A coal conveying gallery fire early warning system and method

Abstract

The application discloses a coal conveying gallery fire early warning system and method, and the early warning method comprises the following steps: acquiring infrared thermal imaging video and image video in the coal conveying gallery; performing structured processing on the infrared thermal imaging video, and converting the processed infrared thermal imaging video into data arranged in a matrix mode to form an infrared matrix; excluding the highest temperature and the lowest temperature of interference items of the coal conveying gallery in the infrared matrix; and locating the highest temperature square point in the infrared matrix. The application obtains temperature data in the coal conveying gallery in a matrix format through infrared thermal imaging, establishes a temperature matrix, and the matrix data after excluding the highest temperature Tmax and the lowest temperature Tmin can more accurately reflect the actual temperature distribution in the coal conveying gallery; meanwhile, the application finds the highest temperature square point in the temperature matrix, compares the highest temperature square point with a corresponding area in the infrared video, and can obtain comprehensive information of temperature change of the area.

Description

TECHNICAL FIELD

[0001] The present application relates to a coal conveying gallery fire early warning system and method. BACKGROUND

[0002] The coal conveying system plays a vital role in thermal power generation, which provides fast and stable raw coal supply for the unit, which is one of the prerequisites for stable operation of the unit. However, the coal conveying system often needs to deal with various challenges during operation. One of the most serious problems is the fire risk. Due to the harsh working environment of the coal conveying system, easy wear and tear and other factors, early identification and early warning of equipment abnormalities, defects and failures become extremely difficult, thereby increasing the risk of fire occurring without timely discovery. This has brought serious hidden dangers to the safety production of coal-fired power plants.

[0003] Before the fire occurs, the fire source point often shows abnormal changes such as high temperature or gradual temperature rise. If it can be discovered and handled at an extremely early stage, it can avoid equipment damage caused by fire and reduce economic losses suffered by enterprises. SUMMARY

[0004] The main purpose of the present application is to provide a coal conveying gallery fire early warning system and method to achieve the purpose of early discovery of the fire source.

[0005] The purpose of the present application can be achieved by adopting the following technical solutions:

[0006] A coal conveying gallery fire early warning method, comprising the following steps:

[0007] Obtain infrared thermal imaging video and image video in the coal conveying gallery;

[0008] Structurally process the infrared thermal imaging video, and convert it into data arranged in a matrix form after processing to constitute an infrared matrix;

[0009] Exclude the highest temperature and the lowest temperature of the interference items of the coal conveying gallery in the infrared matrix;

[0010] Position the square point of the highest temperature in the infrared matrix, and calculate the temperature rising speed and the maximum temperature of the square point within a preset time;

[0011] Record the temperature rising speed of the surrounding square points of the square point of the highest temperature, and divide the surrounding square points based on the temperature rising speed;

[0012] Record the temperature diffusion of the surrounding square points of the square point of the highest temperature;

[0013] Based on the image video, record the smoke distribution and diffusion;

[0014] Based on the recorded data for analysis, the analysis results are abnormal, and the warning module sends different alarm signals according to different levels of abnormalities, and uploads related data.

[0015] Preferably, the infrared matrix data is constituted, specifically:

[0016] The infrared thermal imaging video is decomposed into continuous multiple thermal images;

[0017] Each thermal image is divided into a grid structure, and each cell of the grid structure is mapped to a region in the corresponding thermal image.

[0018] The pixel value of each cell of the thermal image is obtained, and the pixel value is converted into a temperature value.

[0019] According to the temperature value of each cell of the thermal image, a two-dimensional array is constructed, which is the infrared matrix.

[0020] Preferably, each cell of the grid structure is mapped to a region in the corresponding thermal image, specifically:

[0021] Determine the space ratio:

[0022] According to the distance from the infrared thermal imaging video acquisition module to the coal conveying corridor, the area corresponding to each pixel point in the actual environment is calculated.

[0023] Establish a coordinate system:

[0024] A coordinate system is established in the actual environment, taking one end of the coal conveying corridor as the origin, and aligning the field of view of the infrared thermal imaging video acquisition module with the coordinate system.

[0025] Map the cell to the actual region:

[0026] For each cell in the infrared matrix, according to its position in the image and the previously determined space ratio, the position and area of the cell in the actual coordinate system are calculated.

[0027] Preferably, the image video records the smoke distribution and diffusion, specifically:

[0028] Through color space conversion, mark the gray pixel points as 1, and the rest of the pixel points as 0.

[0029] According to the image video, calculate the total amount of pixel points 1 and its growth rate per second.

[0030] Preferably, the method further comprises the following steps:

[0031] For different alarm signals, after manual processing, update the analysis strategy according to the processing result.

[0032] The application discloses a fire early warning system in a coal conveying gallery, and relates to a fire early warning method in the coal conveying gallery.

[0033] An infrared thermal imaging video acquisition module is arranged, which is used for acquiring infrared thermal imaging video of objects in the coal conveying gallery and sending the infrared thermal imaging video data to the data processing module.

[0034] An image video acquisition module is arranged, which is used for acquiring image video in the coal conveying gallery and sending the video to the data processing module.

[0035] A data processing module is arranged, which is used for converting the received infrared thermal imaging video into infrared matrix data, wherein the infrared matrix data is a two-dimensional array containing temperature values of different squares of a single-frame thermal image of the infrared thermal imaging video.

[0036] An interference term exclusion module is arranged, which is used for excluding the highest and lowest temperatures of interference terms in the coal conveying gallery from the infrared matrix data.

[0037] The data processing module is internally provided with a temperature processing unit and a smoke processing unit.

[0038] The temperature processing unit is used for determining the square point of the highest temperature in the infrared matrix data after excluding the highest and lowest temperatures of the interference terms, locating the corresponding area in the infrared thermal imaging video, and calculating and recording the temperature rising speed and the maximum temperature of the area within a preset time.

[0039] The temperature processing unit is also used for recording the temperature rising conditions of the surrounding square points of the square point of the highest temperature and performing weight division based on the temperature rising conditions, wherein the temperature rising conditions include the conditions of the rising speed being less than a first preset speed and the rising speed being greater than a second preset speed, the first preset speed is less than the second preset speed, the weight of the slow rising speed is low, and the weight of the fast rising speed is high.

[0040] The temperature processing unit is also used for recording the temperature diffusion conditions of the surrounding square points of the square point of the highest temperature.

[0041] The smoke processing unit is used for calculating and recording the smoke distribution and diffusion conditions according to the image video.

[0042] The data processing module is used for sending the recorded data to an analysis module, the analysis module uses different models to analyze the respective data, and the analysis module sends the analysis results to a warning module after the analysis, and the warning module judges the warning level according to the analysis results.

[0043] Preferably, the analysis module is built-in with different models, the different models including interference item models of each component in the coal conveying gallery, and the recorded data is judged according to each interference model whether the temperature belongs to the range of itself, if not, an abnormal temperature is obtained;

[0044] The analysis result includes that the highest temperature exceeds a first preset threshold value, the temperature increasing rate exceeds a second preset threshold value, and the smoke area exceeds a third preset threshold value.

[0045] Preferably, the early warning module is built-in with a self-learning model, which is used to import the processed record value into the corresponding interference item model after the early warning module issues an alarm and after processing, so as to perform iterative updating.

[0046] Preferably, the method further comprises a storage module, which is used to store the infrared thermal imaging video and the image video.

[0047] Preferably, when the cognitive agent is applied in the coal conveying gallery, the included angle between the cognitive agent and the top of the coal conveying gallery is 30°.

[0048] The present application has the beneficial technical effects that: the present application obtains the temperature data in the coal conveying gallery in a matrix format through infrared thermal imaging, establishes a temperature matrix, and the matrix data after excluding the highest temperature Tmax and the lowest temperature Tmin can more accurately reflect the actual temperature distribution in the coal conveying gallery;

[0049] Meanwhile, the present application can obtain comprehensive information of temperature change in the region, including temperature rising speed and maximum temperature, by finding the highest temperature square point in the temperature matrix and comparing with the corresponding region in the infrared video, which are the key indicators for evaluating fire risk;

[0050] The present application further records the temperature rising and diffusion of the square points around the highest temperature square point in the temperature matrix, so as to predict the fire development and issue fire warning early. DETAILED DESCRIPTION

[0051] In order to make the skilled in the art more clear and definite the technical solutions of the present application, the present application is further described in detail below in combination with embodiments, but the embodiments of the present application are not limited thereto.

[0052] The fire warning method in the coal conveying gallery provided by the present embodiment comprises the following steps:

[0053] 1) A cognitive agent is installed on the top of the coal conveying gallery, and the included angle between the collection direction of the collection end of the cognitive agent and the top of the coal conveying gallery is 30°, so as to improve the accuracy of fire source detection to a certain extent, even in the case that the fire source is blocked by the coal conveying equipment or coal;

[0054] 2) The cognitive agent continuously collects infrared thermal imaging videos and image videos, which are stored and used for subsequent model calculations;

[0055] 3) The infrared thermal imaging videos are structured and converted into matrix-arranged data to form infrared matrix data. The infrared matrix is a two-dimensional array containing the temperature values of each grid point, and each grid point corresponds to a region in the actual environment;

[0056] The specific steps for constructing the infrared matrix data are as follows:

[0057] The infrared thermal imaging videos are decomposed into independent thermal image frames to obtain multiple consecutive thermal images. In this process, each thermal image frame can be regarded as an independent static image, and each image contains a large number of pixel points;

[0058] Each thermal image frame is divided into a grid structure, which is a 9x9 grid, i.e., each thermal image frame is converted into 81 independent regions, and each grid maps to a region in the corresponding image. After segmentation, each thermal image frame is converted into a set of grids. This step is mainly to reduce the complexity of the data for subsequent processing;

[0059] The pixel values of each grid of the thermal image are obtained, and these values are converted to temperatures to obtain the temperature values corresponding to each grid. The mapping relationship between pixel values and temperature can be obtained in advance by calibrating the pixel values and the corresponding temperature values. In this step, since each grid contains multiple pixels, the values of these pixels need to be processed, such as taking the average, and then mapping the average temperature value to obtain the temperature value of the grid;

[0060] According to the temperature values of each grid, a two-dimensional array is constructed, which is the infrared matrix. In this matrix, each position (element) corresponds to a grid in the original thermal image, and the value of the element is the temperature of the grid. In this way, through the matrix, the temperature distribution of the original thermal image can be clearly seen;

[0061] 4) From the infrared matrix data, the highest temperature and the lowest temperature of the interference items in the coal conveying gallery, such as belts, bearings, water vapor, and coal heat, are excluded to avoid misleading in subsequent processing; max and the lowest T min

[0062] 5) From the infrared matrix data, the grid point with the highest temperature is found, and its real-time performance in the corresponding region of the infrared thermal imaging video is compared. The temperature rising speed and the maximum temperature of the grid point with the highest temperature within a predetermined time are recorded;

[0063] ​Finding the cell with the highest temperature in the matrix constructed from the infrared thermal imaging data can determine the area with the highest temperature in the actual environment;

[0064] Comparing the cell with the highest temperature found with the real-time performance of the same area in the infrared thermal imaging video, where the infrared thermal imaging video can provide dynamic information about temperature changes, while the infrared matrix data provides static temperature information. By comparing the two, more comprehensive information about temperature changes can be obtained;

[0065] Based on the comparison results, record the temperature rising speed and the maximum temperature of the cell with the highest temperature within a preset time. These two parameters are important indicators for evaluating fire risk. The temperature rising speed reflects the trend of temperature change. If the rising speed is too fast, it may indicate the occurrence of fire. The maximum temperature directly reflects the heat of the current environment;

[0066] 6) Continuously monitor the cell with the highest temperature in the infrared matrix and its neighborhood, and compare the corresponding area in the infrared thermal imaging video. Record the temperature rising of the surrounding cells of the cell with the highest temperature and weight the surrounding temperature. For example, in a 3x3 grid, position 5 has the highest temperature, positions 1-3 have low rising speed and low weight, and positions 7-9 have high rising speed and high weight. This facilitates the prediction of the rising direction of the flame;

[0067] In this step, the main target of comparing the cell in the infrared thermal imaging video is to observe and record the temperature changes of the surrounding cells of the cell with the highest temperature. This is because the fire source usually spreads to the surrounding, causing the temperature to rise;

[0068] The process of comparing the corresponding area in the infrared thermal imaging video not only obtains dynamic information (the temperature rising speed and pattern of the surrounding cells of the cell with the highest temperature), but also clearly monitors the process and pattern of fire spread. In addition, by comparing the cell in the infrared thermal imaging video, the fire risk of the surrounding cells can be more accurately evaluated, which is helpful for weight division. For example, if the temperature rising speed of a cell is very fast, it needs to be given a higher weight, so as to judge the trend of fire spread direction;

[0069] 7) Find the cell with the highest temperature from the infrared matrix, compare the cell in the infrared thermal imaging video, and record the spread of the surrounding cells;

[0070] The cell with the highest temperature in the infrared matrix usually represents the location of the (future) fire source. Then, record the temperature changes of the surrounding cells of this cell with the highest temperature. This is because the fire source usually spreads to the surrounding, which is manifested as the gradual rise of the temperature of the surrounding cells;

[0071] The rising rate of the surrounding squares can be calculated by comparing the infrared thermal imaging video images frame by frame;

[0072] According to the records of the diffusion, the development of the fire can be more accurately predicted before the fire occurs, the fire warning can be given earlier, and the emergency response personnel can be provided with sufficient reaction time, so that the occurrence of the fire can be more effectively prevented, and the possible loss can be reduced;

[0073] 8) Smoke detection is performed through image video, and smoke distribution and diffusion are recorded;

[0074] The process mainly obtains the distribution of smoke in space and the change trend of the smoke over time through continuous analysis of video frames of image video;

[0075] The smoke distribution refers to the area covered by the smoke, the size of the area, and the position of the area in each image frame;

[0076] The smoke diffusion refers to the change of the smoke distribution over time, that is, the moving path of the smoke and the expansion or contraction of the covered area;

[0077] 9) For the recorded data, different models are used to analyze the respective data to determine the abnormal temperature in the coal conveying gallery, and different alarm signals are sent according to different levels of abnormalities and relevant data is uploaded if an abnormality is detected;

[0078] The different models include a coal temperature rising model, a coal belt temperature rising model, and other interference term models, which are used to determine whether the recorded data belongs to the temperature within the scope of the model, and if not, an abnormal temperature is obtained;

[0079] The relevant data includes positioning (gallery position of the monitoring device), location (monitoring picture area), abnormal temperature, and structured abnormal position (monitoring picture position) temperature matrix;

[0080] 10) When the alarm is manually processed, the processed record value is imported into the model, that is, into the model of the specified device (such as the coal belt temperature rising model), and the model is iteratively updated. Iteration is for relearning. For example, if the coal belt abnormal temperature is misjudged, the model is iterated, and the boundary value of the model is expanded. If it is really on fire, the model is iterated to improve the accuracy.

[0081] In this embodiment, there are 30 frames per second, and the first frame of the first second is compared with the first frame of the second second, the second frame of the first second is compared with the second frame of the second second, and so on, to calculate the temperature rising speed of the frame number interval of one second;

[0082] For the square with the highest temperature, the temperature rising speed in the preset time is calculated;

[0083] For the surrounding squares of the square with the highest temperature, calculate the temperature increase rate thereof within a preset time;

[0084] For the surrounding squares of the square with the highest temperature, calculate the diffusion speed thereof, the diffusion speed referring to the number of squares diffused per frame;

[0085] For the diffused squares, a weight grading can also be performed, with 90% of the highest temperature within a preset time being the highest weight, 70% being the medium weight, and 40% being the low weight. For example, if the highest temperature is 100 degrees, then 90-100 degrees is the highest weight (more than 90% of the highest temperature), and this area is the most dangerous area.

[0086] In the present embodiment, for step 9), the following is specifically performed:

[0087] 1) Collect the highest temperature of the square, the temperature increase rate of the square with the highest temperature, the diffusion of the surrounding of the square with the highest temperature, the temperature increase rate of the surrounding of the square with the highest temperature, the smoke coverage position, the smoke area, and the smoke increase rate data;

[0088] The temperature increase rate of the square with the highest temperature = (the highest temperature - the highest temperature at the previous time) / the time interval;

[0089] The temperature increase rate of the surrounding of the square with the highest temperature = (the average temperature of the surrounding of the square with the highest temperature - the average temperature of the surrounding of the square with the highest temperature at the previous time) / the time interval;

[0090] The smoke increase rate = (the current smoke area - the smoke area at the previous time) / the time interval;

[0091] If the highest temperature exceeds a first preset threshold value, a high-level warning level can be set, and the threshold value is 1.5 times the normal operating temperature of the coal conveying gallery;

[0092] If the temperature increase rate exceeds a second preset threshold value, a medium-level warning level can be set, and the threshold value is 3 times the normal temperature change speed;

[0093] If the smoke area exceeds a third preset threshold value, a medium-level or high-level warning level can be set, and the threshold value is 10% of the total area of the video frame;

[0094] If the smoke increase rate exceeds a fourth preset threshold value, a medium-level or high-level warning level can be set, and the threshold value is 3 times the normal smoke area change speed.

[0095] In the present embodiment, the length of the coal conveying gallery is 60 meters, and the width is 10 meters, and each square point corresponds to an area in the actual environment, specifically as follows:

[0096] Step 3.1 determines the space ratio:

[0097] The resolution of the infrared thermal imaging camera is determined to be 1920x1080 pixels, and the distance from the camera to the coal conveying gallery is such that each pixel corresponds to an area of 1 cm x 1 cm in the actual environment;

[0098] Step 3.2 Establish the coordinate system:

[0099] A Cartesian coordinate system is established in the coal conveying gallery, with one end of the coal conveying gallery as the origin, and the coordinate system aligned with the length and width of the coal conveying gallery;

[0100] Step 3.3 Map the grid to the actual area:

[0101] Step 3.3.1 Calculate the projection length

[0102] Since the camera and the top of the coal conveying gallery form a 30-degree angle, the projection length of the camera's field of view on the surface of the coal conveying gallery needs to be calculated:

[0103] Let the distance from the camera to the top of the coal conveying gallery be h meters, then the projection length L is calculated by the following formula:

[0104]

[0105] Step 3.3.2 Calculate the actual size of a single grid

[0106] Let the infrared matrix be a 9x9 grid, then the actual length l occupied by each grid in the projection length is calculated by the following formula:

[0107]

[0108] Step 3.3.3 Determine the actual position of the grid

[0109] For each grid in the infrared matrix, its position in the actual coordinate system is calculated by the following formula:

[0110] Let the coordinates of the grid be (i, j) in the infrared matrix, then its position (X, Y) in the actual coordinate system is calculated by the following formula:

[0111] X = i x l

[0112] Y = j x l;

[0113] Step 3.3.4 Calculate the area of the actual area corresponding to the grid

[0114] The actual length l occupied by each grid in the projection length, the area A of the actual area corresponding to each grid is calculated by the following formula:

[0115] A = l x l;

[0116] This area A represents the actual size of each grid in the coal conveying corridor, which is very important for assessing the scale of the fire and allocating resources;

[0117] In this way, each grid point is associated with an actual area in the environment;

[0118] Step 3.3.4 stores mapping information

[0119] Create a data structure for each grid to store its position, size, and area information in the actual coordinate system, which will be used for subsequent fire detection and early warning;

[0120] Step 3.4 calibration and verification

[0121] Place an object with a known temperature in the coal conveying corridor at the coordinates (20 meters, 5 meters), observe the temperature value of the corresponding grid in the infrared matrix, and compare it with the actual temperature of the object. If there is a deviation, you can adjust the spatial scale or recalibrate the camera to correct it;

[0122] Through the above steps, the correspondence between each grid point in the infrared matrix and the actual area in the coal conveying corridor is established, and the tilt angle and spatial scale of the camera are considered. This mapping is crucial for the accuracy of the fire warning system, and by calculating the area of each grid, the scale of the fire can be more effectively assessed and resources can be allocated more efficiently;

[0123] In addition, the calculation of the area of the actual area corresponding to each grid in the above steps has the following technical effects:

[0124] In the case of a fire, by calculating the area of each grid and knowing which grids detect fire, the scale of the fire can be estimated;

[0125] In the case of emergency response, knowing the area affected by the fire can help optimize resource allocation, such as for larger fires, more firefighters and equipment may be needed.

[0126] In this embodiment, there is a specific mapping relationship between pixel value and temperature, which can be obtained in advance by calibrating pixel values and corresponding temperature values. The calibration example is as follows:

[0127] Let pixel value 100 correspond to temperature 20℃

[0128] Let pixel value 150 correspond to temperature 30℃

[0129] Let pixel value 200 correspond to temperature 40℃

[0130] Suppose there are four pixels in a grid, and their pixel values are 110, 120, 130, and 140, respectively. First, calculate the average of these pixel values: (110+120+130+140) / 4=125. Then, according to the mapping relationship, it can be estimated that the pixel value 125 corresponds to a temperature of about 25°C.

[0131] In this embodiment, a two-dimensional array is constructed according to the temperature value of each grid, specifically:

[0132] A. Initialize the two-dimensional array

[0133] Create a 9x9 two-dimensional array, which will be used to store the temperature value of each grid;

[0134] B. Traverse each grid

[0135] Suppose the grid being processed is in the 2nd row and 3rd column, and there are four pixels in this grid, with pixel values of 110, 120, 130, and 140, respectively;

[0136] For each grid in each frame of thermal image, perform the following steps:

[0137] Calculate the average pixel value:

[0138] Calculate the average of all pixel values in the grid: (110+120+130+140) / 4=125;

[0139] Map to temperature:

[0140] Using the mapping relationship between pixel value and temperature, convert the average pixel value to temperature. Suppose the mapping relationship is linear, that is, for every 10 pixel value increase, the temperature increases by 5°C. Then, the pixel value 125 corresponds to a temperature of 25°C;

[0141] Store the calculated temperature value 25°C in the [2][3] position of the two-dimensional array;

[0142] C. Repeat step B

[0143] For other grids in the thermal image, repeat step 2 until the temperature values of all grids are stored in the two-dimensional array;

[0144] D. After completing step C, each element in the two-dimensional array corresponds to a grid in the original thermal image, and the value of the element is the temperature of that grid;

[0145] Through the above process, a 9x9 two-dimensional array is obtained, which can clearly show the temperature distribution of the original thermal image.

[0146] In this embodiment, from the infrared matrix data, the interference items in the coal conveying corridor are excluded, specifically:

[0147] Step 4.1 Establish the temperature range database of interference items:

[0148] Obtain the temperature range of interference items such as belts, bearings, water vapor, and coal heat in the actual operating environment, for example, the temperature of the belt during normal operation is usually 30-50 degrees Celsius, the normal operating temperature of the bearing is 40-80 degrees Celsius, and the temperature of the water vapor may be around 100 degrees Celsius. Store these data in the database for subsequent comparison;

[0149] Step 4.2 Analyze the infrared matrix data:

[0150] Traverse each element in the infrared matrix to obtain its temperature value. Since the infrared matrix is composed of a two-dimensional array in step 3), each element represents the temperature value of a square, so the temperature value of each element in the two-dimensional array can be obtained by looping through it;

[0151] Step 4.3 Compare the temperature value with the interference item temperature range in the database:

[0152] For each element in the infrared matrix, compare its temperature value with the interference item temperature range established in step 4.1. If the temperature value falls within the temperature range of the interference item, mark the element as an interference item;

[0153] Step 4.4 Process the elements marked as interference items:

[0154] For elements marked as interference items, two processing methods can be adopted: one is to set their temperature value to a specific value, such as the minimum temperature Tmin, to indicate that this area is not a high-temperature area of fire; the second is to directly remove the element from the infrared matrix to reduce the complexity of subsequent analysis;

[0155] Step 4.5 Repeat steps 4.2 to 4.4 until all elements in the infrared matrix are processed:

[0156] Continue to traverse the next element in the infrared matrix and repeat steps 4.2 to 4.4 until all elements are analyzed and processed.

[0157] In this embodiment, the infrared thermal imaging video pays more attention to heat sources and temperature changes, which is of great value for detecting fire sources and predicting fire development. While ordinary image videos can capture color, shape, texture, and other features, which are more suitable for smoke detection, as the color, concentration, shape change, and other features of smoke can be obtained from image videos, while the infrared thermal imaging has relatively weak ability to capture these features.

[0158] In this embodiment, through step 8), the image video is analyzed to obtain the information of the smoke, and a reaction can be made in the initial stage of the fire, so as to reduce the possible loss.

[0159] In this embodiment, the "highest temperature square point in the infrared matrix" mentioned in steps 5-7 can be the same square point, or can change to another position over time, for example, the coordinate point (4, 5) is originally the highest temperature, but the material next to it is more easily heat-conducting, and (4, 6) can be the highest temperature.

[0160] A coal conveying gallery fire warning system, the warning system is applied to a coal conveying gallery fire warning method, the warning system comprises a cognitive agent, and the cognitive agent comprises:

[0161] An infrared thermal imaging video acquisition module, which is used to acquire infrared thermal imaging video of objects in the coal conveying gallery, and send the infrared thermal imaging video data to a data processing module;

[0162] An image video acquisition module, which is used to acquire image video in the coal conveying gallery, and send the video to the data processing module;

[0163] A data processing module, which is used to convert the received infrared thermal imaging video into infrared matrix data, the infrared matrix data being a two-dimensional array containing temperature values of different squares of a single-frame thermal image of the infrared thermal imaging video;

[0164] An interference term exclusion module, which is used to exclude the highest and lowest temperatures of interference terms in the infrared matrix data;

[0165] The data processing module is built-in with a temperature processing unit and a smoke processing unit;

[0166] The temperature processing unit is used to determine the square point of the highest temperature in the infrared matrix data after excluding the highest and lowest temperatures of the interference terms, locate the corresponding area in the infrared thermal imaging video, and calculate and record the temperature rising speed and the maximum temperature of the area within a preset time;

[0167] The temperature processing unit is also used to record the temperature rising conditions of the square points around the highest temperature square point and perform weight division based on the temperature rising conditions, wherein the temperature rising conditions include the conditions of the rising speed being less than a first preset speed and the rising speed being greater than a second preset speed, the first preset speed is less than the second preset speed, and the weight of the slow rising speed is low, and the weight of the fast rising speed is high;

[0168] The temperature processing unit is also used to record the temperature diffusion conditions of the square points around the highest temperature square point;

[0169] The smoke processing unit is used for calculating and recording smoke distribution and diffusion according to the image video;

[0170] The data processing module is used for sending the recorded data to the analysis module, the analysis module uses different models to analyze respective data, and the analysis module sends analysis results to the early warning module, and the early warning module judges the early warning level according to the analysis results.

[0171] In the embodiment, the analysis module is built-in different models, the different models include a coal temperature rising model, a coal belt temperature rising model and other interference term models, whether the recorded data belongs to the temperature within the range of each model is judged, if not, an abnormal temperature is obtained;

[0172] The analysis results include that the highest temperature exceeds a first preset threshold value, the temperature increasing speed exceeds a second preset threshold value, and the smoke area exceeds a third preset threshold value.

[0173] In the embodiment, the early warning module is built-in a self-learning model, the self-learning model is used for importing the processed record value into the corresponding interference term model after the early warning module issues an alarm and is processed, so as to perform iterative updating.

[0174] In the embodiment, a storage module is further included, the module is used for storing the infrared thermal imaging video and the image video.

[0175] The above is only further embodiments of the present application, but the protection scope of the present application is not limited to this, any skilled person in the art in the disclosed range of the present application, according to the technical solution and the concept of the present application, equivalent replacement or change, all belong to the protection scope of the present application.

Claims

1. A method for early warning of fires in a coal conveying corridor, characterized in that, Includes the following steps: Acquire infrared thermal imaging videos and images of the coal conveying corridor; The infrared thermal imaging video is structured and then converted into data arranged in a matrix to form an infrared matrix. Exclude the highest and lowest temperatures of the coal conveying corridor from the interference terms in the infrared matrix; Locate the grid point with the highest temperature in the infrared matrix, and calculate the temperature rise rate and maximum temperature of that grid point within a preset time. Record the rate of temperature increase of the surrounding grid points of the grid point with the highest temperature, and assign weights to the surrounding grid points based on this rate of temperature increase. Record the temperature diffusion of the grid points surrounding the grid point with the highest temperature; record the distribution and diffusion of smoke based on images and videos; Based on the analysis of the recorded data, if the analysis results are abnormal, the early warning module will issue different alarm signals according to the different levels of abnormality and upload relevant data. The infrared matrix data specifically comprises: Infrared thermal imaging video is decomposed into continuous multi-frame thermal images; Each frame of thermal image is divided into a grid structure, and each square of the grid structure is mapped to a region in the corresponding thermal image. Obtain the pixel value of each square in the thermal image and convert these pixel values ​​into temperature values; Based on the temperature value of each square in the thermal image, a two-dimensional array is constructed, which is the infrared matrix; Each square of the grid structure maps to a region in the corresponding thermal image, specifically: Determine the spatial proportions: Based on the distance from the infrared thermal imaging video acquisition module to the coal conveying corridor, the area corresponding to each pixel in the actual environment is calculated. Establish a coordinate system: Establish a coordinate system in the actual environment, with one end of the coal conveying corridor as the origin, and align the field of view of the infrared thermal imaging video acquisition module with this coordinate system; Mapping squares to actual areas: For each square in the infrared matrix, its position and area in the actual coordinate system are calculated based on its position in the image and the previously determined spatial proportions.

2. The fire early warning method for a coal conveying corridor according to claim 1, characterized in that, The process of recording the distribution and diffusion of smoke based on images and videos specifically includes: By converting the color space, gray pixels are marked as 1, and the remaining pixels are marked as 0; Calculate the total number of pixels and their growth rate based on 30 frames per second for images and videos.

3. A fire early warning method for a coal conveying corridor according to claim 1, characterized in that... It lies in, It also includes the following steps: For different alarm signals, after manual processing, the analysis strategy is updated based on the processing results.

4. A fire early warning system for a coal conveying corridor, wherein the early warning system is applied to a fire early warning method for a coal conveying corridor according to any one of claims 1 to 3, characterized in that, The early warning system includes a cognitive agent, which includes: Infrared thermal imaging video acquisition module: This module is used to acquire infrared thermal imaging videos of objects in the coal conveying corridor and send the infrared thermal imaging video data to the data processing module. Image and video acquisition module: This module is used to acquire images and videos inside the coal conveying corridor and send the videos to the data processing module; The data processing module is used to convert the received infrared thermal imaging video into infrared matrix data, which is a two-dimensional array containing the temperature values ​​of different squares in a single frame of the infrared thermal imaging video. Interference removal module: This module is used to remove the highest and lowest temperatures, which are interference items in the coal conveying corridor, from the infrared matrix data. The data processing module has a built-in temperature processing unit and a smoke processing unit; The temperature processing unit is used to determine the grid point with the highest temperature in the infrared matrix data after excluding the highest and lowest temperatures of the interference items, locate the corresponding area in the infrared thermal imaging video, and calculate and record the temperature rise rate and maximum temperature of the area within a preset time. The temperature processing unit is also used to record the temperature rise of the grid points surrounding the highest temperature grid point and to perform weight division based on the temperature rise. The temperature rise includes cases where the rise rate is less than a first preset speed and cases where the rise rate is greater than a second preset speed. The first preset speed is less than the second preset speed. The weight of the slower rise rate is low, and the weight of the faster rise rate is high. The temperature processing unit is also used to record the temperature diffusion of the grid points surrounding the grid point with the highest temperature. The smoke processing unit is used to calculate and record the distribution and diffusion of smoke based on images and videos; The data processing module sends the recorded data to the analysis module. The analysis module uses different models to analyze the data and then sends the analysis results to the early warning module. The early warning module determines the early warning level based on the analysis results.

5. A fire early warning system for a coal conveying corridor according to claim 4, characterized in that, The analysis module has different built-in models, including interference models for each component in the coal conveying corridor. Based on each interference model, it determines whether the recorded data belongs to its own temperature range. If not, it determines the abnormal temperature. The analysis results include the highest temperature exceeding a first preset threshold, the temperature increase exceeding a second preset threshold, and the smoke area exceeding a third preset threshold.

6. A fire early warning system for a coal conveying corridor according to claim 4, characterized in that, The early warning module has a built-in self-learning model. After the early warning module issues an alarm and processes it, the processed recorded values ​​are imported into the corresponding interference item model for iterative updates.

7. A fire early warning system for a coal conveying corridor according to claim 4, characterized in that, It also includes a storage module for storing the infrared thermal imaging video and the image video.

8. A fire early warning system for a coal conveying corridor according to claim 4, characterized in that, When the cognitive agent is applied in the coal conveying corridor, the angle between the cognitive agent and the top of the coal conveying corridor is 30°.

Images