A fire monitoring method and system based on a mine belt conveyor

Abstract

The application provides a fire monitoring method and system based on a mine belt conveyor, which realizes accurate identification and early warning of fire risk through multi-dimensional data fusion. The total journey of the mine belt conveyor is divided into multiple sections according to a preset length, and thermal imaging data, environmental temperature data, carbon monoxide concentration data and carbon dioxide concentration data are collected in each section. At the same time, the belt temperature data is collected at the head and tail of the belt conveyor, and the belt temperature difference is calculated. Through a preset fire identification algorithm, the multi-dimensional monitoring data and the belt temperature difference data of each section are comprehensively analyzed to generate corresponding section fire probability values. When the section fire probability value exceeds the preset fire warning threshold, the system defines the section as a fire risk section and issues a warning. By adopting a segmented monitoring strategy, combining thermal imaging technology, gas concentration detection and temperature difference analysis and other monitoring means, accurate positioning and intelligent early warning of the fire risk of the mine belt conveyor are realized.

Description

Technical Field

[0001] The present application relates to the field of mining belt conveyors, and in particular to a fire monitoring method and system based on mining belt conveyors. Background Art

[0002] As a key piece of equipment for underground transportation, fire monitoring technology for mining belt conveyors is crucial to ensuring safe production in mines. Currently, the main fire monitoring methods include temperature monitoring, smoke monitoring, and carbon monoxide monitoring.

[0003] Temperature monitoring technology primarily uses infrared thermal imagers, fiber optic temperature sensors, and thermocouples to monitor the real-time temperature of key belt conveyor components. Infrared thermal imagers clearly report abnormal belt temperature fluctuations caused by friction, and immediately trigger an alarm if the temperature exceeds the warning threshold. However, this method has significant drawbacks: First, temperature sensors are susceptible to the high temperature, high humidity, and strong electromagnetic interference in underground environments, resulting in reduced measurement accuracy and increased false alarm rates. Second, the installation location and number of sensors make it difficult to achieve full coverage, easily creating monitoring blind spots. Third, simple temperature monitoring cannot distinguish between normal frictional heating and abnormal fire signs, resulting in frequent misjudgments.

[0004] Smoke monitoring technology uses devices such as ionization and photoelectric smoke sensors to detect smoke particles in the air. However, in mine environments, this technology faces serious challenges: coal dust, blasting smoke, and dust generated by mechanical operations can trigger smoke sensors, causing a large number of false alarms. The sensors are sensitive to air flow, and the complex ventilation system underground can affect smoke diffusion and detection effectiveness. Furthermore, the sensors are difficult to maintain and clean, and their detection sensitivity decreases significantly after long-term use.

[0005] Carbon monoxide monitoring technology determines fire conditions by detecting carbon monoxide concentrations in combustion products. Its main drawbacks include: multiple sources of carbon monoxide exist underground, such as diesel exhaust and blasting operations, creating complex interference signals; changes in carbon monoxide concentrations are slow, meaning that by the time abnormal concentrations are detected, the fire may have already reached a more serious stage; and gas sensors have limited stability and service life in the harsh underground environment, requiring frequent replacement and calibration. Summary of the Invention

[0006] The present application provides a fire monitoring method based on a mining belt conveyor, comprising the following steps:

[0007] A1, dividing the preset total travel length of the mining belt conveyor according to the preset travel section length to form each belt travel section;

[0008] A2, collects corresponding section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data in each belt travel section;

[0009] A3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor;

[0010] A4, collecting corresponding head belt temperature data at the head of the mining belt conveyor;

[0011] A5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data;

[0012] A6 generates a corresponding fire probability value for each belt travel section using a preset fire identification algorithm based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data, and belt forward temperature difference data corresponding to each belt travel section;

[0013] A7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section.

[0014] By adopting the above technical solution, the fire monitoring method based on mining belt conveyors can realize the positioning and early warning of fire risks of mining belt conveyors by monitoring the total travel of the belt conveyors in sections, combining multi-dimensional data collection such as thermal imaging, ambient temperature, carbon monoxide and carbon dioxide concentrations, and the temperature difference between the head and tail, thereby improving the accuracy and reliability of fire detection, being able to identify potential fire sections in a timely manner, and providing more comprehensive and intelligent protection for mine safety production.

[0015] Optionally, step A5 includes the following steps:

[0016] A501 calculates the corresponding one-way belt travel time based on the quotient of the total travel length and the preset belt running speed;

[0017] A502 obtains the current time and defines it as the belt head time. The corresponding belt tail time is calculated by subtracting the belt one-way travel time from the belt head time.

[0018] A503 determines the corresponding belt head monitoring time window based on the belt head moment and the belt one-way movement time;

[0019] A504 determines the corresponding monitoring time window for the belt's outward tail according to the belt's outward tail moment and the belt's one-way travel time.

[0020] A505, obtains the corresponding belt head time window temperature data according to the belt head monitoring time window and the belt head temperature data;

[0021] A506, obtaining the belt head time window temperature data according to the belt head monitoring time window and the head belt temperature data;

[0022] A507, obtaining the belt one-way temperature difference data by subtracting the corresponding belt temperature values in the belt head time window temperature data and the belt one-way tail time window temperature data.

[0023] By adopting the above technical scheme, the fire monitoring method based on the mine belt conveyor can calculate the belt one-way moving time length, and set the monitoring time windows of the head and tail based on the time synchronization principle, so as to ensure that the temperature data of the same belt at the head and tail can be accurately corresponded, thereby calculating the one-way temperature difference data which truly reflects the temperature change of the belt in the running process, effectively eliminating the data misplacement problem caused by the belt movement, improving the accuracy and reliability of the temperature difference data, and providing more accurate basic data support for subsequent fire risk judgment.

[0024] Optionally, the fire identification algorithm comprises the following steps:

[0025] B1, generating corresponding section thermal imaging feature data according to the section thermal imaging data through a preset image feature extraction algorithm;

[0026] B2, obtaining corresponding section time window temperature data, section time window carbon monoxide concentration data and section time window carbon dioxide concentration data according to the belt head monitoring time window and the section environment temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data, respectively;

[0027] B3, generating corresponding section temperature feature data, section carbon monoxide concentration feature data, section carbon dioxide concentration feature data and belt one-way temperature difference feature data according to the section time window temperature data, section time window carbon monoxide concentration data, section time window carbon dioxide concentration data and belt one-way temperature difference data through a preset data feature extraction algorithm, respectively;

[0028] B4, generating section comprehensive feature data set by combining the section thermal imaging feature data, section temperature feature data, section carbon monoxide concentration feature data, section carbon dioxide concentration feature data and belt one-way temperature difference feature data;

[0029] B5, inputting the section comprehensive feature data set into the pre-trained fire probability identification model to generate corresponding section fire probability value.

[0030] By adopting the above technical solution, the fire monitoring method based on mining belt conveyors can extract and combine the thermal imaging image features with multi-dimensional sensor data such as ambient temperature, gas concentration, and belt temperature difference to form a comprehensive feature data set, which is then input into a pre-trained machine learning model to achieve diversified variable analysis. This not only improves the accuracy and robustness of fire identification, but also effectively reduces the false alarm rate. Through the adaptive ability of the deep learning model, it can better identify fire signs in complex mining environments, providing more reliable and intelligent technical support for mine belt conveyor fire early warning.

[0031] Optionally, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0032] C1, calculate the corresponding section time window average temperature based on the section time window temperature data corresponding to each belt travel section;

[0033] C2, generating belt time window ambient temperature distribution data according to the average temperature of the section time window corresponding to each belt travel section;

[0034] C3, calculate the average carbon monoxide concentration of the corresponding section time window according to the carbon monoxide concentration data of each belt travel section;

[0035] C4, generating belt time window carbon monoxide concentration distribution data according to the average carbon monoxide concentration in the time window corresponding to each belt travel section;

[0036] C5, calculate the average carbon dioxide concentration of the corresponding section time window based on the carbon dioxide concentration data of each belt travel section;

[0037] C6, generating belt time window carbon dioxide concentration distribution data according to the average carbon dioxide concentration in the time window corresponding to each belt travel section in sequence;

[0038] C7, determines the corresponding high temperature position according to the belt window ambient temperature distribution data;

[0039] C8, determines the corresponding high carbon monoxide concentration position based on the carbon monoxide concentration distribution data of the belt time window;

[0040] C9, determines the corresponding high-concentration position of carbon dioxide based on the carbon dioxide concentration distribution data of the belt time window;

[0041] C10, comprehensively determine the corresponding abnormal location based on the high temperature location, high carbon monoxide concentration location, and high carbon dioxide concentration location;

[0042] C11: If the abnormal location is located in a fire risk section, the fire risk section is defined as a high-risk fire section, and the abnormal location is defined as a fire risk location.

[0043] By adopting the above technical solution, the fire monitoring method based on mining belt conveyors can locate the abnormally high value areas of each parameter by constructing the ambient temperature, carbon monoxide and carbon dioxide concentration distribution data of the entire belt conveyor, and realize the positioning of fire risk positions through spatial correlation analysis of multi-parameter abnormal positions. It not only improves the spatial accuracy of fire risk identification, but also reduces the risk of false alarms through dual verification machines, providing mine rescue and maintenance personnel with more accurate fire risk location information, and significantly improving the pertinence and efficiency of emergency response.

[0044] Optionally, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0045] D1, respectively set up corresponding wireless communication modules in each belt travel section;

[0046] D2, connecting the wireless communication modules of each belt travel section wirelessly to form a data relay network;

[0047] D3, collect corresponding risk section video data in high-risk fire sections;

[0048] D4, sending the risk section video data to a preset control background via the data relay network.

[0049] By adopting the above technical solution, the fire monitoring method based on mining belt conveyors can solve the problem of long-distance data transmission in the complex environment of mines by deploying wireless communication modules in each belt travel section and building a data relay network, and realize real-time and reliable transmission of fire monitoring data. At the same time, the video monitoring function is activated for the identified high-risk fire sections to provide intuitive on-site images for the control background, which not only ensures the timeliness and integrity of the monitoring data, but also enhances the ability to confirm fire risks through visualization means, providing mine safety management personnel with more comprehensive and intuitive decision-making support, and significantly improving the accuracy and timeliness of fire emergency response.

[0050] Optionally, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0051] E1, obtain the current time and define it as the belt tail time, and calculate the corresponding belt return head time based on the belt tail time minus the belt one-way movement time;

[0052] E2, determine the corresponding belt tail monitoring time window according to the belt tail moment and the belt one-way movement time;

[0053] E3, determine the corresponding belt return head monitoring time window according to the belt return head moment and the belt one-way movement time;

[0054] E4, obtaining corresponding belt tail time window temperature data according to the belt tail monitoring time window and the tail belt temperature data;

[0055] E5, obtaining the corresponding belt return head time window temperature data according to the belt return head monitoring time window and the head belt temperature data;

[0056] E6, subtracting the corresponding belt temperature values ​​in the belt tail time window temperature data and the belt return head time window temperature data to obtain the belt return temperature difference data;

[0057] E7, calculating the corresponding belt return average temperature difference based on the average of each belt temperature difference value in the belt return temperature difference data;

[0058] E8. If the average temperature difference of the belt return is greater than a preset return temperature difference threshold, a return belt cooling operation is performed on the mining belt conveyor.

[0059] By adopting the above technical solution, the fire monitoring method based on the mining belt conveyor can realize the monitoring of the return operating temperature status of the belt conveyor by performing time-synchronized temperature difference calculation on the belt return path. When the average return temperature difference exceeds the safety threshold, the cooling operation is automatically triggered. The preventive cooling measures can effectively prevent the fire risk caused by overheating of the belt, realize the transition from passive monitoring to active protection, and provide a more complete protection strategy for the safe operation of the mining belt conveyor.

[0060] Optionally, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0061] F1, calculate the corresponding average temperature difference of the belt outward according to the average temperature difference value of each belt in the belt outward temperature difference data;

[0062] F2, calculate the corresponding belt circulation average temperature difference based on the sum of the belt outward average temperature difference and the belt return average temperature difference;

[0063] F3, determine the corresponding belt cycle time window based on twice the belt one-way moving time;

[0064] F4, collects the belt circulation average temperature difference within the belt circulation time window and generates the corresponding belt circulation average temperature difference time window data;

[0065] F5, determine the corresponding belt circulation average temperature difference trend data based on the belt circulation average temperature difference time window data;

[0066] F6, adjusts and controls the return belt cooling operation according to the trend data of the average temperature difference of the belt circulation.

[0067] By adopting the above technical solution, the fire monitoring method based on the mining belt conveyor can realize dynamic tracking and prediction of the overall thermal state of the belt conveyor by calculating the average temperature difference of the belt's outward and return cycles and monitoring the temperature difference change trend within the complete cycle time window. The cooling operation is performed based on the temperature difference trend data, which not only avoids the energy waste caused by excessive cooling, but also can perform forward-looking control according to the development trend of the belt's thermal state, making the cooling operation more accurate and efficient, realizing the intelligent and energy-saving thermal management of the belt conveyor, and providing a scientific and economical temperature control strategy for the long-term safe and stable operation of the mining belt conveyor.

[0068] The present application also provides a fire monitoring system based on a mining belt conveyor, comprising:

[0069] Multiple belt travel section monitoring modules;

[0070] Belt temperature monitoring module;

[0071] Belt cooling module;

[0072] Processing control module;

[0073] Wherein, each of the belt travel section monitoring module, the belt temperature monitoring module and the belt cooling module is data-connected to the processing control module respectively;

[0074] The belt travel section monitoring module includes a thermal imaging module, an ambient temperature monitoring module, a gas concentration monitoring module, a video monitoring module, and a wireless communication module. The thermal imaging module, the ambient temperature monitoring module, the gas concentration monitoring module, and the video monitoring module are respectively data-connected to the wireless communication module, and the wireless communication module is data-connected to the processing control module.

[0075] Wherein, the belt temperature monitoring module includes a tail belt temperature detection module and a head belt temperature detection module, the tail belt temperature detection module and the head belt temperature detection module are respectively data-connected to the processing control module, the tail belt temperature detection module is arranged at the tail of the preset mining belt conveyor, and the head belt temperature detection module is arranged at the head of the mining belt conveyor;

[0076] The fire monitoring system based on the mining belt conveyor further includes a fire monitoring strategy, including the following steps:

[0077] G1, dividing the preset total travel length of the mining belt conveyor according to the preset travel segment lengths to form each belt travel segment by the processing control module;

[0078] G2, collecting corresponding section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data in each belt travel section through each belt travel section monitoring module;

[0079] G3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor through the belt temperature monitoring module;

[0080] G4, collecting corresponding head belt temperature data at the head of the mining belt conveyor through the belt temperature monitoring module;

[0081] G5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data;

[0082] G6, the processing control module generates a corresponding section fire probability value based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data and belt forward temperature difference data corresponding to each belt travel section using a preset fire identification algorithm;

[0083] G7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section.

[0084] By adopting the above technical solution, the fire monitoring system based on the mining belt conveyor can realize the positioning and early warning of the fire risk of the mining belt conveyor by monitoring the total stroke of the belt conveyor in sections, combining multi-dimensional data collection such as thermal imaging, ambient temperature, carbon monoxide and carbon dioxide concentrations, and the temperature difference between the head and tail, thereby improving the accuracy and reliability of fire detection, being able to identify potential fire sections in a timely manner, and providing more comprehensive and intelligent protection for mine safety production.

[0085] In summary, this application includes at least one of the following beneficial technical effects:

[0086] 1. By monitoring the total travel of the belt conveyor in sections and combining multi-dimensional data collection such as thermal imaging, ambient temperature, carbon monoxide and carbon dioxide concentrations, as well as the temperature difference between the head and tail, the fire risk of the mining belt conveyor can be located and warned. This improves the accuracy and reliability of fire detection, enables timely identification of potential fire sections, and provides more comprehensive and intelligent protection for mine safety production.

[0087] 2. By calculating the one-way movement duration of the belt and setting the monitoring time windows at the head and tail based on the principle of time synchronization, it is ensured that the temperature data at the head and tail of the same section of the belt can accurately correspond, thereby calculating the forward temperature difference data that truly reflects the temperature changes of the belt during operation, effectively eliminating the data misalignment problem caused by belt movement, improving the accuracy and reliability of the temperature difference data, and providing more accurate basic data support for subsequent fire risk judgment.

[0088] 3. By extracting and combining thermal imaging image features with multi-dimensional sensor data such as ambient temperature, gas concentration, and belt temperature difference, a comprehensive feature data set can be formed and input into a pre-trained machine learning model to achieve diversified variable analysis. This not only improves the accuracy and robustness of fire identification, but also effectively reduces the false alarm rate. Through the adaptive ability of the deep learning model, it can better identify fire signs in complex mining environments, providing more reliable and intelligent technical support for mine belt conveyor fire warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 It is a process schematic diagram of a fire monitoring method based on a mining belt conveyor of the present invention.

[0090] Figure 2 The diagram is a schematic diagram of the principle of a fire monitoring system based on a mining belt conveyor according to the present invention. DETAILED DESCRIPTION

[0091] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0092] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0093] refer to Figure 1 The present invention provides a fire monitoring method based on a mining belt conveyor, which is used to monitor the fire disaster of the mining belt conveyor, comprising the following steps:

[0094] A1, dividing the preset total travel length of the mining belt conveyor according to the preset travel section length to form each belt travel section;

[0095] The travel segment length is a pre-set length value used to divide the mining belt conveyor into multiple segments and perform fire monitoring on each segment. Since mining belt conveyors are long, typically hundreds to thousands of meters long, it is difficult to perform global monitoring, so segment-by-segment monitoring is required.

[0096] The total stroke length is the total length of the mineral transported by the mining belt conveyor;

[0097] The belt travel sections are the divided sections of the mining belt conveyor.

[0098] A2, collects corresponding section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data in each belt travel section;

[0099] The section thermal imaging data is the thermal imaging data corresponding to the belt travel section;

[0100] The section ambient temperature data is the ambient temperature data corresponding to the belt travel section;

[0101] The section carbon monoxide concentration data is the carbon monoxide concentration data corresponding to the belt travel section;

[0102] The section carbon dioxide concentration data is the carbon dioxide concentration data corresponding to the belt travel section.

[0103] A3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor;

[0104] The tail belt temperature data is the surface temperature data of the belt at the tail of the mining belt conveyor. The tail of the mining belt conveyor is the starting end of ore transportation.

[0105] A4, collecting corresponding head belt temperature data at the head of the mining belt conveyor;

[0106] The head belt temperature data is the surface temperature data of the belt at the head of the mining belt conveyor, and the head of the mining belt conveyor is the end of ore transportation.

[0107] A5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data;

[0108] The temperature difference data of the belt's forward journey is the temperature difference between the tail belt temperature data and the head belt temperature data, that is, the temperature change value that occurs when the belt loads ore and runs from the tail to the head, which can reflect the temperature change of the belt during the ore transportation process.

[0109] A6 generates a corresponding fire probability value for each belt travel section using a preset fire identification algorithm based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data, and belt forward temperature difference data corresponding to each belt travel section;

[0110] The fire recognition algorithm is a pre-set algorithm that can judge the fire situation based on data collected from various aspects. The fire recognition algorithm can be a self-designed algorithm or a recognition model generated by training based on sample data;

[0111] The section fire probability value is the probability value of fire occurring in the corresponding section.

[0112] A7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section;

[0113] The fire warning threshold is a pre-set reference value used to determine the fire probability value of a section;

[0114] The fire risk section is a belt travel section where the fire probability value is greater than the fire warning threshold, that is, a section with a higher probability of fire occurring.

[0115] Through the above steps, the fire monitoring method based on the mining belt conveyor can realize the positioning and early warning of the fire risk of the mining belt conveyor by monitoring the total stroke of the belt conveyor in sections, combining multi-dimensional data collection such as thermal imaging, ambient temperature, carbon monoxide and carbon dioxide concentrations, and the temperature difference between the head and tail, thereby improving the accuracy and reliability of fire detection, being able to timely identify potential fire sections, and providing more comprehensive and intelligent protection for mine safety production.

[0116] Furthermore, the step A5 includes the following steps:

[0117] A501 calculates the corresponding one-way belt travel time based on the quotient of the total travel length and the preset belt running speed;

[0118] The belt running speed is the pre-set moving speed of the belt transporting the minerals, i.e. the belt speed;

[0119] The one-way movement time of the belt is the time required for the same section of the belt to move from the tail to the head.

[0120] A502 obtains the current time and defines it as the belt head time. The corresponding belt tail time is calculated by subtracting the belt one-way travel time from the belt head time.

[0121] The belt head moment is the current moment;

[0122] The time at the tail end of the belt's outward journey is the time obtained by subtracting the one-way movement time of the belt from the head end of the belt, that is, the time when the part of the belt currently located at the head of the mining belt conveyor is at the tail end. For example, the current time is 10:30:00, and the one-way movement time of the belt is 40 seconds, then the corresponding time at the tail end of the belt's outward journey is 10:23:20.

[0123] A503 determines the corresponding belt head monitoring time window based on the belt head moment and the belt one-way movement time;

[0124] The belt head monitoring time window is the time range from the belt head moment to the one-way movement duration of the belt. For example, if the belt head moment is 10:30:00 and the one-way movement duration of the belt is 400 seconds, the corresponding belt head monitoring time window is 10:23:20 to 10:30:00.

[0125] A504 determines the corresponding monitoring time window for the belt's outward tail according to the belt's outward tail moment and the belt's one-way travel time.

[0126] The monitoring window for the outgoing tail of the belt is the time range from the end of the outgoing belt to the one-way movement time of the belt. For example, if the end of the outgoing belt is 10:23:20 and the one-way movement time of the belt is 400 seconds, the corresponding monitoring window for the outgoing tail of the belt is 10:16:40 to 10:23:20.

[0127] A505, obtains the corresponding belt head time window temperature data according to the belt head monitoring time window and the belt head temperature data;

[0128] The belt head time window temperature data is the belt temperature data corresponding to the belt head monitoring time window in the head belt temperature data.

[0129] A506, based on the belt outward tail monitoring time window and the tail belt temperature data, obtains the corresponding belt outward tail time window temperature data;

[0130] The belt outward tail time window temperature data is the belt temperature data corresponding to the belt outward tail monitoring time window in the tail belt temperature data.

[0131] A507, subtracts the corresponding belt temperature values ​​in the belt head window temperature data and the belt forward tail window temperature data to obtain the belt forward temperature difference data;

[0132] The belt outward temperature difference data is a data set of the difference between the corresponding belt temperature values ​​in the belt head time window temperature data and the belt outward tail time window temperature data, that is, the belt temperature values ​​of the same order in the belt head time window temperature data and the belt outward tail time window temperature data are subtracted. For example, the first belt temperature value in the belt head time window temperature data is subtracted from the first belt temperature value in the belt outward tail time window temperature data to obtain the first belt temperature difference value. The differences are calculated in sequence to finally obtain the belt outward temperature difference data.

[0133] Through the above steps, the fire monitoring method based on the mining belt conveyor can calculate the one-way movement time of the belt and set the monitoring time windows at the head and tail based on the principle of time synchronization, thereby ensuring that the temperature data of the same section of the belt at the head and tail can accurately correspond, thereby calculating the outbound temperature difference data that truly reflects the temperature changes of the belt during operation, effectively eliminating the data misalignment problem caused by belt movement, improving the accuracy and reliability of the temperature difference data, and providing more accurate basic data support for subsequent fire risk judgment.

[0134] Furthermore, the fire identification algorithm includes the following steps:

[0135] B1, generating corresponding segment thermal imaging feature data through a preset image feature extraction algorithm based on the segment thermal imaging data;

[0136] The image feature extraction algorithm is a pre-set feature extraction algorithm used to extract image feature data from a thermal image, for example, a feature extraction algorithm based on the grayscale value of a thermal imaging grayscale image to reflect the temperature feature in the image;

[0137] The segment thermal imaging feature data is feature data corresponding to the segment thermal imaging image data.

[0138] B2, according to the belt head monitoring time window, obtain the corresponding section time window temperature data, section time window carbon monoxide concentration data and section time window carbon dioxide concentration data respectively from the section ambient temperature data, section time window carbon monoxide concentration data and section time window carbon dioxide concentration data;

[0139] The section time window temperature data is the data corresponding to the belt head monitoring time window in the section ambient temperature data;

[0140] The section time window carbon monoxide concentration data is the data corresponding to the belt head monitoring time window in the section carbon monoxide concentration data;

[0141] The section time window carbon dioxide concentration data is the data corresponding to the belt head monitoring time window in the section carbon dioxide concentration data.

[0142] B3, based on the section time window temperature data, the section time window carbon monoxide concentration data, the section time window carbon dioxide concentration data and the belt forward temperature difference data, respectively, generates corresponding section temperature characteristic data, section carbon monoxide concentration characteristic data, section carbon dioxide concentration characteristic data and belt forward temperature difference characteristic data through a preset data feature extraction algorithm;

[0143] The data feature extraction algorithm is a pre-set feature extraction algorithm used to extract feature data from the data;

[0144] The section temperature characteristic data is the characteristic data corresponding to the section time window temperature data;

[0145] The segment carbon monoxide concentration characteristic data is the characteristic data corresponding to the segment time window carbon monoxide concentration data;

[0146] The segment carbon dioxide concentration characteristic data is the characteristic data corresponding to the segment time window carbon dioxide concentration data;

[0147] The belt forward temperature difference characteristic data is characteristic data corresponding to the belt forward temperature difference data.

[0148] B4, generating a comprehensive segment feature data set based on the segment thermal imaging feature data, segment temperature feature data, segment carbon monoxide concentration feature data, segment carbon dioxide concentration feature data, and belt forward temperature difference feature data;

[0149] The section comprehensive feature data set includes the section thermal imaging feature data, section temperature feature data, section carbon monoxide concentration feature data, section carbon dioxide concentration feature data and belt forward temperature difference feature data.

[0150] B5, input the pre-trained fire probability recognition model according to the segment comprehensive feature dataset to generate the corresponding segment fire probability value;

[0151] The fire probability recognition model is a pre-trained learning model that can output the corresponding segment fire probability value based on the input segment comprehensive feature data set; the fire probability recognition model can be trained based on collected historical fire data or simulated fire data collected in an experimental environment.

[0152] Through the above steps, the fire monitoring method based on mining belt conveyors can extract and combine the thermal imaging image features with multi-dimensional sensor data such as ambient temperature, gas concentration, and belt temperature difference to form a comprehensive feature data set, which is then input into a pre-trained machine learning model to achieve diversified variable analysis. This not only improves the accuracy and robustness of fire identification, but also effectively reduces the false alarm rate. Through the adaptive ability of the deep learning model, it can better identify fire signs in complex mining environments, providing more reliable and intelligent technical support for mine belt conveyor fire early warning.

[0153] Furthermore, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0154] C1, calculate the corresponding section time window average temperature based on the section time window temperature data corresponding to each belt travel section;

[0155] The section time window average temperature is the average value of each sample data in the section time window temperature data corresponding to the belt travel section.

[0156] C2, generating belt time window ambient temperature distribution data according to the average temperature of the section time window corresponding to each belt travel section;

[0157] The belt time window ambient temperature distribution data is the data formed by sequentially combining the average time window temperatures corresponding to each belt travel section, reflecting the temperature distribution of each section; for example, the belt time window ambient temperature distribution data can be obtained by arranging the average time window temperatures of each section in sequence from the tail to the head of the belt conveyor.

[0158] C3, calculate the average carbon monoxide concentration of the corresponding section time window according to the carbon monoxide concentration data of each belt travel section;

[0159] The average carbon monoxide concentration in the segment time window is the average value of each sample data in the segment time window carbon monoxide concentration data corresponding to the belt travel segment.

[0160] C4, generating belt time window carbon monoxide concentration distribution data according to the average carbon monoxide concentration in the time window corresponding to each belt travel section;

[0161] The belt time window carbon monoxide concentration distribution data is data formed by sequentially combining the average carbon monoxide concentrations in the section time windows corresponding to each belt travel section, reflecting the carbon monoxide concentration distribution in each section.

[0162] C5, calculate the average carbon dioxide concentration of the corresponding section time window based on the carbon dioxide concentration data of each belt travel section;

[0163] The average carbon dioxide concentration in the segment time window is the average value of each sample data in the segment time window carbon dioxide concentration data corresponding to the belt travel segment.

[0164] C6, generating belt time window carbon dioxide concentration distribution data according to the average carbon dioxide concentration in the time window corresponding to each belt travel section in sequence;

[0165] The belt time window carbon dioxide concentration distribution data is data formed by sequentially combining the average carbon dioxide concentrations in the section time windows corresponding to each belt travel section, reflecting the carbon dioxide concentration distribution in each section.

[0166] C7, determines the corresponding high temperature position according to the belt window ambient temperature distribution data;

[0167] The high-temperature position is the estimated position with the highest temperature on the mining belt conveyor. The corresponding curve can be fitted through the belt time window ambient temperature distribution data and a peak value can be estimated in the curve. The corresponding high-temperature position can be obtained by further judging the specific position of the section. For example, each section is arranged in order according to its length to form an X-axis, and the temperature value in the belt time window ambient temperature distribution data is used as the Y-axis. Then, a corresponding curve can be fitted according to the belt time window ambient temperature distribution data. Assuming that its peak value falls on the third section of the mining belt conveyor, the specific position of the high-temperature position in the third section can be estimated based on the X-axis span of the third section and the X-axis position corresponding to the peak value.

[0168] C8, determines the corresponding high carbon monoxide concentration position based on the carbon monoxide concentration distribution data of the belt time window;

[0169] The carbon monoxide high concentration position is an estimated position where the carbon monoxide concentration of the mining belt conveyor is the highest.

[0170] C9, determines the corresponding high-concentration position of carbon dioxide based on the carbon dioxide concentration distribution data of the belt time window;

[0171] The carbon dioxide high concentration position is an estimated position of the mining belt conveyor where the carbon dioxide concentration is the highest.

[0172] C10, comprehensively determine the corresponding abnormal location based on the high temperature location, high carbon monoxide concentration location, and high carbon dioxide concentration location;

[0173] The abnormal position is comprehensively judged based on the high temperature position, high carbon monoxide concentration position and high carbon dioxide concentration position. For example, the center position of the three is taken, or considering that the belt operation will drive the air flow movement and affect the carbon monoxide concentration and carbon dioxide concentration, the position can be estimated and corrected through a certain algorithm.

[0174] C11: If the abnormal location is located in a fire risk section, define the fire risk section as a high-risk fire section, and define the abnormal location as a fire risk location;

[0175] High-risk fire areas are areas where fire or fire hazards are more likely to occur;

[0176] Fire risk locations are locations in high-risk fire areas where fires may occur or where fire hazards exist.

[0177] Through the above steps, the fire monitoring method based on mining belt conveyors can locate the abnormally high value areas of each parameter by constructing the ambient temperature, carbon monoxide and carbon dioxide concentration distribution data of the entire belt conveyor, and realize the fire risk location through spatial correlation analysis of multi-parameter abnormal positions. It not only improves the spatial accuracy of fire risk identification, but also reduces the risk of false alarms through double verification machines, providing mine rescue and maintenance personnel with more accurate fire risk location information, and significantly improving the pertinence and efficiency of emergency response.

[0178] Furthermore, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0179] D1, respectively set up corresponding wireless communication modules in each belt travel section;

[0180] An independent wireless communication module is provided in each belt travel section.

[0181] D2, connecting the wireless communication modules of each belt travel section wirelessly to form a data relay network;

[0182] The data relay network is a network formed by the mutual data connection between the wireless communication modules of each belt travel section. Since the mining belt conveyor is usually a linear structure and the distance span is large, an effective communication line can be formed through the wireless communication modules of each belt travel section. For example, the wireless communication modules of two adjacent belt travel sections can be connected to each other for communication, thereby realizing effective data transmission.

[0183] D3, collect corresponding risk section video data in high-risk fire sections;

[0184] Risk section video data refers to video data collected in high-risk fire sections.

[0185] D4, sending the risk section video data to a preset control background via the data relay network;

[0186] The video data of the risk section is sent to the control background for staff to confirm and make decisions.

[0187] Through the above steps, the fire monitoring method based on the mine belt conveyor can solve the problem of long-distance data transmission in the complex environment of the mine by deploying wireless communication modules in each belt travel section and building a data relay network, realize real-time and reliable transmission of fire monitoring data, and start video monitoring function for the identified high-risk fire section, providing intuitive on-site pictures for the control background. Not only ensures the timeliness and integrity of the monitoring data, but also enhances the confirmation ability of fire risk through visual means, provides more comprehensive and intuitive decision support for mine safety management personnel, and significantly improves the accuracy and timeliness of fire emergency response.

[0188] Further, the fire monitoring method based on the mine belt conveyor further comprises the following steps:

[0189] E1, obtaining the current time and defining it as the tail time of the belt, and calculating the corresponding head time of the return trip of the belt by subtracting the single-trip moving time of the belt from the tail time of the belt;

[0190] The tail time of the belt is the current time.

[0191] The head time of the return trip of the belt is the time obtained by subtracting the single-trip moving time of the belt from the tail time of the belt, that is, the time when the part of the belt currently located at the tail of the mine belt conveyor is located at the head. The principle is the same as the example in the description of step A502.

[0192] E2, determining the corresponding tail monitoring time window of the belt according to the tail time of the belt and the single-trip moving time of the belt;

[0193] The tail monitoring time window of the belt is the time range from the tail time of the belt back to the single-trip moving time of the belt. For reference, see the example in the description of step A503.

[0194] E3, determining the corresponding head monitoring time window of the return trip of the belt according to the head time of the return trip of the belt and the single-trip moving time of the belt;

[0195] The head monitoring time window of the return trip of the belt is the time range from the head time of the return trip of the belt back to the single-trip moving time of the belt. For reference, see the example in the description of step A504.

[0196] E4, obtaining the corresponding tail time window temperature data of the belt according to the tail monitoring time window of the belt and the tail belt temperature data;

[0197] The tail time window temperature data of the belt is the corresponding belt temperature data in the tail monitoring time window of the belt and the tail belt temperature data.

[0198] E5, obtaining the corresponding head time window temperature data of the return trip of the belt according to the head monitoring time window of the return trip of the belt and the head belt temperature data;

[0199] The belt return head time window temperature data is the belt temperature data corresponding to the belt return head monitoring time window in the head belt temperature data.

[0200] E6, subtract the corresponding belt temperature value in the belt tail time window temperature data and the belt return head time window temperature data to obtain the belt return temperature difference data;

[0201] The belt return temperature difference data is a data set of the difference between the corresponding belt temperature values ​​in the belt tail time window temperature data and the belt return head time window temperature data, that is, the belt temperature values ​​of the same order in the belt tail time window temperature data and the belt return head time window temperature data are subtracted. Please refer to the example in the step A507 description.

[0202] E7, calculating the corresponding belt return average temperature difference based on the average of each belt temperature difference value in the belt return temperature difference data;

[0203] The average belt return temperature difference is the average of the belt temperature difference values ​​in the belt return temperature difference data.

[0204] E8, if the average temperature difference of the belt return is greater than a preset return temperature difference threshold, performing a return belt cooling operation on the mining belt conveyor;

[0205] The return temperature difference threshold is a pre-set reference value used to determine the size of the average temperature difference of the belt return;

[0206] Normally, after the return belt unloads the mineral load, its surface temperature will gradually decrease until it returns to the tail and reloads the mineral load. In general, the belt temperature at the tail of the belt conveyor is lower than that at the head, that is, the average temperature difference of the belt return is a negative number. When the average temperature difference of the belt return is positive or close to 0, it means that there is some reason that causes heat to accumulate on the belt. If the belt cooling intervention is not carried out in time, the continuous accumulation of heat may cause a fire hazard.

[0207] The return belt cooling operation can be carried out in various feasible ways. For example, by blowing cold air across the surface of the return belt, or by flushing the surface of the return belt with cold water, the surface temperature of the belt can be reduced. On the other hand, the attached mineral debris can be removed to reduce the load of the belt conveyor. Furthermore, the temperature in the mine can be further appropriately reduced.

[0208] Through the above steps, the fire monitoring method based on the mining belt conveyor can realize the monitoring of the return operating temperature status of the belt conveyor by performing time-synchronized temperature difference calculation on the belt return path. When the average return temperature difference exceeds the safety threshold, the cooling operation is automatically triggered. The preventive cooling measures can effectively prevent the fire risk caused by belt overheating, realize the transition from passive monitoring to active protection, and provide a more complete protection strategy for the safe operation of the mining belt conveyor.

[0209] Furthermore, the fire monitoring method based on a mining belt conveyor further comprises the following steps:

[0210] F1, calculate the corresponding average temperature difference of the belt outward according to the average temperature difference value of each belt in the belt outward temperature difference data;

[0211] The average temperature difference of the belt forward is the average of the temperature difference values ​​of each belt in the belt forward temperature difference data.

[0212] F2, calculate the corresponding belt circulation average temperature difference based on the sum of the belt outward average temperature difference and the belt return average temperature difference;

[0213] The average temperature difference of the belt circulation is the sum of the average temperature difference of the belt's outward journey and the average temperature difference of the belt's return journey. It can reflect the temperature change on the belt surface of the belt conveyor during a reciprocating cycle. When the average temperature difference of the belt circulation is greater than 0, it means that heat continues to accumulate on the belt.

[0214] F3, determine the corresponding belt cycle time window based on twice the belt one-way moving time;

[0215] The belt cycle time window is the time window corresponding to twice the one-way movement time of the belt, which represents the time window required for the belt conveyor to complete one reciprocating cycle.

[0216] F4, collects the belt circulation average temperature difference within the belt circulation time window and generates the corresponding belt circulation average temperature difference time window data;

[0217] The belt circulation average temperature difference time window data is a data set of the belt circulation average temperature difference within the belt circulation time window.

[0218] F5, determine the corresponding belt circulation average temperature difference trend data based on the belt circulation average temperature difference time window data;

[0219] The belt circulation average temperature difference trend data is a change trend of the belt circulation average temperature difference time window data. For example, a linear equation is fitted based on the belt circulation average temperature difference time window data and a corresponding trend slope is calculated.

[0220] F6, adjusts and controls the return belt cooling operation according to the trend data of the average temperature difference of the belt cycle;

[0221] Adjust the cooling intensity of the return belt cooling operation based on the belt circulation average temperature difference trend data.

[0222] Through the above steps, the fire monitoring method based on the mining belt conveyor can realize dynamic tracking and prediction of the overall thermal state of the belt conveyor by calculating the average temperature difference of the belt's outward and return cycles and monitoring the temperature difference change trend within the complete cycle time window. The cooling operation is performed based on the temperature difference trend data, which not only avoids the energy waste caused by excessive cooling, but also can perform forward-looking control according to the development trend of the belt's thermal state, making the cooling operation more accurate and efficient, realizing the intelligent and energy-saving thermal management of the belt conveyor, and providing a scientific and economical temperature control strategy for the long-term safe and stable operation of the mining belt conveyor.

[0223] refer to Figure 2 The present invention also provides a fire monitoring system based on a mining belt conveyor, comprising:

[0224] Multiple belt travel section monitoring modules 10;

[0225] Belt temperature monitoring module 20;

[0226] Belt cooling module 30;

[0227] Processing control module 40;

[0228] Wherein, each of the belt travel section monitoring module 10, the belt temperature monitoring module 20 and the belt cooling module 30 is respectively data-connected to the processing control module 40;

[0229] The belt travel section monitoring module 10 includes a thermal imaging module 11, an ambient temperature monitoring module 12, a gas concentration monitoring module 13, a video monitoring module 14, and a wireless communication module 15. The thermal imaging module 11, the ambient temperature monitoring module 12, the gas concentration monitoring module 13, and the video monitoring module 14 are respectively data-connected to the wireless communication module 15, and the wireless communication module 15 is data-connected to the processing control module 40.

[0230] The belt temperature monitoring module 20 includes a tail belt temperature detection module 21 and a head belt temperature detection module 22. The tail belt temperature detection module 21 and the head belt temperature detection module 22 are respectively data-connected to the processing control module 40. The tail belt temperature detection module 21 is set at the tail of the preset mining belt conveyor, and the head belt temperature detection module 22 is set at the head of the mining belt conveyor.

[0231] The fire monitoring system based on the mining belt conveyor further includes a fire monitoring strategy, including the following steps:

[0232] G1, dividing the preset total travel length of the mining belt conveyor according to the preset travel segment lengths to form each belt travel segment through the processing control module 40;

[0233] G2, collecting corresponding segment thermal imaging data, segment ambient temperature data, segment carbon monoxide concentration data and segment carbon dioxide concentration data in each belt travel segment through each of the belt travel segment monitoring modules 10;

[0234] G3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor through the belt temperature monitoring module 20;

[0235] G4, collecting corresponding head belt temperature data at the head of the mining belt conveyor through the belt temperature monitoring module 20;

[0236] G5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data;

[0237] G6, the processing control module 40 generates a corresponding section fire probability value based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data, and belt forward temperature difference data corresponding to each belt travel section using a preset fire identification algorithm;

[0238] G7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section.

[0239] Through the above technical solution, the fire monitoring system based on the mining belt conveyor can realize the positioning and early warning of the fire risk of the mining belt conveyor by monitoring the total stroke of the belt conveyor in sections, combining multi-dimensional data collection such as thermal imaging, ambient temperature, carbon monoxide and carbon dioxide concentrations, and the temperature difference between the head and tail, thereby improving the accuracy and reliability of fire detection, being able to identify potential fire sections in a timely manner, and providing more comprehensive and intelligent protection for mine safety production.

[0240] The above are all preferred embodiments of the present application and are not intended to limit the scope of protection of this application. Unless otherwise specified, any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features. In other words, unless otherwise specified, each feature is merely an example of a series of equivalent or similar features.

Claims

1. A fire monitoring method based on a mining belt conveyor, characterized in that: The following steps are involved: A1, dividing the preset total travel length of the mining belt conveyor according to the preset travel section length to form each belt travel section; A2, collects corresponding section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data in each belt travel section; A3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor; A4, collecting corresponding head belt temperature data at the head of the mining belt conveyor; A5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data; A6 generates a corresponding fire probability value for each belt travel section using a preset fire identification algorithm based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data, and belt forward temperature difference data corresponding to each belt travel section; A7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section; Step A5 includes the following steps: A501 calculates the corresponding one-way belt travel time based on the quotient of the total travel length and the preset belt running speed; A502 obtains the current time and defines it as the belt head time. The corresponding belt tail time is calculated by subtracting the belt one-way travel time from the belt head time. A503 determines the corresponding belt head monitoring time window based on the belt head moment and the belt one-way movement time; A504 determines the corresponding monitoring time window for the belt's outward tail according to the belt's outward tail moment and the belt's one-way travel time. A505, obtains the corresponding belt head time window temperature data according to the belt head monitoring time window and the belt head temperature data; A506, based on the belt forward tail monitoring time window and the tail belt temperature data, obtains the corresponding belt forward tail time window temperature data; A507, subtracts the corresponding belt temperature values ​​in the belt head window temperature data and the belt forward tail window temperature data to obtain the belt forward temperature difference data; The fire identification algorithm includes the following steps: B1, generating corresponding segment thermal imaging feature data through a preset image feature extraction algorithm based on the segment thermal imaging data; B2, according to the belt head monitoring time window, obtain the corresponding section time window temperature data, section time window carbon monoxide concentration data and section time window carbon dioxide concentration data respectively from the section ambient temperature data, section time window carbon monoxide concentration data and section time window carbon dioxide concentration data; B3, based on the section time window temperature data, the section time window carbon monoxide concentration data, the section time window carbon dioxide concentration data and the belt forward temperature difference data, respectively, generates corresponding section temperature characteristic data, section carbon monoxide concentration characteristic data, section carbon dioxide concentration characteristic data and belt forward temperature difference characteristic data through a preset data feature extraction algorithm; B4, generating a comprehensive segment feature data set based on the segment thermal imaging feature data, segment temperature feature data, segment carbon monoxide concentration feature data, segment carbon dioxide concentration feature data, and belt forward temperature difference feature data; B5, input the pre-trained fire probability recognition model according to the segment comprehensive feature dataset to generate the corresponding segment fire probability value.

2. The fire monitoring method based on a mining belt conveyor according to claim 1, characterized in that: Further comprising the steps of: C1, calculate the corresponding section time window average temperature based on the section time window temperature data corresponding to each belt travel section; C2, generating belt time window ambient temperature distribution data according to the average temperature of the section time window corresponding to each belt travel section; C3, calculate the average carbon monoxide concentration of the corresponding section time window according to the carbon monoxide concentration data of each belt travel section; C4, generating belt time window carbon monoxide concentration distribution data according to the average carbon monoxide concentration in the time window corresponding to each belt travel section; C5, calculate the average carbon dioxide concentration of the corresponding section time window based on the carbon dioxide concentration data of each belt travel section; C6, generating belt time window carbon dioxide concentration distribution data according to the average carbon dioxide concentration of the section time window corresponding to each belt travel section in sequence; C7, determines the corresponding high temperature position according to the belt window ambient temperature distribution data; C8, determines the corresponding high carbon monoxide concentration position based on the carbon monoxide concentration distribution data of the belt time window; C9, determines the corresponding high-concentration position of carbon dioxide based on the carbon dioxide concentration distribution data of the belt time window; C10, comprehensively determine the corresponding abnormal location based on the high temperature location, high carbon monoxide concentration location, and high carbon dioxide concentration location; C11: If the abnormal location is located in a fire risk section, the fire risk section is defined as a high-risk fire section, and the abnormal location is defined as a fire risk location.

3. The fire monitoring method based on a mining belt conveyor according to claim 2, characterized in that: Further comprising the steps of: D1, respectively set up corresponding wireless communication modules in each belt travel section; D2, connecting the wireless communication modules of each belt travel section wirelessly to form a data relay network; D3, collect corresponding risk section video data in high-risk fire sections; D4, sending the risk section video data to a preset control background via the data relay network.

4. The fire monitoring method based on a mining belt conveyor according to claim 2, characterized in that: Further comprising the steps of: E1, obtain the current time and define it as the belt tail time, and calculate the corresponding belt return head time based on the belt tail time minus the belt one-way movement time; E2, determine the corresponding belt tail monitoring time window according to the belt tail moment and the belt one-way movement time; E3, determine the corresponding belt return head monitoring time window according to the belt return head moment and the belt one-way movement time; E4, obtaining corresponding belt tail time window temperature data according to the belt tail monitoring time window and the tail belt temperature data; E5, obtaining the corresponding belt return head time window temperature data according to the belt return head monitoring time window and the head belt temperature data; E6, subtract the corresponding belt temperature value in the belt tail time window temperature data and the belt return head time window temperature data to obtain the belt return temperature difference data; E7, calculating the corresponding belt return average temperature difference based on the average of each belt temperature difference value in the belt return temperature difference data; E8. If the average temperature difference of the belt return is greater than a preset return temperature difference threshold, a return belt cooling operation is performed on the mining belt conveyor.

5. The fire monitoring method based on a mining belt conveyor according to claim 4, characterized in that: Further comprising the steps of: F1, calculate the corresponding average temperature difference of the belt outward according to the average temperature difference value of each belt in the belt outward temperature difference data; F2, calculate the corresponding belt circulation average temperature difference based on the sum of the belt outward average temperature difference and the belt return average temperature difference; F3, determine the corresponding belt cycle time window based on twice the belt one-way moving time; F4, collects the belt circulation average temperature difference within the belt circulation time window and generates the corresponding belt circulation average temperature difference time window data; F5, determine the corresponding belt circulation average temperature difference trend data based on the belt circulation average temperature difference time window data; F6, adjusts and controls the return belt cooling operation according to the trend data of the average temperature difference of the belt circulation.

6. A fire monitoring system based on a mining belt conveyor, characterized in that: include: Multiple belt travel section monitoring modules; Belt temperature monitoring module; Belt cooling module; Processing control module; Wherein, each of the belt travel section monitoring module, the belt temperature monitoring module and the belt cooling module is data-connected to the processing control module respectively; The belt travel section monitoring module includes a thermal imaging module, an ambient temperature monitoring module, a gas concentration monitoring module, a video monitoring module, and a wireless communication module. The thermal imaging module, the ambient temperature monitoring module, the gas concentration monitoring module, and the video monitoring module are respectively data-connected to the wireless communication module, and the wireless communication module is data-connected to the processing control module. Wherein, the belt temperature monitoring module includes a tail belt temperature detection module and a head belt temperature detection module, the tail belt temperature detection module and the head belt temperature detection module are respectively data-connected to the processing control module, the tail belt temperature detection module is arranged at the tail of the preset mining belt conveyor, and the head belt temperature detection module is arranged at the head of the mining belt conveyor; The fire monitoring system based on the mining belt conveyor further includes a fire monitoring strategy, including the following steps: G1, dividing the preset total travel length of the mining belt conveyor according to the preset travel segment lengths to form each belt travel segment by the processing control module; G2, collecting corresponding section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data and section carbon dioxide concentration data in each belt travel section through each belt travel section monitoring module; G3, collecting corresponding tail belt temperature data at the tail of the mining belt conveyor through the belt temperature monitoring module; G4, collecting corresponding head belt temperature data at the head of the mining belt conveyor through the belt temperature monitoring module; G5, calculates the corresponding belt forward temperature difference data based on the tail belt temperature data and the head belt temperature data; G6, the processing control module generates a corresponding section fire probability value based on the section thermal imaging data, section ambient temperature data, section carbon monoxide concentration data, section carbon dioxide concentration data and belt forward temperature difference data corresponding to each belt travel section using a preset fire identification algorithm; G7: If the fire probability value of a section is greater than the preset fire warning threshold, the corresponding belt travel section is defined as a fire risk section; Calculate the corresponding one-way movement time of the belt based on the quotient of the total stroke length and the preset belt running speed; Get the current time and define it as the belt head time, and calculate the corresponding belt outward tail time based on the belt head time minus the belt one-way travel time; Determine the corresponding belt head monitoring time window according to the belt head moment and the belt one-way movement time; Determine the corresponding monitoring time window of the belt's outward tail according to the belt's outward tail moment and the belt's one-way movement time; Obtain corresponding belt head time window temperature data according to the belt head monitoring time window and the head belt temperature data; Obtain the corresponding belt outward tail time window temperature data according to the belt outward tail monitoring time window and the tail belt temperature data; The belt forward temperature difference data is obtained by subtracting the corresponding belt temperature values ​​in the belt head time window temperature data and the belt forward tail time window temperature data.

Images