An Internet of Things-based regional fire extinguishing security monitoring system

Abstract

An Internet of Things type regional fire extinguishing security monitoring system includes a regional fire extinguishing device group, a video monitoring group and a host computer centralized monitoring system; the regional fire extinguishing device group is composed of eight upgraded and transformed mine-used regional automatic powder spraying fire extinguishing devices, which are arranged in various electromechanical equipment chambers in coal mines; the video monitoring group is composed of 16 mine-used intrinsically safe cameras, which are arranged in various electromechanical equipment chambers in coal mines, collect video images of electromechanical chambers, and provide video data for fire image recognition. The host computer centralized monitoring system includes image monitoring and equipment parameter monitoring. Image monitoring combined with graphic recognition algorithm can analyze in real time whether a fire occurs in the image, and equipment parameter monitoring helps staff to remotely understand the equipment status. Advantages of the present invention: multi-dimensional monitoring of whether a fire occurs in the coal mine. Once a fire is detected, the fire extinguishing device is quickly turned on to achieve remote centralized monitoring of the fire extinguishing equipment, and to prevent the occurrence of fires in coal mines to the greatest extent.

Description

Technical Field

[0001] The present invention relates to the field of monitoring systems, and in particular to an Internet of Things type regional fire extinguishing security monitoring system. Background Art

[0002] The controller of the automatic powder spray fire extinguishing device for mining area is connected with the intrinsically safe pressure transmitter, intrinsically safe solenoid valve, sound and light alarm and intrinsically safe flame sensor respectively, which can realize active and rapid fire extinguishing and explosion suppression, and is the main security equipment for fire prevention and extinguishing in coal mines. In a certain period of time, it can achieve good fire extinguishing and explosion suppression effects. However, in recent years, with the gradual acceleration of the pace of intelligent upgrading and transformation of the coal mining industry, the automatic powder spray fire extinguishing device for mining area cannot meet the development needs of intelligent, information-based, less-manned and unmanned coal mines. The main problems are as follows:

[0003] Inspection personnel need to be arranged regularly to check whether the powder spray bottle is leaking, and the equipment maintenance process is cumbersome;

[0004] The control unit does not have a communication interface, so the alarm information cannot be transmitted to the ground centralized control when a fire occurs;

[0005] Sensor failures cannot be diagnosed autonomously, and the inspection personnel and the centralized control center cannot be notified;

[0006] Flame detection relies on flame sensors and smoke sensors, which have a small detection area, slow detection speed, single detection method, and the possibility of false detection and missed detection.

[0007] The automatic powder spray fire extinguishing devices in mining areas cannot be intelligently interconnected, making it impossible to achieve joint prevention and control of underground coal mine fires. Summary of the invention

[0008] The purpose of the present invention is that the Internet of Things type regional fire extinguishing security monitoring system utilizes cutting-edge technologies such as image recognition technology, digital sensor technology, automatic control technology, and intelligent interconnection technology to achieve intelligent security, joint prevention and control, and remote monitoring of underground coal mine fires, and specially provides an Internet of Things type regional fire extinguishing security monitoring system.

[0009] The present invention provides an Internet of Things-based regional fire extinguishing security monitoring system, which is characterized in that: the Internet of Things-based regional fire extinguishing security monitoring system includes a regional fire extinguishing device group, a video monitoring group and a host computer centralized monitoring system;

[0010] The Internet of Things regional fire extinguishing security monitoring system is an intelligent security, joint control and remote monitoring system that integrates image recognition technology, digital sensor technology, automatic control technology and intelligent interconnection technology. It is an important and effective technology and equipment to ensure safe production in coal mines. This system is mainly composed of regional fire extinguishing device group, video monitoring group and host computer system.

[0011] 1. The regional fire extinguishing device group consists of eight upgraded and renovated mine-use regional automatic powder spraying fire extinguishing devices, which are arranged in the electromechanical equipment chambers in the coal mine and are responsible for the fire prevention and extinguishing work of the electromechanical chambers. The mine-use regional automatic powder spraying fire extinguishing device adopts the following new technologies:

[0012] (1) The power supply integration technology is adopted to integrate the substation and power supply into one design, which reduces the number of explosion-proof boxes and facilitates installation and maintenance;

[0013] (2) The analog signal isolation and anti-interference design is adopted. The T1 part in the schematic diagram of the power supply integrated regional fire extinguishing network port substation circuit is a dual-channel analog signal isolation circuit. In the figure, pins 1 and 2 of HN1 are isolated signal inputs. When the input signal is a 4-20mA current type, it is transformed by resistor IV to light up the input light-emitting tube. The input signal size is proportional to the current flowing through the light-emitting tube. The light emitted by the light-emitting tube will induce a current proportional to the light-emitting tubes at pins 1 and 2 on the light-emitting tubes at pins 3 and 4, and induce a current proportional to the light intensity at pins 1 and 2 on pins 5 and 6. In the circuit, U1 adds a voltage follower design at the signal input, which can reduce impedance and avoid the influence of the subsequent circuit on the input signal. The first half of HN1 and op amp U1 and the second half and op amp U2 in the circuit use isolated power supply. This can block the interference from the sensor to the microcontroller part.

[0014] (3) The network optical port design is adopted, and the network port circuit adopts the W5300 chip. The bus method is used to configure its registers with the core controller. The configuration process mainly includes power-on initialization, general register configuration, and socket register configuration. After the substation is powered on, the 66th pin of W5300 is pulled low for 3us and then restored to a high level for 15ms. After the phase-locked loop is stable, the register is configured. The local port number is set through SO_PORTR in the port register, and the communication protocol is set through P3-P0 in S0_MR under the mode register. Then the control register S0_CR is set high to execute the open command. At this time, if S0_SSR in the Socket0 register changes to 0x13, the initialization is completed. Continue to send the monitoring instruction 0X02 to S0_CR. At this time, if the S0_SSR state changes to monitoring 0x14, the monitoring task is completed. W5300 is set as TCP server; finally, wait for S0_SSR to change to SOCK_ESTABLISHED (0x17). If S0_SSR changes to the expected value, Socket0 connection is established and data communication can be performed. In order to realize data transmission through the external optical port, the commonly used 2 optical 3 electrical module on the market is used to realize the expansion of electrical port to optical port. In the figure, the 92 and 93 pins of U5 are used to connect the mesh module, which is connected through the serial port and the MESH network port. When a watch or a patrol device that supports mesh is connected, it can warn the perception of device information and realize the Internet of Things.

[0015] 2. The video surveillance group consists of 16 intrinsically safe cameras for mining, which are arranged in the electromechanical equipment chambers in the coal mine. They are used to collect video images of the electromechanical chambers and provide video data for fire image recognition. The image recognition algorithm is based on deep learning and convolutional neural network construction. The construction process is as follows: first, a large amount of labeled fire image data is normalized, and a 50-layer ResNet convolutional neural network is constructed using the PyTorch library. Then, the image is loaded for model training, the error is calculated, the optimizer is called for optimization, and finally the model is tested using test data. The above process is repeated until the model prediction accuracy reaches 93%. The image recognition algorithm is an important part of the multi-dimensional flame recognition technology. Through multi-dimensional technical means such as flame sensors, smoke sensors, and image recognition, zero false detection and zero missed detection of coal mine fires can be achieved, and accurate identification of coal mine fires can be achieved.

[0016] 3. The centralized monitoring system of the upper computer includes two parts: image monitoring and equipment parameter monitoring. Image monitoring combined with graphic recognition algorithm can analyze in real time whether a fire has occurred in the image. Equipment parameter monitoring helps staff to understand the equipment status remotely.

[0017] The IoT-based regional fire extinguishing security monitoring system deploys regional fire extinguishing device groups and video monitoring groups in each equipment chamber in the coal mine. Through flame sensors, smoke sensors and image recognition algorithms, it can identify whether a fire has occurred in the coal mine in multiple dimensions, and prevent false detection and missed detection of mine fires. At the same time, through the technical upgrade of the core control board and the upper computer monitoring system, the intelligent interconnection of equipment is realized, breaking the information island in the coal mine, realizing the interconnection of everything in the coal mine and on the ground, and promoting the development of informationization and intelligence in coal mines.

[0018] Advantages of the present invention:

[0019] Coal mine fire is one of the major disasters that threaten the safe production of coal mines. Fire accidents will not only produce a large amount of high-temperature smoke and toxic and harmful gases, endangering the lives of underground workers, but also burn a large number of underground equipment, consume coal resources, and bring huge disasters to the mine. The Internet of Things regional fire extinguishing security monitoring system uses flame sensors, carbon monoxide smoke sensors and graphic recognition algorithms to monitor whether there is a fire in the coal mine in multiple dimensions. Once a fire is detected, the fire extinguishing device is quickly turned on to extinguish the flame. At the same time, the regional fire extinguishing device controller can be connected to the underground optical fiber ring network through the optical fiber interface to realize remote centralized monitoring of the fire extinguishing equipment and prevent the occurrence of underground coal mine fires to the greatest extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments:

[0021] Figure 1 This is the architecture diagram of the power supply integrated regional fire extinguishing network port substation architecture;

[0022] Figure 2 It is the T1 part in the circuit schematic diagram of the power supply integrated area fire extinguishing network substation;

[0023] Figure 3 This is a schematic diagram of the structure of an IoT-based regional fire extinguishing security monitoring system;

[0024] Figure 4 Schematic diagram of the process of building a convolutional neural network model for image recognition. DETAILED DESCRIPTION

[0025] Example

[0026] The present invention provides an Internet of Things-based regional fire extinguishing security monitoring system, which is characterized in that: the Internet of Things-based regional fire extinguishing security monitoring system includes a regional fire extinguishing device group, a video monitoring group and a host computer centralized monitoring system;

[0027] The Internet of Things regional fire extinguishing security monitoring system is an intelligent security, joint control and remote monitoring system that integrates image recognition technology, digital sensor technology, automatic control technology and intelligent interconnection technology. It is an important and effective technology and equipment to ensure safe production in coal mines. This system is mainly composed of regional fire extinguishing device group, video monitoring group and host computer system.

[0028] 1. The regional fire extinguishing device group consists of eight upgraded and renovated mine-use regional automatic powder spraying fire extinguishing devices, which are arranged in the electromechanical equipment chambers in the coal mine and are responsible for the fire prevention and extinguishing work of the electromechanical chambers. The mine-use regional automatic powder spraying fire extinguishing device adopts the following new technologies:

[0029] (1) The power supply integration technology is adopted to integrate the substation and power supply into one design, which reduces the number of explosion-proof boxes and facilitates installation and maintenance;

[0030] (2) The analog signal isolation and anti-interference design is adopted. The T1 part in the schematic diagram of the power supply integrated regional fire extinguishing network port substation circuit is a dual-channel analog signal isolation circuit. In the figure, pins 1 and 2 of HN1 are isolated signal inputs. When the input signal is a 4-20mA current type, it is transformed by resistor IV to light up the input light-emitting tube. The input signal size is proportional to the current flowing through the light-emitting tube. The light emitted by the light-emitting tube will induce a current proportional to the light-emitting tubes at pins 1 and 2 on the light-emitting tubes at pins 3 and 4, and induce a current proportional to the light intensity at pins 1 and 2 on pins 5 and 6. In the circuit, U1 adds a voltage follower design at the signal input, which can reduce impedance and avoid the influence of the subsequent circuit on the input signal. The first half of HN1 and op amp U1 and the second half and op amp U2 in the circuit use isolated power supply. This can block the interference from the sensor to the microcontroller part.

[0031] (3) The network optical port design is adopted, and the network port circuit adopts the W5300 chip. The bus method is used to configure its registers with the core controller. The configuration process mainly includes power-on initialization, general register configuration, and socket register configuration. After the substation is powered on, the 66th pin of W5300 is pulled low for 3us and then restored to a high level for 15ms. After the phase-locked loop is stable, the register is configured. The local port number is set through SO_PORTR in the port register, and the communication protocol is set through P3-P0 in S0_MR under the mode register. Then the control register S0_CR is set high to execute the open command. At this time, if S0_SSR in the Socket0 register changes to 0x13, the initialization is completed. Continue to send the monitoring instruction 0X02 to S0_CR. At this time, if the S0_SSR state changes to monitoring 0x14, the monitoring task is completed. W5300 is set as TCP server; finally, wait for S0_SSR to change to SOCK_ESTABLISHED (0x17). If S0_SSR changes to the expected value, Socket0 connection is established and data communication can be performed. In order to realize data transmission through the external optical port, the commonly used 2 optical 3 electrical module on the market is used to realize the expansion of electrical port to optical port. In the figure, the 92 and 93 pins of U5 are used to connect the mesh module, which is connected through the serial port and the MESH network port. When a watch or a patrol device that supports mesh is connected, it can warn the perception of device information and realize the Internet of Things.

[0032] 2. The video surveillance group consists of 16 intrinsically safe cameras for mining, which are arranged in the electromechanical equipment chambers in the coal mine. They are used to collect video images of the electromechanical chambers and provide video data for fire image recognition. The image recognition algorithm is based on deep learning and convolutional neural network construction. The construction process is as follows: first, a large amount of labeled fire image data is normalized, and a 50-layer ResNet convolutional neural network is constructed using the PyTorch library. Then, the image is loaded for model training, the error is calculated, the optimizer is called for optimization, and finally the model is tested using test data. The above process is repeated until the model prediction accuracy reaches 93%. The image recognition algorithm is an important part of the multi-dimensional flame recognition technology. Through multi-dimensional technical means such as flame sensors, smoke sensors, and image recognition, zero false detection and zero missed detection of coal mine fires can be achieved, and accurate identification of coal mine fires can be achieved.

[0033] 3. The centralized monitoring system of the upper computer includes two parts: image monitoring and equipment parameter monitoring. Image monitoring combined with graphic recognition algorithm can analyze in real time whether a fire has occurred in the image. Equipment parameter monitoring helps staff to understand the equipment status remotely.

[0034] The IoT-based regional fire extinguishing security monitoring system deploys regional fire extinguishing device groups and video monitoring groups in each equipment chamber in the coal mine. Through flame sensors, smoke sensors and image recognition algorithms, it can identify whether a fire has occurred in the coal mine in multiple dimensions, and prevent false detection and missed detection of mine fires. At the same time, through the technical upgrade of the core control board and the upper computer monitoring system, the intelligent interconnection of equipment is realized, breaking the information island in the coal mine, realizing the interconnection of everything in the coal mine and on the ground, and promoting the development of informationization and intelligence in coal mines.

Claims

1. An Internet of Things regional fire extinguishing security monitoring system, characterized by: The Internet of Things-based regional fire extinguishing security monitoring system includes a regional fire extinguishing device group, a video monitoring group and a host computer centralized monitoring system; The regional fire extinguishing device group is composed of eight upgraded and renovated mine-use regional automatic powder spraying fire extinguishing devices, which are arranged in the electromechanical equipment chambers in the coal mine and are responsible for the fire prevention and extinguishing work of the electromechanical chambers; the mine-use regional automatic powder spraying fire extinguishing device adopts the following new scheme, adopts the power supply integration technology, and designs the substation and power supply as one, reducing the number of explosion-proof boxes and facilitating installation and maintenance; adopts analog signal isolation and anti-interference design, and the T1 part in the circuit schematic diagram of the power supply integrated regional fire extinguishing network port substation is a dual-channel analog signal isolation circuit; Pins 1 and 2 of HN1 are isolated signal inputs. When the input signal is a 4-20mA current type, it lights up the input light-emitting tube after being transformed by resistor IV. The size of the input signal is proportional to the size of the current flowing through the light-emitting tube. The light emitted by the light-emitting tube will induce a current proportional to the light-emitting tubes at pins 1 and 2 on the light-emitting tubes at pins 3 and 4, and a current proportional to the light intensity of pins 1 and 2 on pins 5 and 6; In the circuit, U1 adds a voltage follower design at the signal input to reduce impedance and avoid the influence of the subsequent circuit on the input signal; the first half of HN1 and the operational amplifier U1 and the second half and the operational amplifier U2 in the circuit adopt isolated power supply; this can block the interference from the sensor to the single-chip microcomputer part; the network optical port design is adopted, and the network port circuit adopts the W5300 chip, which and the core controller use the bus method to configure its registers; the configuration process is power-on initialization, general register configuration, and socket register configuration; after the substation is powered on, the 66th pin of W5300 is pulled down for 3us and then restored to a high level for 15ms to ensure that the register is configured after the phase-locked loop is stable; The local port number is set through SO_PORTR in the port register, and the communication protocol is set through P3-P0 in S0_MR under the mode register; then the control register S0_CR is set high to execute the open command. At this time, if S0_SSR in the Socket0 register changes to 0x13, the initialization is completed; continue to send the monitoring instruction 0X02 to S0_CR. At this time, if the S0_SSR state changes to monitoring 0x14, the monitoring task is completed; W5300 is set as a TCP server; finally, wait for S0_SSR to change to SOCK_ESTABLISHED (0x17). If S0_SSR changes to the expected value, the Socket0 connection is established and data communication can be carried out; in order to realize data transmission to the external optical port, the commonly used 2-optical 3-electrical module on the market is used to realize the expansion of electrical port to optical port; the 92 and 93 pins of U5 in the figure are used to connect the mesh module, which is connected through the serial port and the MESH network port. When a watch or a patrol device that supports mesh is connected, it can warn the perception of device information and realize the Internet of Things.

2. The Internet of Things regional fire extinguishing security monitoring system according to claim 1 is characterized in that: The video monitoring group is composed of 16 intrinsically safe cameras for mining, which are arranged in the electromechanical equipment chambers in the coal mine to collect video images of the electromechanical chambers and provide video data for fire image recognition. The image recognition algorithm is constructed based on deep learning and convolutional neural networks. The construction process is as follows: first, a large amount of labeled fire image data is normalized, and a 50-layer ResNet convolutional neural network is constructed using the PyTorch library. Then, the image is loaded for model training, the error is calculated, the optimizer is called for optimization, and finally, the model is tested using test data. The above process is repeated until the model prediction accuracy reaches 93%. The image recognition algorithm is an important part of the multi-dimensional flame recognition technology. Through flame sensors, smoke sensors, and image recognition multi-dimensional technical means, zero false detection and zero missed detection of coal mine fires can be achieved, and accurate recognition of coal mine fires can be achieved.

3. The Internet of Things regional fire extinguishing security monitoring system according to claim 1 is characterized in that: The host computer centralized monitoring system includes two parts: image monitoring and equipment parameter monitoring. Image monitoring combined with a graphic recognition algorithm can analyze in real time whether a fire has occurred in an image, and equipment parameter monitoring helps staff to remotely understand the equipment status.

Images