An industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging

Abstract

The present invention discloses an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging, comprising: a pipeline blockage area identification module for identifying the pipeline blockage area; a thermal imaging feature index acquisition module for acquiring the thermal imaging feature indexes of the pipeline blockage area; a blockage degree evaluation module for constructing an evaluation model of the pipeline blockage degree to evaluate the blockage degree; a pipeline blockage prediction module for predicting the time point when the blockage degree is higher than a preset degree threshold; a pipeline blockage area migration prediction module for predicting the blockage degree when the position of the pipeline changes; a pipeline blockage area aggregation prediction module for predicting the aggregation position and aggregation time of the pipeline blockage area; and a preventive measure optimization module for continuously optimizing the preventive measures against blockage risks and the preventive measures against blockage migration risks. The present invention provides a comprehensive and efficient monitoring and management solution for industrial transmission pipeline blockages, improving the safety and maintenance efficiency of industrial transmission pipeline operations, and reducing the occurrence of blockage risks and pipeline failures.

Description

Technical Field

[0001] The present invention relates to the field of industrial equipment monitoring, and particularly to an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging. Background Art

[0002] Industrial equipment fault monitoring is one of the important links to ensure the normal operation of industrial production systems. Especially in industries such as petroleum, natural gas, and chemical industries, equipment failures not only lead to production stagnation but may also bring serious safety hazards and economic losses. The manifestations of industrial equipment failures are diverse, including problems such as wear, overheating, and aging of mechanical components, as well as blockages and abnormal fluid flow in transmission pipelines. Among them, the blockage fault of industrial transmission pipelines is one of the most common, hidden, and far-reaching problems in industrial production. At present, traditional pipeline monitoring technologies such as pressure sensors and flow meters can detect the blockage situation inside the pipeline indirectly, but usually only have obvious monitoring signals when the blockage problem is relatively serious and affects the overall operation of the pipeline. This lagging detection method often makes it impossible to detect and handle problems in the early stage in a timely manner, resulting in the further deterioration of the blockage. In addition, traditional monitoring means also have limitations in the spatial positioning of pipeline blockages. The complex structure of the pipeline, such as elbows, valves, branch points, and height change points, will affect the position and distribution of the blockage, and traditional flow and pressure monitoring equipment often has difficulty in determining the specific blockage position. Especially in complex pipeline systems, the accurate positioning of the blockage area is crucial for effectively dealing with and preventing blockages, but traditional technologies are difficult to achieve such fine spatial recognition. In addition, the pipeline blockage problem is dynamic, and the blockage may migrate in the pipeline with the flow of the fluid, resulting in changes in the blockage area. Traditional fixed-point monitoring means are difficult to track the migration of the blockage in real time, resulting in the gradual aggregation of some small-scale blockage areas after migration, forming larger blockage areas, and ultimately causing serious pipeline blockages or system failures. Therefore, the lack of real-time monitoring of the dynamic migration of blockages is another major defect of traditional detection technologies. In addition, as the scale of industrial pipeline systems continues to expand, a single detection method is difficult to meet the comprehensive, real-time, and efficient monitoring requirements. When multiple small-scale blockage areas gradually migrate and aggregate, the lack of comprehensive monitoring means leads to the lag of risk prevention, increasing the maintenance cost and failure rate of the pipeline system. Therefore, there is an urgent need for a system that can accurately identify the blockage area, evaluate the blockage degree, predict the migration trend of the blockage, and take corresponding preventive measures in a timely manner. Summary of the Invention

[0003] In view of the problems existing in the above-mentioned prior art, the present invention provides an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging, which mainly includes:

[0004] The pipeline blockage area identification module is used to obtain the thermal imaging image of the industrial transmission pipeline, identify the temperature abnormal area where the pipeline is blocked, and mark it as the pipeline blockage area;

[0005] The thermal imaging feature index acquisition module is used to extract the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction of the thermal imaging image of the pipeline blockage area, and form the thermal imaging feature index of the pipeline blockage area;

[0006] The blockage degree evaluation module is used to construct a pipeline blockage degree evaluation model according to the thermal imaging feature indexes of the historical blockage areas of the industrial transmission pipeline, evaluate the blockage degree of the pipeline blockage area of the industrial transmission pipeline, and formulate blockage risk prevention measures for the pipeline blockage areas where the blockage degree is higher than the preset degree threshold;

[0007] The pipeline blockage prediction module is used to determine the heat diffusion speed of the pipeline blockage area at different time periods according to the temperature values of the pipeline blockage area at different times, construct a thermal imaging feature index prediction model, predict the time point when the blockage degree of the pipeline blockage area where the heat diffusion speed is greater than the preset speed threshold is higher than the preset degree threshold, and implement blockage risk prevention measures in advance;

[0008] The pipeline blockage area migration prediction module is used to judge whether the pipeline blockage area migrates according to the blockage position of the pipeline blockage area, and based on the migration data of the pipeline blockage area, predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes, and implement blockage migration risk prevention measures for the pipeline blockage areas where the blockage degree is higher than the preset degree threshold;

[0009] The pipeline blockage area aggregation prediction module is used to predict the aggregation position and aggregation time of the pipeline blockage area according to the migration speed and migration direction of each pipeline blockage area, and implement blockage migration risk prevention measures for the aggregation area of the pipeline blockage area in advance;

[0010] The prevention measure optimization module is used to continuously monitor the thermal imaging image of the industrial transmission pipeline, evaluate the implementation effects of the blockage risk prevention measures and the blockage migration risk prevention measures, and continuously optimize the prevention measures.

[0011] Furthermore, the pipeline blockage area identification module, which is used to obtain the thermal imaging image of the industrial transmission pipeline, identify the temperature abnormal area where the pipeline is blocked, and mark it as the pipeline blockage area, includes:

[0012] Perform a comprehensive scan of the industrial transmission pipeline according to a predetermined scanning path and speed, obtain and save the thermal imaging image of the industrial transmission pipeline; convert the thermal imaging image of the industrial transmission pipeline into a grayscale image, set a temperature threshold as the segmentation criterion, perform binary processing on the grayscale image, and mark the image area with a temperature higher than the set temperature threshold as the temperature anomaly area; through the historical monitoring database of the industrial transmission pipeline, obtain the thermal imaging images of the historical temperature anomaly areas of the industrial transmission pipeline, mark whether pipeline blockage has occurred, and use a convolutional neural network for model training to construct a pipeline blockage recognition model; according to the newly obtained thermal imaging image of the temperature anomaly area, use the pipeline blockage recognition model to identify the temperature anomaly area where pipeline blockage has occurred and mark it as the pipeline blockage area.

[0013] Further, the thermal imaging feature index acquisition module is used to extract the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction of the thermal imaging image of the pipeline blockage area, and form the thermal imaging feature index of the pipeline blockage area, including:

[0014] According to the thermal imaging image of the pipeline blockage area, extract the temperature values of each pixel point in the thermal imaging image to obtain the temperature distribution data of each pixel point; according to the temperature values in the thermal imaging image, use the central difference method to calculate the temperature change rate between adjacent pixel points to determine the average temperature gradient value of the pipeline blockage area; according to the temperature values in the thermal imaging image, calculate the difference between the highest temperature and the lowest temperature to obtain the maximum temperature difference of the pipeline blockage area; use the Sobel operator to calculate the gradient direction of each pixel point to determine the standard deviation of the gradient direction; perform normalization processing on the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction, and combine them into a feature vector to form the thermal imaging feature index of the pipeline blockage area.

[0015] Further, the blockage degree evaluation module is used to construct a pipeline blockage degree evaluation model according to the thermal imaging feature index of the historical blockage area of the industrial transmission pipeline, evaluate the blockage degree of the pipeline blockage area of the industrial transmission pipeline, and formulate blockage risk prevention measures for the pipeline blockage area with a blockage degree higher than the preset degree threshold, including:

[0016] Obtain the thermal imaging characteristic indicators, blockage degree, pipe diameter, material, and location of the historical blocked areas of industrial transmission pipelines from the historical monitoring database of industrial transmission pipelines. Use a recurrent neural network for model training to construct a pipeline blockage degree evaluation model. The materials of the pipeline include steel, stainless steel, and PVC, and the locations of the pipeline include main transmission pipelines, pipe elbows, pipe valves, pipe branch points, pipe connection points, and height change points; According to the thermal imaging characteristic indicators of the pipeline blocked area obtained in real time, combined with the pipe diameter, material, and location of the pipeline, use the pipeline blockage degree evaluation model to evaluate the blockage degree of the pipeline blocked area of the current industrial transmission pipeline; Develop and implement blockage risk prevention measures for pipeline blocked areas with a blockage degree higher than the preset degree threshold, including cleaning debris inside the pipeline, increasing the monitoring frequency, performing thermal imaging detection regularly, and optimizing the transmission pressure or flow rate of the pipeline.

[0017] Further, the pipeline blockage prediction module is used to determine the heat diffusion speed of different time periods in the pipeline blocked area according to the temperature values at different times in the pipeline blocked area, and construct a thermal imaging characteristic indicator prediction model to predict the time point when the blockage degree of the pipeline blocked area with a heat diffusion speed greater than the preset speed threshold is higher than the preset degree threshold, and implement blockage risk prevention measures in advance, including:

[0018] Adopt a regular sampling method to obtain thermal imaging images of pipeline blocked areas with a blockage degree lower than the preset degree threshold at a preset time interval, extract the temperature distribution data at each moment, and determine the temperature value of the pipeline blocked area at each moment; According to the temperature values at different times in the pipeline blocked area, use the heat diffusion speed calculation formula to calculate the heat diffusion speed of different time periods in the pipeline blocked area, where T i and T j are the temperature values at time t i and time t j respectively, and α is the heat diffusion coefficient determined based on the pipeline material properties; If the heat diffusion speed of a certain time period in the pipeline blocked area is greater than the preset speed threshold, obtain the historical thermal imaging characteristic indicators of the pipeline blocked area through the historical monitoring database of industrial transmission pipelines and organize them into time series data; According to the time series data of the historical thermal imaging characteristic indicators, use the ARIMA algorithm for model training to construct a thermal imaging characteristic indicator prediction model, predict the thermal imaging characteristic indicators within a preset future time period, and combine the pipeline blockage degree evaluation model to judge the time point when the blockage degree of the pipeline blocked area is higher than the preset degree threshold, and implement blockage risk prevention measures for the pipeline blocked area in advance.

[0019] Further, the pipeline blockage area migration prediction module is used to determine whether the pipeline blockage area has migrated based on the blockage position of the pipeline blockage area, and predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes based on the migration data of the pipeline blockage area. Implement blockage migration risk prevention measures for pipeline blockage areas with a blockage degree higher than the preset degree threshold, including:

[0020] Determine the blockage position of the pipeline blockage area according to the position with the highest temperature value in the thermal imaging image of the pipeline blockage area; obtain the historical blockage position data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, and combine the current blockage position of the pipeline blockage area to judge whether the pipeline blockage area has migrated; if the pipeline blockage area has migrated, obtain the migration data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, including the migration distance, migration speed and migration direction, and use the ARIMA algorithm to predict the position of the pipeline where the pipeline blockage area is located in the future time period; if it is predicted that the position of the pipeline where the pipeline blockage area is located changes in the future time period, then predict the thermal imaging characteristic index when the position of the pipeline where the pipeline blockage area is located changes according to the thermal imaging characteristic index prediction model; according to the blockage migration prediction formula Predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes. Among them, S(t) is the blockage degree at time t, S1 is the blockage degree before migration, D1 and D2 are the diameters of the pipeline before and after migration respectively, k is the coefficient of the influence of the pipe diameter change on the blockage degree, obtained by fitting historical data, M1 and M2 are the material parameters of the pipeline before and after migration respectively, defined according to the friction coefficient and corrosion resistance characteristics of the material, l is the influence coefficient of the material change on the blockage degree, obtained by fitting historical data, F z Is an influence factor related to the pipeline position z, obtained by fitting historical data, m represents the number of pipeline positions passed during the migration process; implement blockage migration risk prevention measures for pipeline blockage areas with a blockage degree higher than the preset degree threshold, including optimizing the interval time of fluid transportation, using a pipeline regulating device to guide the blockage to migrate in a specified direction, and optimizing the flow rate by adjusting the power of the pump or the opening and closing degree of the valve.

[0021] Further, the pipeline blockage area aggregation prediction module is used to predict the aggregation position and aggregation time of the pipeline blockage area according to the migration speed and migration direction of each pipeline blockage area, and implement blockage migration risk prevention measures for the aggregation area of the pipeline blockage area in advance, including:

[0022] If there are migrations of several pipeline blockage areas in an industrial transmission pipeline with blockage degrees lower than a preset degree threshold, then according to the blockage migration prediction formula, predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes; if the predicted blockage degree when the position of the pipeline where the pipeline blockage area is located changes is lower than the preset degree threshold, obtain the blockage position data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline; according to the blockage position data of the pipeline blockage area, use the ARIMA algorithm to predict the aggregation position and aggregation time of the pipeline blockage area; based on the predicted aggregation position and aggregation time of the pipeline blockage area, implement blockage migration risk prevention measures in advance for the aggregation area of the pipeline blockage area.

[0023] Further, the prevention measure optimization module is used to continuously monitor the thermal imaging images of the industrial transmission pipeline, evaluate the implementation effects of the blockage risk prevention measures and the blockage migration risk prevention measures, and continuously optimize the prevention measures, including:

[0024] Continuously monitor the industrial transmission pipeline through infrared thermal imaging technology, and regularly obtain the thermal imaging images and corresponding temperature data of the industrial transmission pipeline; by comparing the thermal imaging images of the pipeline before and after implementing the blockage risk prevention measures and the blockage migration risk prevention measures, evaluate the effectiveness of the prevention measures; if the blockage degree is reduced and the temperature distribution of the blockage area returns to normal, it is determined that the blockage risk prevention measures and the blockage migration risk prevention measures are effective; if the blockage degree is not reduced, optimize the blockage risk prevention measures and the blockage migration risk prevention measures, including increasing the pipeline cleaning frequency, adjusting the transmission pressure or changing the flow rate, or changing the cleaning method for the blockage area; implement the optimized blockage risk prevention measures and the blockage migration risk prevention measures, evaluate the effectiveness of the new prevention measures, and continuously optimize the blockage risk prevention measures and the blockage migration risk prevention measures based on the effectiveness evaluation results of the prevention measures.

[0025] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0026] The present invention provides an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging. By applying infrared thermal imaging technology, the present invention can monitor the temperature distribution of industrial transmission pipelines in real time, accurately identify and mark the temperature abnormal areas as pipeline blockage areas. According to the extracted temperature characteristic indexes, the blockage degree of the pipeline is further evaluated to ensure that the blockage problem can be detected and processed in the early stage. The present invention can dynamically monitor the migration of blockage substances, predict the position change of the blockage area in the pipeline and the change trend of the blockage degree, and ensure that preventive measures are taken for the areas with higher blockage degree after migration to prevent the blockage from intensifying. For the situation where multiple small-scale blockage substances gradually gather, the present invention can predict their gathering position and time to ensure intervention before the blockage substances form a gathering and avoid greater risks caused by the gathering of blockage substances. The present invention provides a comprehensive and efficient pipeline blockage monitoring and management solution, improving the safety and maintenance efficiency of pipeline operation and reducing the blockage risk and the occurrence of pipeline failures. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flowchart of an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging according to the present invention;

[0028] Figure 2 is a schematic diagram of an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging according to the present invention;

[0029] Figure 3 is another schematic diagram of an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] As Figures 1 - 3 , an industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging in this embodiment may specifically include:

[0032] Step S101, a pipeline blockage area identification module, configured to obtain a thermal imaging image of an industrial transmission pipeline, identify a temperature abnormal area where pipeline blockage occurs, and mark it as a pipeline blockage area.

[0033] Perform a comprehensive scan of the industrial transmission pipeline according to a predetermined scan path and speed, obtain and save the thermal imaging image of the industrial transmission pipeline. Convert the thermal imaging image of the industrial transmission pipeline into a grayscale image, set a temperature threshold as the segmentation criterion, perform binary processing on the grayscale image, and mark the image area with a temperature higher than the set temperature threshold as the temperature anomaly area. Through the historical monitoring database of the industrial transmission pipeline, obtain the thermal imaging images of the historical temperature anomaly areas of the industrial transmission pipeline, and mark whether pipeline blockage has occurred, and use a convolutional neural network for model training to construct a pipeline blockage recognition model. According to the newly obtained thermal imaging image of the temperature anomaly area, use the pipeline blockage recognition model to identify the temperature anomaly area where pipeline blockage has occurred and mark it as the pipeline blockage area.

[0034] Exemplarily, during the monitoring of the industrial transmission pipeline, perform a comprehensive scan of the pipeline at a speed of 1 meter per second according to a predetermined scan path, and obtain and save the thermal imaging images of a pipeline with a total length of 500 meters. After converting the obtained thermal imaging image into a grayscale image, set the temperature threshold to 70 degrees Celsius, perform binary processing on the areas with a temperature higher than 70 degrees Celsius, and mark these areas as the temperature anomaly areas. Subsequently, through the historical monitoring database of the industrial transmission pipeline, obtain the thermal imaging images of the corresponding pipeline in the past three years, and mark which temperature anomaly areas have caused pipeline blockage. Use these historical data with annotations to train a convolutional neural network to construct a pipeline blockage recognition model. When an area with a temperature higher than 70 degrees Celsius in the newly obtained thermal imaging image of the pipeline is detected, use this model for analysis, and successfully identify one of the temperature anomaly areas, which is located 150 meters from the starting point of the pipeline, and mark it as the pipeline blockage area.

[0035] Step S102, the thermal imaging feature index acquisition module is used to extract the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction of the thermal imaging image of the pipeline blockage area, and form the thermal imaging feature index of the pipeline blockage area.

[0036] According to the thermal imaging image of the pipeline blockage area, extract the temperature values of each pixel point in the thermal imaging image to obtain the temperature distribution data of each pixel point. According to the temperature values in the thermal imaging image, use the central difference method to calculate the temperature change rate between adjacent pixel points to determine the average temperature gradient value of the pipeline blockage area. According to the temperature values in the thermal imaging image, calculate the difference between the highest temperature and the lowest temperature to obtain the maximum temperature difference of the pipeline blockage area. Use the Sobel operator to calculate the gradient direction of each pixel point to determine the standard deviation of the gradient direction. Standardize the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction, and combine them into a feature vector to form the thermal imaging feature index of the pipeline blockage area.

[0037] Exemplarily, during the process of monitoring an industrial transmission pipeline, a thermal imaging image of the pipeline blockage area was extracted, and the temperature value of each pixel point was extracted from it, generating a thermal imaging image with a resolution of 200x200 pixels. By analyzing the temperature distribution of each pixel point in this image, the temperature range of the entire area was obtained to be between 60 degrees Celsius and 120 degrees Celsius. Furthermore, the temperature change rate between adjacent pixel points was calculated through the central difference method, and the average temperature gradient of this pipeline blockage area was determined to be 10 degrees Celsius per pixel. The highest temperature in this area was 120 degrees Celsius, and the lowest temperature was 60 degrees Celsius, so the maximum temperature difference in this area was 60 degrees Celsius. The Sobel operator was used to calculate the gradient direction of each pixel point in the thermal imaging image, and the standard deviation of the gradient direction was determined to be 15 degrees, indicating that there was a certain degree of volatility in the change direction of different temperature gradients. The highest temperature of 120 degrees Celsius, the average temperature gradient of 10 degrees Celsius, the maximum temperature difference of 60 degrees Celsius, and the standard deviation of the gradient direction of 15 degrees were normalized and combined into a feature vector to describe the thermal imaging feature index of this pipeline blockage area.

[0038] Step S103, the blockage degree evaluation module is used to construct a pipeline blockage degree evaluation model based on the thermal imaging feature indexes of the historical blockage areas of the industrial transmission pipeline, evaluate the blockage degree of the pipeline blockage areas of the industrial transmission pipeline, and formulate blockage risk prevention measures for the pipeline blockage areas where the blockage degree is higher than the preset degree threshold.

[0039] Through the historical monitoring database of the industrial transmission pipeline, the thermal imaging feature indexes, blockage degree, pipe diameter, material, and location of the historical blockage areas of the industrial transmission pipeline were obtained. A pipeline blockage degree evaluation model was constructed using a recurrent neural network for model training. The materials of the pipeline include steel, stainless steel, and PVC, and the locations of the pipeline include the main transmission pipeline, pipeline elbows, pipeline valves, pipeline branch points, pipeline connection points, and height change points. According to the thermal imaging feature indexes of the pipeline blockage area obtained in real time, combined with the pipe diameter, material, and location of the pipeline, the pipeline blockage degree evaluation model was used to evaluate the blockage degree of the pipeline blockage area of the current industrial transmission pipeline. Blockage risk prevention measures were formulated and implemented for the pipeline blockage areas where the blockage degree is higher than the preset degree threshold, including cleaning the debris inside the pipeline, increasing the monitoring frequency, regularly performing thermal imaging detection, optimizing the transmission pressure or flow rate of the pipeline.

[0040] Exemplarily, during the monitoring of industrial transmission pipelines, thermal imaging characteristic indexes, blockage degrees, pipe diameters, materials, and location data of 20 different pipeline blockage areas in the past five years were obtained through the historical monitoring database. The pipe diameter ranges from 15 to 50 cm, the materials include steel, stainless steel, and PVC, etc., and the locations cover main transmission pipelines, pipe elbows, pipe valves, pipe branch points, pipe connection points, and height change points, etc. By inputting these data into a recurrent neural network, the training of a pipeline blockage degree evaluation model was completed, and a pipeline blockage degree evaluation model was constructed. During real-time monitoring, thermal imaging characteristic indexes of a main transmission pipeline were obtained, with the highest temperature being 85 °C, the average temperature gradient being 12 °C, the maximum temperature difference being 45 °C, and the standard deviation of the gradient direction being 18 degrees. Considering that the material of this pipeline is stainless steel, the pipe diameter is 30 cm, and its special location at the pipe elbow, these data were input into the trained pipeline blockage degree evaluation model, and it was evaluated that the blockage degree of the blocked area of this pipeline is 35%, higher than the preset degree threshold of 30%. It was decided to implement blockage risk prevention measures, including cleaning the debris inside the pipeline, increasing the monitoring frequency to once a week, regularly conducting thermal imaging inspections, and appropriately adjusting the transmission pressure of the pipeline to relieve the blockage situation.

[0041] Step S104, a pipeline blockage prediction module, is used to determine the heat diffusion speed of different time periods in the pipeline blockage area according to the temperature values at different times in the pipeline blockage area, and construct a thermal imaging characteristic index prediction model to predict the time point when the blockage degree of the pipeline blockage area with a heat diffusion speed greater than the preset speed threshold is higher than the preset degree threshold, and implement blockage risk prevention measures in advance.

[0042] Adopt a regular sampling method to obtain thermal imaging images of the pipeline blockage area with a blockage degree lower than the preset degree threshold at a preset time interval, extract the temperature distribution data at each moment, and determine the temperature value of the pipeline blockage area at each moment. According to the temperature values at different times in the pipeline blockage area, use the heat diffusion speed calculation formula Calculate the heat diffusion speed v of different time periods in the pipeline blockage area d , where T i and T j are respectively t i moment and t jThe temperature value at a moment, where α is the thermal diffusivity, which is determined based on the properties of the pipeline material. If the thermal diffusion rate in a certain period of the pipeline blockage area is greater than the preset speed threshold, the historical thermal imaging feature index of the pipeline blockage area is obtained from the historical monitoring database of the industrial transmission pipeline and organized into time series data. According to the time series data of the historical thermal imaging feature index, the ARIMA algorithm is used for model training to construct a thermal imaging feature index prediction model, predict the thermal imaging feature index in a preset future time period, and combine it with the pipeline blockage degree evaluation model to determine the time point when the blockage degree of the pipeline blockage area is higher than the preset degree threshold, and implement blockage risk prevention measures for the pipeline blockage area in advance.

[0043] Exemplarily, during the monitoring of the industrial transmission pipeline, a thermal imaging image of a pipeline blockage area with a blockage degree of 10% is obtained at a preset time interval of every half hour. At each sampling, the temperature distribution data of the pipeline blockage area is extracted, including, at time t i = 10:00, the temperature T i of the pipeline blockage area is 80 °C, and at time t j = 11:00, the temperature T j rises to 95 °C. Combining the stainless steel thermal diffusivity α = 0.02 of the pipeline material, the thermal diffusion speed calculation formula is used to calculate the thermal diffusion speed v d of 0.3 °C / minute. By comparing with the preset speed threshold of 0.2 °C / minute, it is determined that the thermal diffusion speed in this period is greater than the preset threshold. Therefore, the historical thermal imaging feature index of the pipeline blockage area in the past six months is extracted from the historical monitoring database of the industrial transmission pipeline and organized into time series data. By inputting these historical data into the ARIMA model for training, the thermal imaging feature index of the pipeline area in the next two hours is predicted, and combined with the pipeline blockage degree evaluation model, it is judged that the blockage degree of the pipeline may reach 30% severity in the next two hours. Therefore, preventive measures against blockage risks such as cleaning the pipeline, adjusting the transmission pressure, and increasing the monitoring frequency are taken in advance.

[0044] Step S105, the pipeline blockage area migration prediction module, is used to determine whether the pipeline blockage area has migrated according to the blockage position of the pipeline blockage area, and based on the migration data of the pipeline blockage area, predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes, and implement blockage migration risk prevention measures for the pipeline blockage area with a blockage degree higher than the preset degree threshold.

[0045] Determine the blockage location of the pipeline blockage area based on the position with the highest temperature value in the thermal imaging image of the pipeline blockage area. Obtain the historical blockage location data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, and combine it with the current blockage location of the pipeline blockage area to determine whether the pipeline blockage area has migrated. If the pipeline blockage area has migrated, obtain the migration data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, including the migration distance, migration speed, and migration direction, and use the ARIMA algorithm to predict the position of the pipeline where the pipeline blockage area is located in the future time period. If it is predicted that the position of the pipeline where the pipeline blockage area is located in the future time period has changed, then according to the thermal imaging feature index prediction model, predict the thermal imaging feature index when the position of the pipeline where the pipeline blockage area is located changes. According to the blockage migration prediction formula Predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes, where S(t) is the blockage degree at time t, S1 is the blockage degree before migration, D1 and D2 are the diameters of the pipeline before and after migration respectively, k is the coefficient of the influence of the pipe diameter change on the blockage degree, obtained by fitting historical data, M1 and M2 are the material parameters of the pipeline before and after migration respectively, defined according to the friction coefficient and corrosion resistance characteristics of the material, l is the influence coefficient of the material change on the blockage degree, obtained by fitting historical data, F z is the influence factor related to the pipeline position z, obtained by fitting historical data, and m represents the number of pipeline positions passed during the migration process. Implement blockage migration risk prevention measures for pipeline blockage areas with a blockage degree higher than the preset degree threshold, including optimizing the interval time of fluid transportation, using pipeline regulating devices to guide the blockage to migrate in a specified direction, and optimizing the flow rate by adjusting the power of the pump or the opening and closing degree of the valve.

[0046] Exemplarily, during the monitoring of an industrial transmission pipeline, the blockage location of a certain section of the pipeline is determined to be at 120 meters from the starting point of the main transmission pipeline through the position with the highest temperature value in the thermal imaging image. According to the historical monitoring database of the industrial transmission pipeline, it is found that the blocked area has migrated from the straight section of the main transmission pipeline 100 meters from the starting point to the pipeline elbow position 120 meters from the starting point. It is calculated that the migration distance of the blockage is 20 meters, the migration speed is 5 meters per hour, and the migration direction has changed along the pipeline transportation direction. Using the ARIMA algorithm to predict, within the next 2 hours, the blockage may continue to migrate to 140 meters and enter a pipeline branch point position. Since the position of the pipeline has migrated from the straight section of the main transmission pipeline to the pipeline elbow and may then migrate to the pipeline branch point, it is necessary to re-evaluate the blockage risk. After migration, the diameter of the pipeline has decreased from 30 cm to 20 cm at the pipeline elbow. At the same time, the pipeline material has also changed from stainless steel with a friction coefficient of 0.05 of the original material to PVC with a friction coefficient of 0.1. If the blockage degree S1 before migration is 60%, the influence coefficient k of the pipe diameter change on the blockage degree is 0.8, the influence coefficient l of the material change on the blockage degree is 0.5, and the blockage has passed through two key positions when migrating from the main transmission pipeline to the pipeline elbow, namely the pipeline valve with an influence factor F1 = 1.1 and the pipeline branch point with an influence factor F2 = 1.05. Using the blockage migration prediction formula Predict the blockage degree when the position of the pipeline where the blocked area is located changes. Among them, S(t) is the blockage degree at time t, S1 is the blockage degree before migration, D1 and D2 are the diameters of the pipeline before and after migration respectively, k is the coefficient of the influence of the pipe diameter change on the blockage degree, obtained by fitting historical data, M1 and M2 are the material parameters of the pipeline before and after migration respectively, defined according to the friction coefficient and corrosion resistance characteristics of the material, l is the influence coefficient of the material change on the blockage degree, obtained by fitting historical data, F z is the influence factor related to the pipeline position z, such as the main transmission pipeline, pipeline elbow, pipeline valve, pipeline branch point, pipeline connection point, and height change point, etc., obtained by fitting historical data. The influence factor F of the pipeline position Z will be adjusted as the pipeline position changes. m represents the number of pipeline positions passed through during the migration process. The predicted blockage degree when the position of the pipeline where the blocked area is located changes is 70.14%. This indicates that the blockage situation after migration is very serious. Therefore, it is necessary to immediately implement blockage risk prevention measures, including adjusting the power of the pump, optimizing the interval time of fluid transportation, and using pipeline regulating devices to guide the blockage to a safe position. At the same time, increase the monitoring frequency of the pipeline elbow and branch point to prevent the blockage from further intensifying.

[0047] Step S106: The pipeline blockage area aggregation prediction module is used to predict the aggregation position and aggregation time of the pipeline blockage areas based on the migration speed and migration direction of each pipeline blockage area, and implement blockage migration risk prevention measures in advance for the aggregation areas of the pipeline blockage areas.

[0048] If there are several pipeline blockage areas with blockage degrees lower than the preset degree threshold migrating in the industrial transmission pipeline, then according to the blockage migration prediction formula, predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes. If the predicted blockage degree when the position of the pipeline where the pipeline blockage area is located changes is lower than the preset degree threshold, obtain the blockage position data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline. According to the blockage position data of the pipeline blockage area, use the ARIMA algorithm to predict the aggregation position and aggregation time of the pipeline blockage area. Based on the predicted aggregation position and aggregation time of the pipeline blockage area, implement blockage migration risk prevention measures in advance for the aggregation area of the pipeline blockage area.

[0049] Exemplarily, during the monitoring of the industrial transmission pipeline, it is found that the blockage degrees of multiple blockage areas in the pipeline are all lower than the preset 30% threshold, and these blockage areas are located at positions 150 meters, 200 meters, and 250 meters away from the starting point of the pipeline respectively. Through analysis using the blockage migration prediction formula, it is predicted that these areas with low blockage degrees will all migrate to the pipeline branch point position 250 meters away from the starting point within the next 3 hours. Since the blockage degrees of these blockage areas are still lower than the preset 30% threshold after migration, the historical blockage position data of these blockage areas is obtained through the historical monitoring database of the industrial transmission pipeline. Combining the data of the past few months, it is observed that the blockage gradually migrates in the pipeline at a speed of 5 meters per hour. Use the ARIMA algorithm to analyze and model the historical position data of the blockage area, and predict that within the next 5 hours, two of these areas with low blockage degrees will aggregate at the pipeline branch point 300 meters away from the starting point of the pipeline, and the aggregation time is approximately 5 hours later. Based on this prediction result, immediately implement blockage migration risk prevention measures in advance at the branch point position of the pipeline, including slowing down the migration speed of the blockage by adjusting the fluid flow rate, and installing a temporary adjustment device at the pipeline branch point to guide the blockage to migrate to a safe area outside the main transmission pipeline. At the same time, increase the real-time monitoring frequency of this area to prevent serious blockage caused by the aggregation of the blockage.

[0050] Step S107: The prevention measure optimization module is used to continuously monitor the thermal imaging images of the industrial transmission pipeline, evaluate the implementation effects of the blockage risk prevention measures and the blockage migration risk prevention measures, and continuously optimize the prevention measures.

[0051] Continuously monitor industrial transmission pipelines through infrared thermal imaging technology, and regularly obtain thermal imaging images and corresponding temperature data of the industrial transmission pipelines. By comparing the thermal imaging images of the pipeline before and after implementing the blockage risk prevention measures and the blockage migration risk prevention measures, evaluate the effectiveness of the prevention measures. If the degree of blockage is reduced and the temperature distribution in the blocked area returns to normal, it is determined that the blockage risk prevention measures and the blockage migration risk prevention measures are effective. If the degree of blockage is not reduced, optimize the blockage risk prevention measures and the blockage migration risk prevention measures, including increasing the pipeline cleaning frequency, adjusting the transmission pressure or changing the flow rate, or changing the cleaning method for the blocked area. Implement the optimized blockage risk prevention measures and the blockage migration risk prevention measures, evaluate the effectiveness of the new prevention measures, and continuously optimize the blockage risk prevention measures and the blockage migration risk prevention measures based on the evaluation results of the effectiveness of the prevention measures.

[0052] Exemplarily, when continuously monitoring a certain industrial transmission pipeline, thermal imaging images and temperature data are obtained once an hour using infrared thermal imaging technology. After analysis, it is found that the degree of blockage in a blocked area of the pipeline is 50%, and the temperature in this area is higher than the normal temperature, reaching 85°C. Based on this situation, blockage risk prevention measures are implemented, including adjusting the flow rate and cleaning some of the blockages, and comparing the pipeline conditions before and after the implementation of the prevention measures through thermal imaging images. Within 2 hours after the implementation of the prevention measures, the temperature in the blocked area drops to the normal level of 55°C, and the degree of blockage is reduced to 20%. By comparing the thermal imaging images and the change in the degree of blockage, it is determined that the blockage risk prevention measures are effective this time, the temperature distribution in the blocked area returns to normal, and the blockage risk is significantly reduced. In another blocked area, after the implementation of the prevention measures, the situation in this blocked area does not improve, the degree of blockage remains at 50%, and the temperature in this area is still relatively high. Based on this, it is decided to optimize the current blockage risk prevention measures and the blockage migration risk prevention measures, specifically including increasing the pipeline cleaning frequency from once a week to once every 3 days, adjusting the pipeline transmission pressure from 10 MPa to 8 MPa, and increasing the flow rate from 1.2 m / s to 1.5 m / s to promote the further evacuation of the blockages. In addition, the cleaning method for the blocked area is also changed, adopting a combination of mechanical cleaning and chemical cleaning. After implementing the optimized prevention measures, continuous monitoring for 2 hours shows that the temperature in this blocked area drops to the normal value of 70°C, and the degree of blockage drops to 10%. Based on this result, it is confirmed that the optimized blockage risk prevention measures are effective, and it is decided to continue to adjust the cleaning frequency and flow rate according to the actual situation in the next monitoring cycle to further optimize the prevention measures and ensure the long-term safe operation of the pipeline.

[0053] The above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the technical solution formed by mutually replacing the above features with the technical features (but not limited to) disclosed in the present application that have similar functions.

Claims

1. An industrial equipment fault monitoring and diagnosis system based on infrared thermal imaging, characterized in that, The system includes: A pipeline blockage area identification module, which is used to obtain the thermal imaging image of the industrial transmission pipeline, identify the temperature abnormal area where the pipeline is blocked, and mark it as the pipeline blockage area; A thermal imaging feature index acquisition module, which is used to extract the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction of the thermal imaging image of the pipeline blockage area, and form the thermal imaging feature index of the pipeline blockage area; A blockage degree evaluation module, which is used to construct a pipeline blockage degree evaluation model according to the thermal imaging feature index of the historical blockage area of the industrial transmission pipeline, evaluate the blockage degree of the pipeline blockage area of the industrial transmission pipeline, and formulate blockage risk prevention measures for the pipeline blockage area where the blockage degree is higher than the preset degree threshold; A pipeline blockage prediction module, which is used to determine the heat diffusion speed of different time periods in the pipeline blockage area according to the temperature values of the pipeline blockage area at different times, construct a thermal imaging feature index prediction model, predict the time point when the blockage degree of the pipeline blockage area with a heat diffusion speed greater than the preset speed threshold is higher than the preset degree threshold, and implement blockage risk prevention measures in advance; A pipeline blockage area migration prediction module, which is used to judge whether the pipeline blockage area migrates according to the blockage position of the pipeline blockage area, and predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes based on the migration data of the pipeline blockage area, and implement blockage migration risk prevention measures for the pipeline blockage area where the blockage degree is higher than the preset degree threshold; The pipeline blockage area migration prediction module includes: Determine the blockage location of the pipeline blockage area according to the position with the highest temperature value in the thermal imaging image of the pipeline blockage area; obtain the historical blockage location data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, and combine it with the current blockage location of the pipeline blockage area to determine whether the pipeline blockage area has migrated; if the pipeline blockage area has migrated, obtain the migration data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, including the migration distance, migration speed, and migration direction, and use the ARIMA algorithm to predict the position of the pipeline where the pipeline blockage area is located in the future time period; if it is predicted that the position of the pipeline where the pipeline blockage area is located in the future time period has changed, then according to the thermal imaging feature index prediction model, predict the thermal imaging feature index when the position of the pipeline where the pipeline blockage area is located changes; according to the blockage migration prediction formula , predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes, where S(t) is the blockage degree at time t, S1 is the blockage degree before migration, D1 and D2 are the diameters of the pipeline before and after migration respectively, k is the coefficient of the influence of the pipe diameter change on the blockage degree, obtained by fitting historical data, M1 and M2 are the material parameters of the pipeline before and after migration respectively, defined according to the friction coefficient and corrosion resistance characteristics of the material, l is the influence coefficient of the material change on the blockage degree, obtained by fitting historical data, F z is the influence factor related to the pipeline position z, obtained by fitting historical data, and m represents the number of pipeline positions passed during the migration process; implement blockage migration risk prevention measures for pipeline blockage areas with a blockage degree higher than the preset degree threshold, including optimizing the interval time of fluid transportation, using pipeline regulating devices to guide the blockage to migrate in a specified direction, and optimizing the flow rate by adjusting the power of the pump or the opening and closing degree of the valve; A pipeline blockage area aggregation prediction module, which is used to predict the aggregation position and aggregation time of the pipeline blockage area according to the migration speed and migration direction of each pipeline blockage area, and implement blockage migration risk prevention measures for the aggregation area of the pipeline blockage area in advance; A prevention measure optimization module, which is used to continuously monitor the thermal imaging image of the industrial transmission pipeline, evaluate the implementation effect of the blockage risk prevention measures and the blockage migration risk prevention measures, and continuously optimize the prevention measures.

2. The system according to claim 1, wherein The pipeline blockage area identification module, which is used to obtain the thermal imaging image of the industrial transmission pipeline, identify the temperature abnormal area where the pipeline is blocked, and mark it as the pipeline blockage area, includes: Comprehensively scan the industrial transmission pipeline according to a predetermined scanning path and speed, obtain and save the thermal imaging image of the industrial transmission pipeline; convert the thermal imaging image of the industrial transmission pipeline into a grayscale image, set the temperature threshold as the segmentation standard, perform binary processing on the grayscale image, and mark the image area with a temperature higher than the set temperature threshold as the temperature abnormal area; obtain the thermal imaging image of the historical temperature abnormal area of the industrial transmission pipeline through the historical monitoring database of the industrial transmission pipeline, and mark whether the pipeline is blocked, use a convolutional neural network for model training, and construct a pipeline blockage identification model; according to the newly obtained thermal imaging image of the temperature abnormal area, use the pipeline blockage identification model to identify the temperature abnormal area where the pipeline is blocked, and mark it as the pipeline blockage area.

3. The system according to claim 1, wherein The thermal imaging feature index acquisition module is used to extract the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction of the thermal imaging image of the pipeline blockage area, and form the thermal imaging feature index of the pipeline blockage area, including: According to the thermal imaging image of the pipeline blockage area, extract the temperature values of each pixel point in the thermal imaging image to obtain the temperature distribution data of each pixel point; according to the temperature values in the thermal imaging image, use the central difference method to calculate the temperature change rate between adjacent pixel points to determine the average temperature gradient value of the pipeline blockage area; according to the temperature values in the thermal imaging image, calculate the difference between the highest temperature and the lowest temperature to obtain the maximum temperature difference of the pipeline blockage area; use the Sobel operator to calculate the gradient direction of each pixel point to determine the standard deviation of the gradient direction; standardize the highest temperature, average temperature gradient, maximum temperature difference, and standard deviation of the gradient direction, and combine them into a feature vector to form the thermal imaging feature index of the pipeline blockage area.

4. The system according to claim 1, wherein, The blockage degree evaluation module is used to construct a pipeline blockage degree evaluation model according to the thermal imaging feature index of the historical blockage area of the industrial transmission pipeline, evaluate the blockage degree of the pipeline blockage area of the industrial transmission pipeline, and formulate blockage risk prevention measures for the pipeline blockage area where the blockage degree is higher than the preset degree threshold, including: Through the historical monitoring database of the industrial transmission pipeline, obtain the thermal imaging feature index, blockage degree, pipe diameter, material, and location of the historical blockage area of the industrial transmission pipeline, and use a recurrent neural network for model training to construct a pipeline blockage degree evaluation model. The materials of the pipeline include steel, stainless steel, and PVC, and the locations of the pipeline include the main transmission pipeline, pipeline elbow, pipeline valve, pipeline branch point, pipeline connection point, and height change point; according to the thermal imaging feature index of the pipeline blockage area obtained in real time, combined with the pipe diameter, material, and location of the pipeline, use the pipeline blockage degree evaluation model to evaluate the blockage degree of the pipeline blockage area of the current industrial transmission pipeline; formulate and implement blockage risk prevention measures for the pipeline blockage area where the blockage degree is higher than the preset degree threshold, including cleaning the debris inside the pipeline, increasing the monitoring frequency, regularly performing thermal imaging detection, optimizing the transmission pressure or flow rate of the pipeline.

5. The system according to claim 1, wherein The pipeline blockage prediction module is used to determine the thermal diffusion speed of different time periods in the pipeline blockage area according to the temperature values of different moments in the pipeline blockage area, and construct a thermal imaging feature index prediction model to predict the time point when the blockage degree of the pipeline blockage area where the thermal diffusion speed is greater than the preset speed threshold is higher than the preset degree threshold, and implement blockage risk prevention measures in advance, including: Adopt the method of regular sampling to obtain the thermal imaging images of the pipeline blockage area with the blockage degree lower than the preset degree threshold at a preset time interval, extract the temperature distribution data at each moment, and determine the temperature value of the pipeline blockage area at each moment; According to the temperature values of the pipeline blockage area at different moments, use the thermal diffusion velocity calculation formula , calculate the thermal diffusion velocity of different periods in the pipeline blockage area, where T i and T j are the temperature values at time t i and time t j respectively, α is the thermal diffusion coefficient, which is determined based on the pipeline material properties; If the thermal diffusion velocity of a certain period in the pipeline blockage area is greater than the preset velocity threshold, obtain the historical thermal imaging characteristic indexes of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline, and organize them into time series data; According to the time series data of the historical thermal imaging characteristic indexes, use the ARIMA algorithm for model training to construct a thermal imaging characteristic index prediction model, predict the thermal imaging characteristic indexes within a preset future time period, and combine with the pipeline blockage degree evaluation model to judge the time point when the blockage degree of the pipeline blockage area is higher than the preset degree threshold, and implement blockage risk prevention measures for the pipeline blockage area in advance.

6. The system according to claim 1, wherein The pipeline blockage area aggregation prediction module is used to predict the aggregation location and aggregation time of the pipeline blockage area according to the migration speed and migration direction of each pipeline blockage area, and implement blockage migration risk prevention measures in advance for the aggregation area of the pipeline blockage area, including: If there are migrations of several pipeline blockage areas in an industrial transmission pipeline with a blockage degree lower than a preset degree threshold, then according to the blockage migration prediction formula, predict the blockage degree when the position of the pipeline where the pipeline blockage area is located changes; if the predicted blockage degree when the position of the pipeline where the pipeline blockage area is located changes is lower than the preset degree threshold, obtain the blockage position data of the pipeline blockage area through the historical monitoring database of the industrial transmission pipeline; according to the blockage position data of the pipeline blockage area, use the ARIMA algorithm to predict the aggregation position and aggregation time of the pipeline blockage area; based on the predicted aggregation position and aggregation time of the pipeline blockage area, implement blockage migration risk prevention measures in advance for the aggregation area of the pipeline blockage area.

7. The system according to claim 1, wherein, The prevention measure optimization module is used to continuously monitor the thermal imaging images of the industrial transmission pipeline, evaluate the implementation effects of the blockage risk prevention measures and the blockage migration risk prevention measures, and continuously optimize the prevention measures, including: Continuously monitor the industrial transmission pipeline through infrared thermal imaging technology, and regularly obtain the thermal imaging images and corresponding temperature data of the industrial transmission pipeline; evaluate the effectiveness of the prevention measures by comparing the thermal imaging images of the pipeline before and after implementing the blockage risk prevention measures and the blockage migration risk prevention measures; if the blockage degree is reduced and the temperature distribution of the blockage area returns to normal, it is determined that the blockage risk prevention measures and the blockage migration risk prevention measures are effective; if the blockage degree is not reduced, optimize the blockage risk prevention measures and the blockage migration risk prevention measures, including increasing the pipeline cleaning frequency, adjusting the transmission pressure or changing the flow rate, or changing the cleaning method for the blockage area; implement the optimized blockage risk prevention measures and the blockage migration risk prevention measures, evaluate the effectiveness of the new prevention measures, and continuously optimize the blockage risk prevention measures and the blockage migration risk prevention measures based on the evaluation results of the effectiveness of the prevention measures.

Images