Coal conveying belt double-circular-rail inspection robot and voiceprint fault early warning method

Abstract

The invention relates to a coal conveying belt double-circular-rail inspection robot and a voiceprint fault early warning method. The coal conveying belt double-circular-rail inspection robot comprises an inspection robot body and circular pipe rails. The top of the inspection robot is rotationally connected with walking wheels, and the walking wheels are in rolling connection to the circular pipe track. The inspection robot is hung below the double-circular-tube track through the walking wheels; a voiceprint fault detection early warning device is arranged in the inspection robot, the voiceprint fault detection early warning device comprises a microphone array, a camera, an infrared sensor and a control panel, the control panel is installed in the inspection robot, and the microphone array, the camera and the infrared sensor are installed on the outer side of the inspection robot and connected with the control panel; the control panel is connected with a robot management and control platform through signals. The double-round-pipe track has the advantages that coal ash adhesion is reduced by the aid of the round-face corner-free characteristic, and the problem that the inspection efficiency and the service life of the track are affected by robot operation clamping stagnation due to the fact that a traditional single-rail or I-steel track is prone to accumulation of coal ash and slag bonding is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of industrial robots, and particularly relates to a coal belt double-circular-track inspection robot and a voiceprint fault early warning method. BACKGROUND

[0002] The traditional coal belt inspection robot adopts a single track or an I-beam track, which is prone to accumulate coal ash and slag in a high dust concentration and high humidity environment, causing the robot to run and jam, affecting the inspection efficiency and the track service life. The traditional coal belt inspection robot relies on vision or temperature sensors, but the identification accuracy of hidden mechanical faults such as belt roller wear and bearing jamming is insufficient, and the device voiceprint features cannot be collected in real time for early warning. Therefore, there is an urgent need to design a coal belt inspection robot that does not jam and has high fault identification accuracy. SUMMARY

[0003] The purpose of the application is to overcome the shortcomings of the prior art and provide a coal belt double-circular-track inspection robot and a voiceprint fault early warning method.

[0004] The coal belt double-circular-track inspection robot comprises an inspection robot and a circular pipe track. The top of the inspection robot is rotatably connected with a walking wheel, and the walking wheel is rollingly connected to the circular pipe track. Two parallel circular pipe tracks form a double-circular-pipe track, and the inspection robot is hung below the double-circular-pipe track through the walking wheel. A voiceprint fault detection and early warning device is arranged in the inspection robot. The voiceprint fault detection and early warning device comprises a microphone array, a camera, an infrared sensor and a control board. The control board is installed in the inspection robot, and the microphone array, the camera and the infrared sensor are installed on the outer side of the inspection robot and connected with the control board. The control board is signal-connected with a robot control platform.

[0005] Preferably, the circular pipe track is connected by a plurality of circular pipe tracks; track supports are connected to the middle section and both ends of the side wall of the circular pipe track, and the track supports and the circular pipe track are fixed by chemical bolts; and mounting holes are arranged on the top of the track supports for hoisting in a factory building.

[0006] Preferably, the double-circular-pipe track adopts an arc-shaped transition in the curved track area.

[0007] Preferably, grooves are formed in the middle of the two sides of the walking wheel, and the grooves are matched with the circular pipe track.

[0008] Preferably, the circular pipe track is made of aluminum alloy, and the surface is subjected to anodic oxidation and salt mist treatment.

[0009] The voiceprint fault early warning method of the coal belt double-circular-track inspection robot comprises the following steps:

[0010] Step one, after the control panel loads the fault detection algorithm, the inspection robot is started to move on the circular pipe track, the microphone array collects the voiceprint signal of the belt area, and the control panel locates the voiceprint signal through video data;

[0011] Step two, the control panel extracts the frequency spectrum features of the voiceprint signal through timing analysis;

[0012] Step three, the control panel analyzes the frequency spectrum features of the voiceprint signal in real time through the fault detection algorithm;

[0013] Step four, when the analysis is abnormal, the infrared sensor is started to verify the abnormal area, and after confirming the abnormality, the control panel transmits the fault information to the robot control platform.

[0014] As preferred, in step one, the microphone array is built-in with a noise reduction filter module, which removes the background noise in the voiceprint signal.

[0015] As preferred, in step two, the control panel extracts the frequency spectrum features of the voiceprint signal through fast Fourier transform.

[0016] As preferred, in step three, the fault detection algorithm includes a fault classification model, which labels the video data at normal and abnormal voiceprints, combines the frequency spectrum features of normal and abnormal voiceprint signals with the video data labels to obtain a voiceprint database, and trains a deep convolutional network with the voiceprint database data to obtain the fault classification model; the fault detection algorithm compares and analyzes the frequency spectrum features of the collected voiceprint signal with the fault classification model data.

[0017] The beneficial effects of the present application are:

[0018] 1) The track of the inspection robot is set as a double circular pipe track in the present application, which reduces the adhesion of coal ash by utilizing the edgeless characteristics of the circular surface, solves the problem that the traditional single rail or I-beam track is easy to accumulate coal ash and slag in high dust concentration and high humidity environments, and causes the robot to run and jam, affecting the inspection efficiency and the service life of the track.

[0019] 2) The present application positions the detection area through the camera, collects the frequency spectrum features of the voiceprint signal of the detection area through the microphone array with noise reduction filter, detects the frequency spectrum features of the voiceprint signal of the detection area combined with the fault classification model, realizes early warning of hidden mechanical faults such as belt roller wear, bearing abnormal noise and roller eccentricity, and has high warning accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a schematic diagram of the robot track structure;

[0021] Figure 2 is Figure 1 the local enlarged view of A part in the robot track structure

[0022] Figure 3 This is a diagram of the robot's track layout.

[0023] Explanation of reference numerals in the attached diagram: 1. Circular tube track; 2. Track support; 3. Chemical bolt. Detailed Implementation

[0024] The present invention will be further described below with reference to embodiments. The description of the embodiments below is only for the purpose of helping to understand the present invention. It should be noted that those skilled in the art can make several modifications to the present invention without departing from the principle of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

[0025] Example 1

[0026] As one embodiment, a dual-circular-rail inspection robot for coal conveyor belts is proposed, such as... Figures 1-3 As shown, it includes: an inspection robot and a circular tube track 1; the top of the inspection robot is rotatably connected to a walking wheel, which is rolled on the circular tube track 1. The walking wheel has a groove formed on both sides of its circumference, which is higher than the middle. The groove fits into the circular tube track 1 to prevent the inspection robot from derailing; two parallel circular tube tracks 1 form a double circular tube track, and the inspection robot is suspended below the double circular tube track by the walking wheel; the inspection robot is equipped with a voiceprint fault detection and early warning device, which includes a microphone array, a camera, an infrared sensor, and a control board. The control board is installed inside the inspection robot, and the microphone array, camera, and infrared sensor are installed on the outside of the inspection robot and connected to the control board. The control board is connected to a robot management platform via a signal.

[0027] The circular tube track 1 adopts an anti-ash accumulation design. Its parallel, suspended layout utilizes the rounded, edgeless nature of its surface to reduce coal ash adhesion and avoid wheel jamming caused by single-track slag accumulation. The circular tube track 1 consists of several sections connected end-to-end, each section being 3m long and 32mm in outer diameter. The circular tube track 1 employs flexible installation, with track supports 2 connected to the middle section and both end sidewalls. The track supports 2 and the circular tube track 1 are fixed with chemical bolts 3. Each section of circular tube track 1 is hoisted using chemical bolts 3, with track supports 2 spaced 1.5m apart to enhance stability. The track supports 2 have mounting holes at the top for hoisting within the plant. The circular tube track 1 is modularly assembled, supporting 30°, 45°, and 90° curved track combinations to adapt to complex spatial layouts of coal conveyor bridges. The curved sections of the double circular tube track feature an arc transition. The circular tube track 1 is made of aluminum alloy with anodized and anti-salt spray treatment.

[0028] The machine head and tail of the coal conveying belt machine are provided with various devices for ensuring normal operation of the belt. The daily point inspection and operation and maintenance of the devices are completed manually, which is low in production efficiency and cannot timely discover and report fault conditions. The soundprint fault detection and early warning device carried on the inspection robot is developed to realize all-weather online intelligent inspection of the belt machine device, ensure timely discovery of faults and alarm, and ensure the investment rate of the device.

[0029] Embodiment Two

[0030] As another embodiment, embodiment two is proposed on the basis of embodiment one. A soundprint fault early warning method of a coal conveying belt double-circular-track inspection robot is shown in Figures 1-3 and includes the following steps:

[0031] Step one, after the control board loads the fault detection algorithm, the inspection robot is started to move on the circular pipe track 1, the microphone array collects the soundprint signal of the belt area, and the control board positions the soundprint signal through the video data. Specifically, the microphone array is built-in with a noise reduction filter module, which removes the background noise in the soundprint signal.

[0032] Step two, the control board extracts the frequency spectrum features of the soundprint signal through time sequence analysis. Specifically, the time sequence analysis includes fast Fourier transform, and the control board extracts the frequency spectrum features of the soundprint signal through fast Fourier transform. The frequency spectrum features include the fundamental frequency and harmonic energy distribution.

[0033] Step three, the control board analyzes the frequency spectrum features of the soundprint signal in real time through the fault detection algorithm. Specifically, the fault detection algorithm includes a fault classification model. The frequency spectrum features of the normal and abnormal soundprints and the video data labels are combined to obtain a soundprint database by labeling the labels of the normal and abnormal soundprints. The fault classification model is obtained by training the deep convolution network with the soundprint database data. The fault detection algorithm compares and analyzes the frequency spectrum features of the collected soundprint signal and the fault classification model data. The more sample data the model training has, the higher the fault alarm accuracy of the fault detection algorithm is.

[0034] Step four, when an anomaly is analyzed, the infrared sensor is started to verify the abnormal area. After confirming the anomaly, the control board transmits the fault information to the robot control platform to realize visual monitoring.

[0035] It should be noted that the same or similar parts in this embodiment and embodiment one can be mutually referenced, and will not be described herein.

[0036] The embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts of each embodiment can be mutually referenced.

Claims

1. A coal belt double circular rail inspection robot, characterized in that, The utility model relates to a kind of inspection robot and round tube track. Walking wheel is rotatably connected to the top of the inspection robot, and the walking wheel is rollably connected to the round tube track. Two parallel round tube tracks form double round tube tracks, and the inspection robot is hung below the double round tube tracks by the walking wheel.

2. The coal belt double circular rail inspection robot of claim 1, wherein, The middle section and both ends of the round tube track are connected with track supports on the side walls, and the track supports and the round tube track are fixed by chemical bolts.

3. The coal belt double circular rail inspection robot of claim 1, wherein, The top of the track support is provided with a mounting hole for hoisting in the factory building.

4. The coal belt double circular rail inspection robot of claim 1, wherein, The curved track area of the double round tube track adopts arc transition.

5. The coal belt double circular rail inspection robot of claim 1, wherein, The concave groove is formed in the middle of the two sides of the walking wheel circumference, and the concave groove is matched with the round tube track.

6. The voiceprint fault early warning method of the coal conveying belt double-circular-track inspection robot according to claim 1, characterized in that, The round tube track is made of aluminum alloy, and the surface is treated with anodic oxidation and salt mist resistance. The utility model includes the following steps: Step one: after the control board loads the fault detection algorithm, the inspection robot is started to move on the round tube track, the microphone array collects the voiceprint signal of the belt area, and the control board locates the voiceprint signal through video data; Step two: the control board extracts the frequency spectrum characteristics of the voiceprint signal through time sequence analysis; Step three: the control board analyzes the frequency spectrum characteristics of the voiceprint signal in real time through the fault detection algorithm; 7. The voiceprint fault early warning method of the coal conveying belt double-circular-track inspection robot according to claim 6, characterized in that, Step four: when the analysis is abnormal, the infrared sensor is started to verify the abnormal area, and after confirming the abnormality, the control board transmits the fault information to the robot control platform.

8. The voiceprint fault early warning method of the coal conveying belt double-circular-track inspection robot according to claim 6, characterized in that, In step one, the noise reduction filter module is built-in the microphone array, which removes the background noise in the voiceprint signal.

9. The voiceprint fault early warning method of the coal conveying belt double-circular-track inspection robot according to claim 6, characterized in that, In step two, the control board extracts the frequency spectrum characteristics of the voiceprint signal through fast Fourier transform. In step three, the fault detection algorithm includes a fault classification model, which labels the video data of normal and abnormal voiceprints, combines the frequency spectrum characteristics of normal and abnormal voiceprint signals with the video data labels to obtain a voiceprint database, and trains a deep convolutional network using the voiceprint database data to obtain the fault classification model; The fault detection algorithm compares and analyzes the frequency spectrum characteristics of the collected voiceprint signal with the fault classification model data.

Images