotemporal evolution model is used to extract feature data that is helpful for smoke diffusion identification from the monitoring data, and the trend data of smoke diffusion and potential high-risk area warning information can be obtained. In this way, the use of a multi-fusion smoke alarm recognition model and a smoke distribution spatiotemporal evolution model can not only monitor and alarm fires in real time, but also predict the diffusion path of smoke and potential high-risk areas. This prediction capability is not available in traditional smoke alarm systems, and is of great significance for taking countermeasures in advance and planning evacuation routes.

[0043] Based on the in-depth analysis of the aerosol composition of mine smoke, the present invention fully combines technical means such as multi-wavelength laser scattering and trace gas monitoring, and integrates a variety of sensor data through data fusion technology. The data is analyzed through a deep learning algorithm, and the fire scene can be more accurately identified in the complex environment of the mine, and the trend of smoke diffusion can be predicted. The method has a high degree of intelligence, which can not only realize real-time monitoring and early warning of fires, but also predict the diffusion trend of smoke in advance, which is conducive to relevant personnel to take effective countermeasures in time.

Claims

1. A mine smoke intelligent warning device, comprising an outer shell (1), wherein the front end and the rear end of the outer shell (1) are respectively provided with a main air inlet and a main air outlet connected to the inner cavity thereof; characterized in that: It also includes an aerosol measurement optical chamber (2), a particle high efficiency filter (4), a differential pressure detection component, a trace gas monitoring chamber (3), a piezoelectric micro pump (5), a data acquisition module (6), a communication module (8) and a microprocessor (7); The aerosol measurement optical chamber (2), the particle high efficiency filter (4), the trace gas monitoring chamber (3) and the piezoelectric micro pump (5) are sequentially installed in the inner cavity of the outer shell (1) from front to back; The aerosol measurement optical chamber (2) comprises an optical chamber housing (9), a 650nm laser source (10), a first detection component, a 450nm laser source (11), a second detection component, an 850nm laser source (12), a third detection component and a temperature sensor (19); the optical chamber housing (9) is made of a matte black or black light-absorbing material, and an optical chamber detection cavity is formed therein; the front end and the rear end of the optical chamber housing (9) are respectively provided with an optical chamber inlet and an optical chamber outlet at the parts corresponding to the center of the optical chamber detection cavity, and the optical chamber inlet is connected to the main air inlet on the outer shell (1) through a main air inlet pipeline; the 650nm laser source (10) is connected to the main air inlet on the outer shell (1) through a main air inlet pipeline; the 850nm laser source (12) ... The light source (10) is installed in the middle section of the left inner wall of the optical chamber housing (9), and its laser emission direction is perpendicular to the axial direction of the optical chamber detection cavity and passes through the center of the optical chamber detection cavity; the first detection component is installed in the middle section of the upper inner wall of the optical chamber housing (9), and the axis of its detection surface is perpendicular to the laser emission direction of the 650nm laser source (10) and passes through the center of the optical chamber detection cavity; the first detection component includes a first photoelectric sensing unit (13) and a first light trap (14) coaxially distributed, and the first photoelectric sensing unit (13) is located on the inner side of the first light trap (14); the 450nm laser source (11) is installed in the middle section of the upper inner wall of the optical chamber housing (9), and the axis of its detection surface is perpendicular to the laser emission direction of the 650nm laser source (10) and passes through the center of the optical chamber detection cavity. The optical chamber housing (9) is mounted on the front of the left inner wall, and its laser emission direction passes through the center of the optical chamber detection cavity; the second detection component is mounted on the rear of the right inner wall of the optical chamber housing (9), and its detection surface corresponds to the laser emission direction of the 450nm laser source (11); the second detection component comprises a 450nm filter (20), a second photoelectric sensing unit (17) and a second light trap (18) which are coaxially distributed, and the 450nm filter (20), the second photoelectric sensing unit (17) and the second light trap (18) are sequentially distributed from the inside to the outside; the 850nm laser source (12) is mounted on the optical chamber housing (9) is located at the rear of the left inner wall, and the laser emission direction thereof passes through the center of the optical chamber detection cavity; the third detection component is installed at the front of the right inner wall of the optical chamber housing (9), and the detection surface thereof corresponds to the laser emission direction of the 850nm laser source (12); the third detection component comprises a coaxially distributed 850nm filter (21), a third photoelectric sensing unit (15) and a third light trap (16), and the 850nm filter (21), the third photoelectric sensing unit (15) and the third light trap (16) are sequentially distributed from the inside to the outside; the temperature sensor (19) is installed on the lower inner wall of the optical chamber housing (9); The front end and rear end of the particle high efficiency filter (4) are respectively provided with a filter inlet and a filter outlet, and the filter inlet is connected to the optical chamber outlet of the aerosol measurement optical chamber (2) through a first connecting pipe (34), and the filter outlet is connected to the inlet end of the second connecting pipe (35); The differential pressure detection assembly is arranged in the inner cavity of the outer shell (1) and is located on one side outside the particle high efficiency filter (4). The differential pressure detection assembly comprises a pre-filter connecting pipeline (32), a differential pressure sensor (31) and a post-filter connecting pipeline (33); the inlet end of the pre-filter connecting pipeline (32) is connected to the first connecting pipeline (34), one end of the differential pressure sensor (31) is connected to the outlet end of the pre-filter connecting pipeline (32), the inlet end of the post-filter connecting pipeline (33) is connected to the other end of the differential pressure sensor (31), and the outlet end thereof is connected to the second connecting pipeline (35); The trace gas monitoring chamber (3) comprises a monitoring chamber housing (22), a piezoelectric quartz crystal detection matrix, a quartz crystal oscillation module (27) and a high-precision frequency measurement module (28); a monitoring chamber is formed inside the monitoring chamber housing (22); a monitoring inlet and a monitoring outlet are respectively provided at the front and rear ends of the monitoring chamber housing (22) and are connected to the monitoring chamber, and the monitoring inlet is connected to the outlet end of the second connecting pipeline (35); the piezoelectric quartz crystal detection matrix is ​​installed in the central area of ​​the monitoring chamber, and the piezoelectric quartz crystal detection matrix comprises a sensor array acquisition module (26), a first piezoelectric quartz crystal sensor (23), a second piezoelectric quartz crystal sensor (24) and a first piezoelectric quartz crystal sensor (25). Three piezoelectric quartz crystal sensors (25); three sensing interfaces are arranged on the same side of the sensing array acquisition module (26); CO characteristic adsorption film material, CO2 characteristic adsorption film material and high hydrophilicity characteristic adsorption film material are attached to the first piezoelectric quartz crystal sensor (23), the second piezoelectric quartz crystal sensor (24) and the third piezoelectric quartz crystal sensor (25), respectively, and the three piezoelectric quartz crystal sensors are connected to the sensing array acquisition module (26) through the three sensing interfaces respectively; the quartz crystal chip oscillation module (27) is connected to the piezoelectric quartz crystal detection matrix; the high-precision frequency measurement module (28) is connected to the quartz crystal chip oscillation module (27); The inlet end of the piezoelectric micro pump (5) is connected to the monitoring outlet of the trace gas monitoring chamber (3) via a third connecting pipeline (36), and the outlet end thereof is connected to the main gas outlet of the main housing (1) via a main gas outlet pipeline; The data acquisition module (6), the communication module 1 (8) and the microprocessor (7) are all installed in the inner cavity of the outer shell (1); the data acquisition module (6) is respectively connected to the first photoelectric sensing unit (13), the third photoelectric sensing unit (15), the second photoelectric sensing unit (17), the temperature sensor (19), the pressure difference sensor (31) and the high-precision frequency measurement module (28); and the microprocessor (7) is respectively connected to the data acquisition module (6), the 650nm laser source (10), the 450nm laser source (11), the 850nm laser source (12), the piezoelectric micropump (5) and the communication module 1 (8).

2. The intelligent early warning device for mine smoke according to claim 1 is characterized in that: It also includes an alarm component, which is mounted on the outside of the outer shell (1), and includes a buzzer (29) and a flash device (30); the buzzer (29) and the flash device (30) are both connected to the microprocessor (7).

3. The intelligent early warning device for mine smoke according to claim 2 is characterized in that: A pair of electrodes are mounted opposite to each other on the outer edge of the upper end surface of the quartz crystal chips of the first piezoelectric quartz crystal sensor (23), the second piezoelectric quartz crystal sensor (24) and the third piezoelectric quartz crystal sensor (25), and nano-gold is coated on the portion other than the pair of electrodes.

4. The intelligent early warning device for mine smoke according to claim 3 is characterized in that: The main air inlet of the outer shell (1) and the light chamber inlet of the aerosol measurement light chamber (2) are located at the same horizontal plane, and the main air outlet of the outer shell (1) and the outlet end of the piezoelectric micro pump (5) are located at the same horizontal plane.

5. The intelligent early warning device for mine smoke according to claim 4 is characterized in that: The inner surface of the optical chamber housing (9) is coated with a non-adsorbent material having anti-adhesion properties.

6. The intelligent early warning device for mine smoke according to claim 5 is characterized in that: The temperature sensor (19) is a low power consumption sensor.

7. The intelligent early warning device for mine smoke according to claim 6 is characterized in that: It also includes a display screen, which is mounted on the surface of the outer shell (1) and connected to the microprocessor (7).

8. An intelligent early warning system for mining smoke, characterized in that: It includes a smoke alarm real-time monitoring network, which includes a plurality of intelligent early warning devices for mine smoke, and is characterized in that it also includes a central server and a visual client; A plurality of intelligent smoke alarm devices for mining are respectively arranged in different monitoring areas of the workplace to be monitored, and are connected to the central server via a communication link between the communication module 1 and the communication module 2; The central server includes a smoke alarm monitoring database, a data processing unit and a communication module 2, wherein the data processing unit is connected to the smoke alarm monitoring database and the communication module 2 respectively, and the data processing unit is connected to the smoke alarm real-time monitoring network respectively through the communication module 2; The visualization client is connected to the data processing unit via the communication module 2.

9. A method for intelligent early warning of mining smoke, using the intelligent early warning system for mining smoke as claimed in claim 8, characterized in that: The following steps are involved: Step 1: Arrange multiple intelligent smoke alarm devices for mines in different monitoring areas of the workplace to be monitored, and record the location information of the multiple intelligent smoke alarm devices for mines. At the same time, store the location information of the multiple intelligent smoke alarm devices for mines and the three-dimensional information of the workplace to be monitored in the smoke alarm monitoring database of the central server; Step 2: Use multiple intelligent smoke alarm devices for mining to monitor smoke concentration in multiple monitoring areas respectively, and send the monitoring data to the central server in real time; S21: Control the piezoelectric micro pump (5) to start working, and use the piezoelectric micro pump (5) to provide negative pressure so that air in the environment enters the intelligent early warning system for mining smoke; S22: performing a continuous three-stage detection operation on the dust-containing gas passing through the aerosol measurement optical chamber (2), and at the same time, collecting the temperature signal in the aerosol measurement optical chamber (2) in real time through the temperature sensor (19), and sending it to the data acquisition module (6); In the first detection phase, the 650nm laser source (10) is controlled to start working, and the 650nm laser source (10) is used to emit a first laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit (13) is controlled to start working, and the first photoelectric sensing unit (13) is used to receive a first laser light signal A scattered by the dust-containing gas, and the first laser light signal A is converted into a first laser electrical signal A, which is then sent to the data acquisition module (6). In the second detection phase, the 650nm laser source (10) is controlled to stop working, and the 450nm laser source (11) is controlled to start working, and the 450nm laser source (11) is used to emit a second laser beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit (13) is controlled to start working, and the second photoelectric sensing unit (17) is controlled to start working, and the second photoelectric sensing unit (17) is controlled to start working, and the second laser light signal B scattered by the dust-containing gas is received by the second photoelectric sensing unit (17), and the second laser light signal B is converted into a second laser electrical signal B, which is then sent to the data acquisition module (6). The photoelectric sensing unit (13) receives a second laser light signal C scattered by the dust-containing gas, converts the second laser light signal C into a second laser electrical signal C, and then sends it to the data acquisition module (6); in the third detection stage, the 650nm laser source (10) is controlled to stop working, the 450nm laser source (11) is controlled to stop working, and the 850nm laser source (12) is controlled to start working, and the 850nm laser source (12) is used to emit a third laser light beam to the center of the optical chamber detection cavity. At the same time, the first photoelectric sensing unit (13) is controlled to start working, and the third photoelectric sensing unit (15) is controlled to start working, and the third photoelectric sensing unit (15) is used to receive a third laser light signal D scattered by the dust-containing gas, and the third laser light signal D is converted into a third laser electrical signal D, and then sent to the data acquisition module (6); the first photoelectric sensing unit (13) is used to receive a third laser light signal E scattered by the dust-containing gas, and the third laser light signal E is converted into a third laser electrical signal E, and then sent to the data acquisition module (6); S23: filtering the airflow entering the trace gas monitoring chamber (3) through a particle high efficiency filter (4) to remove particulate matter in the airflow. At the same time, using a quartz chip oscillation module (27) to collect CO frequency characteristics, CO2 frequency characteristics and relative humidity frequency characteristics in the airflow through a piezoelectric quartz crystal detection matrix, and sending them to a high-precision frequency measurement module (28). The high-precision frequency measurement module (28) obtains a CO concentration signal, a CO2 concentration signal and a relative humidity signal according to the CO frequency characteristics, the CO2 frequency characteristics and the relative humidity frequency characteristics, and sends them to a data acquisition module (6); At the same time, a pressure difference signal before and after the airflow passes through the particle high efficiency filter (4) is detected by a pressure difference sensor (31) and sent to a data acquisition module (6); S24: the data acquisition module (6) sends the received first laser electrical signal A, second laser electrical signal B, second laser electrical signal C, third laser electrical signal D, third laser electrical signal E, temperature signal, CO concentration signal, CO2 concentration signal, relative humidity signal and pressure difference signal to the microprocessor (7); The microprocessor (7) uses a built-in light scattering sensor signal correction model, and combines the first laser electrical signal A and the receiving angle of the first photoelectric sensing unit (13) to the 650nm laser source (10) to obtain a 650nm laser electrical signal, combines the second laser electrical signal B, the second laser electrical signal C and the receiving angle of the first photoelectric sensing unit (13) to the 450nm laser source (11) to obtain a 450nm laser electrical signal, combines the third laser electrical signal D, the third laser electrical signal E and the receiving angle of the first photoelectric sensing unit (13) to the 850nm laser source (12) to obtain an 850nm laser electrical signal, and then combines the threshold range calibrated based on the standard material to obtain a 450nm laser electrical signal. The particle size classification algorithm analyzes and processes the voltage pulse frequency and amplitude signals of the 650nm laser electrical signal, the 450nm laser electrical signal and the 850nm laser electrical signal, respectively, and obtains the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition, and further cross-checks and supplements the aerosol particle size characteristic information under the 650nm laser condition, the aerosol particle size characteristic information under the 450nm laser condition and the aerosol particle size characteristic information under the 850nm laser condition, so as to obtain the aerosol wet particle size distribution characteristic information; At the same time, the microprocessor (7) obtains temperature data of the dust-laden airflow according to the temperature signal, obtains CO concentration data and CO2 concentration data according to the CO concentration signal and the CO2 concentration signal, respectively, and obtains relative humidity data of the airflow according to the relative humidity signal; the microprocessor (7) uses a built-in trace gas quantitative detection and correction model, and calibrates the aerosol wet particle size distribution characteristic information in combination with the temperature data and the relative humidity data to obtain the aerosol dry particle size distribution characteristic information, and then calculates the aerosol particle concentration, CO concentration correction data, and CO2 concentration correction data based on the aerosol dry particle size distribution characteristic information, and then performs time series processing on the aerosol dry particle size distribution characteristic information, the aerosol particle concentration, the CO concentration correction data, and the CO2 concentration correction data, and then sends them to the central server; At the same time, the microprocessor (7) obtains pressure difference data according to the pressure difference signal, and compares the pressure difference data with a set pressure difference threshold value. When the pressure difference data is greater than or equal to the set pressure difference threshold value, the flash device (30) in the alarm component is controlled to sound an alarm, and at the same time, the display screen is controlled to display a reminder message for replacing the filter element of the particle high efficiency filter (4); Step 3: After receiving the alarm signal, the data processing unit sends the location information of the mining smoke intelligent alarm device that issued the alarm to the smoke alarm monitoring database for storage. At the same time, after receiving the aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, it sends them to the smoke alarm monitoring database for storage; At the same time, the data processing unit uses the built-in multivariate fusion smoke alarm recognition model to denoise the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and integrates the denoised data to form a comprehensive data view, and then sends it to the visualization client for real-time display; at the same time, the denoised data is used as the input data of the multivariate fusion smoke alarm recognition model, and the multivariate fusion smoke alarm recognition model extracts characteristic data from the input data that is helpful for fire identification, and then analyzes and processes the characteristic data and outputs fire warning result data. When the fire warning result data contains fire conditions in one or several monitoring areas, it outputs the possible cause of the fire and the preliminary judgment information of the type of burning material, and sends a fire warning signal, the possible cause of the fire and the preliminary judgment information of the type of burning material to the mining smoke intelligent alarm device in the corresponding monitoring area; At the same time, after receiving the fire warning signal, the microprocessor (7) in the intelligent smoke alarm device for mining controls the buzzer (29) and the flash device (30) in the alarm component to simultaneously perform sound and light alarm actions, and controls the display screen to display preliminary judgment information on the possible cause of the fire and the type of burning material; At the same time, the data processing unit uses the built-in smoke distribution spatiotemporal evolution model to preprocess the received aerosol dry particle size distribution characteristic information, aerosol particle concentration, CO concentration correction data and CO2 concentration correction data, and then uses the preprocessed data as the input data of the smoke distribution spatiotemporal evolution model. The smoke distribution spatiotemporal evolution model extracts characteristic data from the input data that is helpful for smoke diffusion identification, and then analyzes and processes the characteristic data and outputs smoke diffusion trend data and potential high-risk area warning information; Step 4: The data processing unit sends the location information of multiple mine smoke intelligent alarm devices, the three-dimensional information of the workplace to be monitored, the fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information to the visualization client, and the visualization client realizes the real-time display of the three-dimensional information of the workplace to be monitored and the location information of multiple mine smoke intelligent alarm devices. At the same time, based on the fire warning result data, the alarm position is displayed in the three-dimensional information of the workplace to be monitored. At the same time, the received fire warning result data, the possible cause information of the fire, the combustion material type information, the smoke diffusion trend data and the potential high-risk area warning information are displayed in real time.

10. The intelligent early warning method for mining smoke according to claim 9, characterized in that: In step one, after the smoke alarm monitoring database receives the location information of multiple mine smoke intelligent alarm devices and the three-dimensional information of the workplace, it is stored in the mine smoke intelligent alarm device location information storage area and the workplace three-dimensional information storage area respectively; in step three, after the smoke alarm monitoring database receives the CO concentration correction data, CO2 concentration correction data, aerosol dry particle size distribution characteristic information and aerosol particle concentration, it is stored in the CO concentration correction data storage area, CO2 concentration correction data storage area, aerosol dry particle size distribution characteristic information storage area and aerosol particle concentration storage area respectively.

Images