A method for inspecting coking oven tracks in coal chemical industry

Abstract

The present invention discloses a method for inspecting the track of a coal chemical coke oven. The method for inspecting the track of the coal chemical coke oven includes building a system platform, and the system platform includes a field information interaction layer, an Internet of Things big data platform, a business-level application layer, and a decision-making-level application layer; based on the field information interaction layer in the system platform, building a computer vision recognition system, and using the computer vision recognition system to achieve machine-assisted judgment on the correctness of the cock position; building a digital audio analysis system, and using the digital audio analysis system to achieve machine-assisted judgment on abnormal sounds on site; completing the inspection of the chemical industry track; having the advantages of improving the inspection frequency and inspection quality, etc.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical industry inspection equipment methods, and particularly to a method for inspecting the tracks of a coal chemical coke oven. Background Art

[0002] 61 pairs of exchange cocks are arranged in the underground coke oven chamber. The cocks are connected to the exchange chain through connecting rods and realize synchronous rotation actions of the cocks under the drive of a hydraulic actuator. A rotation switching action is performed every 24 minutes. The cocks have two working positions of +45° and -45°, and inspections need to be carried out after each switching action is completed to confirm whether their rotation positions are normal. At present, the manual visual inspection method is adopted. The length of the basement roadway is 89 meters. The inspection cycle is the same as the action cycle of the cocks, which is 24 minutes, and the single inspection duration is about 5 to 6 minutes. The main explosive gas in the working condition environment is carbon monoxide, and the height that the inspection passage can pass through is about 1.8 meters. During the action process of the cocks, a detonation phenomenon will occur, and abnormal sounds need to be paid attention to during the inspection process. At present, the manual visual inspection method has problems such as low inspection frequency, difficulty in ensuring inspection accuracy, and increasing labor costs year by year.

[0003] The present invention provides a method for inspecting the tracks of a coal chemical coke oven to meet the above requirements. Summary of the Invention

[0004] The technical problem to be solved by the present invention is that the current manual visual inspection method for the underground coke oven chamber and the coke oven corridor has problems such as low inspection frequency, difficulty in ensuring inspection accuracy, and increasing labor costs year by year;

[0005] Provide a method for inspecting the tracks of a coal chemical coke oven, and the method for inspecting the tracks of a coal chemical coke oven includes:

[0006] Build a system platform, and the system platform includes a field information interaction layer, an Internet of Things big data platform, a business-level application layer, and a decision-making-level application layer;

[0007] Based on the field information interaction layer in the system platform, build a computer vision recognition system, and use the computer vision recognition system to realize machine-assisted judgment of the correctness of the cock position;

[0008] Build a digital audio analysis system, and use the digital audio analysis system to realize machine-assisted judgment of abnormal sounds on site;

[0009] Complete the inspection of the chemical track.

[0010] Furthermore, the present application provides a coal chemical coke oven track inspection method, wherein the on-site information interaction layer includes an inspection robot, a pan-tilt camera, a thermal imager, an on-site microphone, a hazardous gas sensor, a temperature and humidity sensor, a smoke detector, an anti-collision radar, an on-site speaker and an sound and light alarm, and the inspection robot, the pan-tilt camera, the thermal imager, the on-site microphone, the hazardous gas sensor, the temperature and humidity sensor, the smoke detector, the anti-collision radar, the on-site speaker and the sound and light alarm are simultaneously connected to the Internet of Things big data platform.

[0011] Furthermore, the present application provides a coal chemical coke oven track inspection method, wherein the Internet of Things big data platform includes an industrial Internet of Things and a field information metadata database, the industrial Internet of Things is used to connect the equipment in the field information interaction layer to the Internet of Things for information interaction, and the field information metadata database is used to store information collected by the field information interaction layer.

[0012] Furthermore, the present application provides a coal chemical coke oven track inspection method, wherein the business-level application layer and the enterprise-level application layer are connected via an enterprise-level local area network, and the business-level application layer includes an inspection equipment management unit, a fault analysis and prediction unit, a production safety information display unit, and a remote inspection unit;

[0013] The decision-making application layer includes a production safety management unit, an emergency event decision-making auxiliary unit, and a joint prevention and control platform scheduling unit.

[0014] Furthermore, the present application provides a method for inspecting a track of a coal chemical coke oven, wherein the method includes building a computer vision recognition system based on the on-site information interaction layer in the system platform, and using the computer vision recognition system to perform machine-assisted judgment on the correctness of the cock position, including:

[0015] Capture videos using a pan-tilt camera and thermal imager, and pre-process the images in the videos;

[0016] Locate and crop the logo in the image;

[0017] Rectify the cropped image;

[0018] Binarize and reduce noise on the cropped and rectified images;

[0019] Extract and process the cropped, rectified, binarized and denoised images;

[0020] Calculate the position of Cock by the formula;

[0021] Upload the calculation results to the business-level application layer;

[0022] Use the business-level application layer to determine whether the cock position is correct.

[0023] Furthermore, a method for inspecting the track of a coal chemical coke oven provided by the present application, wherein the calculation of the cock position by the formula includes:

[0024] Calculate the T face of the cock and the handle face of the cock.

[0025] Furthermore, a method for inspecting the track of a coal chemical coke oven provided by the present application, wherein the calculation of the T face of the cock includes:

[0026] Calculate the T face of the cock using the dot product operation of vectors.

[0027] Furthermore, a method for inspecting the track of a coal chemical coke oven provided by the present application, wherein the calculation of the handle face of the cock includes

[0028] Calculate the handle position of the cock and the bolt position of the cock using the dot product of vectors.

[0029] Furthermore, a method for inspecting the track of a coal chemical coke oven provided by the present application, wherein the establishment of a digital audio analysis system and the use of the digital audio analysis system to achieve machine-assisted judgment of on-site abnormal sounds include:

[0030] Audio signal sampling and quantization;

[0031] Perform FFT transformation to obtain the time-frequency spectrum;

[0032] Extract audio features;

[0033] Compare the feature data;

[0034] Report the comparison result to the business-level application layer..

[0035] Implementing the present invention has the following beneficial effects:

[0036] 1. The present invention can realize the automatic replacement of the current manual inspection method without interfering with the manual inspection, can realize the functions of remote monitoring, recording, and abnormal alarm of the on-site working conditions, and at the same time can improve the inspection frequency, improve the inspection accuracy, and reduce the labor cost. Brief Description of the Drawings

[0037] Figure 1 is a flowchart of a method for inspecting the track of a coal chemical coke oven according to the present invention;

[0038] Figure 2 is a schematic diagram of the calculation of the T face of the cock in a method for inspecting the track of a coal chemical coke oven according to the present invention;

[0039] Figure 3It is a schematic diagram for calculating the cock handle surface in a method for inspecting the coke oven track in coal chemical industry according to the present invention;

[0040] Figure 4 It is a structural diagram of a system platform for a method for inspecting the coke oven track in coal chemical industry according to the present invention. Specific embodiments

[0041] To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0042] Embodiment

[0043] In this embodiment, referring to the attached drawings of the specification Figures 1-3 The technical problem to be solved in this embodiment is that the current manual visual inspection method for the underground coke oven chamber and the coke oven corridor has problems such as low inspection frequency, difficulty in ensuring inspection accuracy, and increasing labor cost year by year;

[0044] A method for inspecting the coke oven track in coal chemical industry is provided. The method for inspecting the coke oven track in coal chemical industry includes:

[0045] Building a system platform, the system platform includes a field information interaction layer, an Internet of Things big data platform, a business-level application layer and a decision-making-level application layer;

[0046] Based on the field information interaction layer in the system platform, building a computer vision recognition system, and using the computer vision recognition system to realize machine-assisted judgment on the correctness of the cock position;

[0047] Building a digital audio analysis system, and using the digital audio analysis system to realize machine-assisted judgment on abnormal sounds on site;

[0048] Completing the inspection of the chemical track.

[0049] In a specific embodiment, the field information interaction layer includes an inspection robot, a pan-tilt camera, a thermal imager, a field microphone, a hazardous gas sensor, a temperature and humidity sensor, a smoke detector, an anti-collision radar, a field speaker and an audible and visual alarm. The inspection robot, the pan-tilt camera, the thermal imager, the field microphone, the hazardous gas sensor, the temperature and humidity sensor, the smoke detector, the anti-collision radar, the field speaker and the audible and visual alarm are simultaneously connected to the Internet of Things big data platform.

[0050] In a specific embodiment, the Internet of Things big data platform includes an industrial Internet of Things and a field information meta-database. The industrial Internet of Things is used to connect the devices in the field information interaction layer to the Internet of Things for information interaction, and the field information meta-database is used to store the information collected by the field information interaction layer.

[0051] In a specific implementation, the business-level application layer and the enterprise-level application layer are network-connected through an enterprise-level local area network. The business-level application layer includes an inspection equipment management unit, a fault analysis and prediction unit, a safety production information display unit, and a remote inspection unit.

[0052] The decision-making level application layer includes a safety production management unit, an emergency event decision-making assistance unit, and a joint defense and control platform-based dispatching unit.

[0053] In a specific implementation, based on the on-site information interaction layer in the system platform, a computer vision recognition system is built. Using the computer vision recognition system to implement machine-assisted judgment on the correctness of the cock position includes:

[0054] Collect videos through a pan-tilt camera and an infrared thermal imager, and preprocess the images in the videos.

[0055] Locate and crop the identifiers in the images.

[0056] Rectify the cropped images.

[0057] Binarize and denoise the images that have been cropped and rectified.

[0058] Extract and process the images that have been cropped, rectified, binarized, and denoised.

[0059] Calculate the position of the cock through a formula.

[0060] Upload the calculation result to the business-level application layer.

[0061] Use the business-level application layer to judge whether the cock position is correct.

[0062] In a specific implementation, the calculation of the position of the cock through a formula includes:

[0063] Calculate the T-letter face of the cock and the handle face of the cock.

[0064] In a specific implementation, the calculation of the T-letter face of the cock includes:

[0065] Calculate the T-letter face of the cock using the dot product operation of vectors, as shown in the attached instructions Figure 3 , where for a certain cock, the cock position OA = +2° (when closed for the first time) and OA' = -2° (when closed for the second time) on both sides of the cock, and the normal range of the threshold is (-5°, +5°). The change amount between the two times is: Δ = POA - POA' = 4°. If it is within the threshold range, the device is operating normally. The calculation logic for OA equal to 2° is: (where the vector OX is the reference value in the horizontal direction when closed).

[0066] In a specific embodiment, the calculation of the cock handle surface includes

[0067] Calculating the cock handle position and the cock bolt position using the dot product of vectors, as shown in the attached instructions Figure 4 , where for a certain cock pair, the handle position AB = -48° (at the first commutation), and the cock handle position A'B' = -46° (assuming the third commutation), and the normal threshold range is (-5°, 5°). The change amount Δ = PAB - PA’B’ = -2°, which is within the threshold range, so the device is operating normally. The calculation logic for AB being equal to -48°: (where the vector OX is the reference value in the horizontal direction when closed), the bolt position relative to the handle position, that is, the angle between the vector AB and the vector ON, is also obtained through the dot product operation of vectors. The calculation formula is as follows (where the vector ON is the angle between the handle facing left and the horizontal direction).

[0068] In a specific embodiment, for the construction of the digital audio analysis system, using the digital audio analysis system to achieve machine-assisted judgment of on-site abnormal sounds includes:

[0069] Sampling and quantifying the audio signal;

[0070] Performing FFT transformation to obtain the time-frequency spectrum;

[0071] Extracting audio features;

[0072] Comparing the feature data;

[0073] Reporting the comparison result to the business-level application layer.

[0074] Implementing the present invention has the following beneficial effects:

[0075] 1. The present invention can achieve automated replacement of the current manual inspection method without interfering with manual inspection, can achieve functions of remote monitoring, recording, and abnormal alarm of on-site working conditions, and at the same time can increase the inspection frequency, improve the inspection accuracy, and reduce the labor cost.

[0076] What is disclosed above are only several preferred embodiments of the present invention. Of course, the scope of the rights of the present invention cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A method for inspecting the tracks of a coal chemical coke oven, characterized in that, Including: Build a system platform, which includes a field information interaction layer, an Internet of Things big data platform, a business-level application layer, and a decision-making-level application layer; Based on the field information interaction layer in the system platform, build a computer vision recognition system, and use the computer vision recognition system to achieve machine-assisted judgment of the correctness of the cock position; Build a digital audio analysis system, and use the digital audio analysis system to achieve machine-assisted judgment of abnormal sounds on site; Complete chemical track inspection; Among them, calculating the position of the cock through formulas includes: Calculating the T surface of the cock and calculating the handle surface of the cock; Among them, the calculation of the T surface of the cock includes: Calculating the T surface of the cock using the dot product operation of vectors; Among them, the calculation of the handle surface of the cock includes: Calculating the handle position of the cock and calculating the bolt position of the cock using vector dot product; Among them, building the digital audio analysis system and using the digital audio analysis system to achieve machine-assisted judgment of abnormal sounds on site includes: Audio signal sampling and quantization; Performing FFT transformation to obtain the time-frequency spectrum; Extracting audio features; Comparing the feature data; Reporting the comparison result to the business-level application layer.

2. The coal chemical coke oven track inspection method according to claim 1, characterized in that The field information interaction layer includes an inspection robot, a pan-tilt camera, a thermal imager, a field microphone, a hazardous gas sensor, a temperature and humidity sensor, a smoke detector, an anti-collision radar, a field speaker, and an audible and visual alarm. The inspection robot, the pan-tilt camera, the thermal imager, the field microphone, the hazardous gas sensor, the temperature and humidity sensor, the smoke detector, the anti-collision radar, the field speaker, and the audible and visual alarm are simultaneously connected to the Internet of Things big data platform.

3. The coal chemical coke oven track inspection method according to claim 2, wherein The Internet of Things big data platform includes an industrial Internet of Things and a field information meta-database. The industrial Internet of Things is used to connect the devices in the field information interaction layer to the Internet of Things for information interaction, and the field information meta-database is used to store the information collected by the field information interaction layer.

4. The coal chemical coke oven track inspection method according to claim 3, characterized in that, The business-level application layer and the decision-making-level application layer are connected through an enterprise-level local area network. The business-level application layer includes an inspection equipment management unit, a fault analysis and prediction unit, a safety production information display unit, and a remote inspection unit; The decision-making-level application layer includes a safety production management unit, an emergency event decision-making assistance unit, and a joint defense and control platform scheduling unit.

5. The coal chemical coke oven track inspection method according to claim 4, wherein Based on the field information interaction layer in the system platform, building a computer vision recognition system and using the computer vision recognition system to achieve machine-assisted judgment of the correctness of the cock position includes: Collecting videos through a pan-tilt camera and a thermal imager, and preprocessing the images in the videos; Locating and cropping the identifiers in the images; Rectifying the cropped images; Binarizing and denoising the cropped, rectified, binarized, and denoised images; Extracting and processing the cropped, rectified, binarized, and denoised images; Calculating the position of the cock through formulas; Uploading the calculation result to the business-level application layer; Using the business-level application layer to judge whether the cock position is correct.

Images