A fault diagnosis method for planetary gears of rocker arms of coal mining machines

Abstract

The present application relates to a fault diagnosis method for the planetary gear of a coal mining machine rocker arm. The method comprises: the above-mentioned planetary gear temperature monitoring device installed on the planetary frame inside the coal mining machine rocker arm collects the temperature flow of the planetary gear in real time during the operation of the coal mining machine, obtains the infrared thermal image of the planetary gear, and then performs image denoising on the infrared thermal image, and then performs segmentation processing and edge refinement on the denoised planetary gear thermal image to obtain an image after the high temperature area is refined, and then according to the image after the high temperature area is refined, the high temperature area is marked on the infrared thermal image to obtain an infrared thermal image marked with the high temperature area of ​​the planetary gear, and the infrared thermal image marked with the high temperature area of ​​the planetary gear is input into the fault diagnosis model trained based on the VGG16 convolutional neural network for fault diagnosis, and the fault diagnosis result of the planetary gear is determined. The accuracy of the fault diagnosis of the planetary gear of the coal mining machine rocker arm is improved.

Description

Technical Field

[0001] The present application relates to the technical field of coal mining machinery, and in particular to a fault diagnosis method for a planetary gear of a rocker arm of a coal mining machine and a planetary gear temperature monitoring device. Background Art

[0002] The existing fault diagnosis of planetary gears in coal mining machine rocker arms relies on sensor data for fault detection and diagnosis. The existing fault diagnosis methods for planetary gears in coal mining machine rocker arms include: traditional diagnosis, signal processing, data-driven fault diagnosis methods, etc.

[0003] The existing diagnostic data acquisition methods have limitations. Traditional methods can directly obtain internal data of coal mining machines, but they are limited to data acquisition in non-working conditions and lack real-time and accuracy. Signal processing and data-driven methods can analyze data under coal mining machine working conditions in real time, but they are restricted by sensor types and working environments, making data processing, analysis and fault diagnosis difficult and the sensor data type is single.

[0004] Therefore, the accuracy of the planetary gear fault diagnosis of the current coal mining machine rocker arm is low. Summary of the invention

[0005] Based on this, it is necessary to provide a planetary gear fault diagnosis method and a planetary gear temperature monitoring device for a coal mining machine rocker arm that can improve the accuracy of planetary gear fault diagnosis of a coal mining machine rocker arm in response to the above technical problems.

[0006] A method for diagnosing a fault of a planetary gear of a rocker arm of a coal mining machine, the method comprising:

[0007] The temperature flow of the planetary gears during the operation of the coal mining machine is collected in real time by a planetary gear temperature monitoring device installed on the planetary carrier inside the rocker arm of the coal mining machine, and an infrared thermal image of the planetary gears is obtained;

[0008] Performing image denoising processing on the infrared thermal image to obtain a denoised thermal image of the planetary gear;

[0009] Segmenting the denoised planetary gear thermal image and performing edge thinning to obtain a high-temperature region thinned image;

[0010] According to the refined image of the high temperature area, the high temperature area is marked on the infrared thermal image to obtain an infrared thermal image with the high temperature area of ​​the planetary gear marked;

[0011] The infrared thermal image marked with the high-temperature area of ​​the planetary gear is input into a fault diagnosis model trained based on a VGG16 convolutional neural network to perform fault diagnosis, and determine the fault diagnosis result of the planetary gear.

[0012] In one embodiment, performing image denoising on the infrared thermal image to obtain a denoised planetary gear thermal image includes:

[0013] The infrared thermal image is input into a bilateral filter for bilateral filtering denoising, and the edge information of the image is maintained while removing the noise, so as to obtain a denoised planetary gear thermal image, wherein the expression of the bilateral filtering is:

[0014]

[0015]

[0016] in, is the updated pixel value of the center pixel of the box, W p is the normalization factor, is the spatial domain kernel, S is the filter window, is the image pixel domain kernel, I(p) is the value of the center pixel point p of the filter window, and I(q) is the value of the pixel point q of the infrared thermal image.

[0017] In one embodiment, the segmentation processing and edge thinning of the denoised planetary gear thermal image to obtain a refined image of the high temperature area includes:

[0018] Using a k-means algorithm to classify pixels with similar features in the denoised planetary gear thermal image, segmenting a high temperature area and a low temperature area in the denoised planetary gear thermal image to obtain a segmented image;

[0019] An opening operation is performed on the high temperature region in the segmented image to achieve edge thinning, thereby obtaining an image after the high temperature region is thinned.

[0020] In one embodiment, the expression of the opening operation is:

[0021]

[0022]

[0023] Where A is the high temperature area in the segmented image, B is the structural element, ⊙ is the corrosion operation on the high temperature area A in the segmented image, z is the pixel coordinate in the structural element, (B) z is the position of the structural element on the segmented image, To dilate the high temperature area A in the segmented image, Is an empty set.

[0024] In one embodiment, the step of marking the high temperature area on the infrared thermal image based on the image after the high temperature area is refined to obtain the infrared thermal image marked with the high temperature area of ​​the planetary gear includes:

[0025] Binarizing the image after the high temperature area is refined and then converting it into data of the same data type as the infrared thermal image to obtain converted data;

[0026] The converted data is multiplied by the infrared thermal image to achieve a masking effect on the high-temperature area of ​​the planetary gear in the infrared thermal image, so as to obtain an infrared thermal image marked with the high-temperature area of ​​the planetary gear.

[0027] In one embodiment, the fault diagnosis model is trained in the following manner:

[0028] Step 1, by means of a planetary gear temperature monitoring device installed on a planetary carrier inside a rocker arm of a coal mining machine, temperature flow samples of the planetary gears in different states during the operation of the coal mining machine are collected to obtain infrared thermal image samples of the planetary gears, wherein the different states include a normal state and various fault states;

[0029] Step 2, performing image denoising processing on the infrared thermal image sample to obtain a denoised planetary gear thermal image sample;

[0030] Step 3, segmenting the denoised planetary gear thermal image samples and performing edge thinning to obtain thinned planetary gear high temperature area samples;

[0031] Step 4, marking the planetary gear high temperature area samples onto the infrared thermal image samples, obtaining infrared thermal image samples marked with the planetary gear high temperature areas, and obtaining a planetary gear fault diagnosis data set;

[0032] Step 5, dividing the planetary gear fault diagnosis data set into a training set, a validation set and a test set according to a preset ratio;

[0033] Step 6, inputting the training set and the validation set into the VGG16 convolutional neural network for training, after multiple rounds of iterations, until the iteration stop condition is met, determining the final network parameters, updating the network parameters of the VGG16 convolutional neural network to the final network parameters, and obtaining a preliminarily trained fault diagnosis model;

[0034] Step 7, input the test set into the preliminarily trained fault diagnosis model to test the accuracy of the preliminarily trained fault diagnosis model. If the accuracy meets the requirement, the trained fault diagnosis model is obtained. If the accuracy does not meet the requirement, return to execute step 1 until the accuracy meets the requirement and the trained fault diagnosis model is obtained.

[0035] In one embodiment, the method further comprises:

[0036] The temperature flow of the planetary gears during the operation of the coal mining machine is collected in real time by the planetary gear temperature monitoring device installed on the planetary frame inside the rocker arm of the coal mining machine, and the full-screen temperature matrix of the infrared thermal image is obtained;

[0037] Determine the temperature value of each pixel point in the high-temperature area of ​​the refined planetary gear according to a predetermined corresponding function relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image;

[0038] A temperature cloud diagram is drawn according to the refined temperature values ​​of each pixel point in the high-temperature area of ​​the planetary gear, so as to visualize the temperature information of the planetary gear during the operation of the coal mining machine.

[0039] In one embodiment, the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image is determined as follows:

[0040] The planetary gear temperature monitoring device installed on the planetary carrier inside the rocker arm of the coal mining machine is used to collect temperature flow samples of the planetary gears in different states during the operation of the coal mining machine, and obtain infrared thermal image full-screen temperature matrix samples corresponding to the infrared thermal image samples of the planetary gears;

[0041] Converting the denoised planetary gear thermal image samples into grayscale image samples;

[0042] Screening out all points with integer temperature values ​​in the full-screen temperature matrix sample of the infrared thermal image, and recording coordinate information of the points with integer temperature values ​​to obtain a coordinate information set;

[0043] Using the grayscale value corresponding to each coordinate information in the coordinate information set, all grayscale values ​​corresponding to each integer temperature value are obtained;

[0044] Count the number of occurrences of each grayscale value among all grayscale values ​​corresponding to each integer temperature value, take the grayscale value with the largest number of occurrences as the grayscale value corresponding to the integer temperature value, and determine each pair of integer temperature value-grayscale value fitting samples;

[0045] Linear regression fitting is applied to each pair of integer temperature value-grayscale value fitting samples to obtain the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image.

[0046] A planetary gear temperature monitoring device, characterized in that the planetary gear temperature monitoring device comprises: an infrared sensor, an outer conductive slip ring mover, an outer conductive slip ring stator, an inner conductive slip ring mover, an inner conductive slip ring stator and a temperature flow receiving device;

[0047] An annular groove is formed on the surface of the planetary frame inside the rocker arm of the coal mining machine near the planetary gear and the sun gear, and the infrared sensor is fixed in the annular groove, so that the infrared sensor monitors the synchronous movement of the planetary gear and the sun gear in real time and takes temperature flow pictures;

[0048] According to the size of the outer conductive slip ring mover, the inner hole of the planet carrier is expanded so that there is a gap between the outer conductive slip ring mover and the inner hole wall of the planet carrier, and the infrared sensor is electrically connected to one end of the outer conductive slip ring mover through a cable, so that the outer conductive slip ring mover rotates synchronously with the planetary gear and the planet carrier;

[0049] The outer conductive slip ring stator is sleeved in the ring of the outer conductive slip ring mover, and the outer conductive slip ring stator is fixed on the outer wall of the output shaft of the coal mining machine rocker arm, so that the outer conductive slip ring stator rotates synchronously with the output shaft;

[0050] The inner conductive slip ring mover is fixedly mounted at the bottom end of the output shaft of the coal mining machine rocker arm in a flange-type manner, a through hole is opened on the output shaft near the location where the outer conductive slip ring stator is mounted, a cable is passed through the through hole to electrically connect one end of the outer conductive slip ring stator with one end of the inner conductive slip ring mover, and the inner conductive slip ring mover rotates synchronously with the outer conductive slip ring stator and the output shaft;

[0051] The inner conductive slip ring stator is sleeved in the inner conductive slip ring mover, and one end of the inner conductive slip ring stator away from the inner conductive slip ring mover is connected to a cable, through which the temperature flow receiving device or the downhole switch is connected. When the cable is connected to the downhole switch, the temperature flow receiving device is connected to the temperature flow receiving device network.

[0052] The above-mentioned fault diagnosis method for the planetary gear of the rocker arm of the coal mining machine is to collect the temperature flow of the planetary gear in real time during the operation of the coal mining machine through the planetary gear temperature monitoring device installed on the planetary frame inside the rocker arm of the coal mining machine, obtain the infrared thermal image of the planetary gear, and then perform image denoising on the infrared thermal image to obtain the denoised planetary gear thermal image, and then perform segmentation processing on the denoised planetary gear thermal image and perform edge refinement to obtain the image after the high temperature area is refined, and then according to the image after the high temperature area is refined, the high temperature area is marked on the infrared thermal image to obtain the infrared thermal image marked with the high temperature area of ​​the planetary gear, and the infrared thermal image marked with the high temperature area of ​​the planetary gear is input into the fault diagnosis model trained based on the VGG16 convolutional neural network for fault diagnosis, and the fault diagnosis result of the planetary gear is determined. Thereby, the accuracy of the fault diagnosis of the planetary gear of the rocker arm of the coal mining machine is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1It is a schematic flow chart of a method for diagnosing a fault of a planetary gear of a rocker arm of a coal mining machine in one embodiment;

[0054] Figure 2 is an infrared thermal image of a planetary gear in one embodiment;

[0055] Figure 3 is a denoised thermal image of a planetary gear in one embodiment;

[0056] Figure 4 is a segmented image in one embodiment;

[0057] Figure 5 An image after refinement of a high temperature region in one embodiment;

[0058] Figure 6 A schematic diagram of a high temperature area of ​​a planetary gear on an original thermal image in one embodiment;

[0059] Figure 7 A histogram of planetary gear fault identification accuracy in one embodiment;

[0060] Figure 8 A temperature cloud diagram of a high temperature area of ​​a planetary gear in one embodiment;

[0061] Fig. 9 A schematic diagram showing the comparison of data measured by the infrared sensor supporting software in one embodiment;

[0062] Fig.10 A schematic diagram of the local structure of a planetary gear temperature monitoring device on a rocker arm of a coal mining machine in one embodiment;

[0063] Fig.11 Another partial structural schematic diagram of a planetary gear temperature monitoring device on a coal mining machine rocker arm in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] In one embodiment, Figure 1 As shown, a fault diagnosis method for a planetary gear of a rocker arm of a coal mining machine is provided, comprising the following steps:

[0066] Step S220, a planetary gear temperature monitoring device installed on a planetary frame inside the rocker arm of the coal mining machine collects the temperature flow of the planetary gear in real time during the operation of the coal mining machine to obtain an infrared thermal image of the planetary gear.

[0067] The infrared sensor of the planetary gear temperature monitoring device is fixedly installed on the planetary frame inside the rocker arm of the coal mining machine, and is used to collect the temperature flow of the planetary gear in real time during the operation of the coal mining machine. The temperature flow collected by the infrared sensor is transmitted to the temperature flow receiving device of the planetary gear temperature monitoring device through the transmission line of the planetary gear temperature monitoring device. The temperature flow receiving device can generate an infrared thermal image of the planetary gear and an infrared thermal image full-screen temperature matrix according to the temperature flow. The infrared thermal image of the planetary gear is as follows: Figure 2 shown.

[0068] In one embodiment, before the planetary gear temperature monitoring device is used, the infrared sensor can be calibrated based on the experimental environment, and the calibrated infrared sensor can then collect the temperature flow of the planetary gear in real time during the operation of the coal mining machine. The accuracy of the collected temperature flow can be improved, and the accuracy of the fault diagnosis result of the planetary gear can be improved.

[0069] Among them, the infrared sensor is calibrated based on the experimental environment to obtain the calibrated infrared sensor. Specifically: input the current meteorological conditions, such as the current temperature, relative humidity, visibility, rainfall intensity, snowfall intensity, and the emissivity of the object to be measured (the emissivity can be filled in by yourself or you can query the emissivity table) and the distance parameters between the object to be measured and the infrared lens of the thermal imager. After the parameters are set, select the optional box to enable "temperature correction", click the "FFC" button, the built-in baffle of the thermal imager opens and closes once, and the camera temperature measurement benchmark is calibrated.

[0070] Specifically, the calibrated infrared sensor is aimed at the running planetary gear, and the calibrated infrared sensor is used to collect and process the image of the planetary gear to obtain the thermal image temperature flow of the planetary gear. The temperature flow is opened with the offline image analysis software (installed in the temperature flow receiving device) of the infrared sensor, and the temperature flow is saved as an infrared thermal image and a full-screen temperature matrix of the infrared thermal image in CSV format.

[0071] The infrared sensor may be D6T-32L-01A or an infrared imaging module with a PICO384 movement.

[0072] Step S240, performing image denoising processing on the infrared thermal image to obtain a denoised thermal image of the planetary gear.

[0073] In one embodiment, performing image denoising processing on the infrared thermal image to obtain a denoised thermal image of the planetary gear includes:

[0074] The infrared thermal image is input into the bilateral filter for bilateral filtering denoising. The edge information of the image is maintained while removing the noise, and the denoised planetary gear thermal image is obtained. The expression of bilateral filtering is:

[0075]

[0076]

[0077] in, is the updated pixel value of the center pixel of the box, W p is the normalization factor, is the spatial domain kernel, S is the filter window, is the image pixel domain kernel, I(p) is the value of the center pixel point p of the filter window, and I(q) is the value of the pixel point q of the infrared thermal image.

[0078] Among them, the denoised thermal image of the planetary gear is as follows: Figure 3 shown.

[0079] Step S260 , segmenting the denoised planetary gear thermal image and performing edge thinning to obtain a thinned image of the high temperature area.

[0080] In one embodiment, the denoised planetary gear thermal image is segmented and edge-refined to obtain a refined image of the high-temperature region, including:

[0081] The k-means algorithm is used to classify pixels with similar features in the denoised planetary gear thermal image, and the high-temperature area and the low-temperature area in the denoised planetary gear thermal image are segmented to obtain the segmented image; the high-temperature area in the segmented image is opened to achieve edge refinement, and the image after the high-temperature area is refined is obtained.

[0082] It should be understood that the k-means algorithm is used to automatically classify data with similar features in the denoised planetary gear thermal image to achieve the effect of segmenting the high temperature area and the low temperature area. The high temperature area segmentation result is subjected to morphological opening operation to achieve the effect of thinning the edge of the moving planetary gear.

[0083] The thermal image is essentially a representation of the temperature of an object. The denoised planetary gear thermal image is input into the k-means algorithm, and the data with similar features in the planetary gear thermal image are automatically classified together. Since the difference between the high temperature area and the low temperature area in the thermal image is obvious, setting the cluster of the k-means algorithm to 2 can achieve a better effect of segmenting the high temperature area and the low temperature area. Figure 4 shown.

[0084] In one embodiment, the expression of the open operation is:

[0085]

[0086]

[0087] Where A is the high temperature area in the segmented image, B is the structural element, ⊙ is the corrosion operation on the high temperature area A in the segmented image, z is the pixel coordinate in the structural element, (B) z is the position of the structural element on the segmented image, To dilate the high temperature area A in the segmented image, Is an empty set.

[0088] Among them, the opening operation consists of two basic operations: erosion and dilation, which are usually used for image thinning.

[0089] Among them, the high temperature area A in the segmented image is corroded, that is, for each pixel point z in the structural element B, the area centered on the point in the high temperature area A is logically "AND" operated with the structural element to obtain the pixel value of the corresponding position in the output image.

[0090] Among them, the high-temperature area A in the segmented image is expanded, that is, for each pixel point z in the structural element B, the area centered on the point in the high-temperature area A is logically "OR" operated with the structural element to obtain the pixel value of the corresponding position in the output image.

[0091] Among them, the high temperature area in the segmented image is opened to achieve edge refinement, and the image after the high temperature area refinement is obtained as follows Figure 5 shown.

[0092] Step S280, based on the refined image of the high temperature area, the high temperature area is marked on the infrared thermal image to obtain an infrared thermal image marked with the high temperature area of ​​the planetary gear.

[0093] In one embodiment, according to the refined image of the high temperature area, the high temperature area is marked on the infrared thermal image to obtain an infrared thermal image marked with the high temperature area of ​​the planetary gear, including:

[0094] Binarize the image after the high temperature area is refined and then convert it into data of the same data type as the infrared thermal image to obtain converted data;

[0095] The converted data is multiplied by the infrared thermal image to achieve a masking effect on the high-temperature area of ​​the planetary gear in the infrared thermal image, and an infrared thermal image marked with the high-temperature area of ​​the planetary gear is obtained.

[0096] Specifically, the image after the high temperature area refinement obtained after the morphological opening operation is binarized, and the value of the high temperature area is set to 1, and the value of other areas is set to 0. The binarization result is converted into the unit8 data type that is the same as the infrared thermal image data type of the planetary gear. The converted data is multiplied by the infrared thermal image of the planetary gear to achieve the mask effect of the high temperature area of ​​the planetary gear, and the high temperature area of ​​the planetary gear on the original thermal image is obtained, such as Figure 6 shown.

[0097] Step S300: Input the infrared thermal image marked with the high-temperature area of ​​the planetary gear into the fault diagnosis model trained based on the VGG16 convolutional neural network to perform fault diagnosis and determine the fault diagnosis result of the planetary gear.

[0098] In one embodiment, the fault diagnosis model is trained as follows:

[0099] Step 1, by installing a planetary gear temperature monitoring device on the planetary frame inside the rocker arm of the coal mining machine, collect temperature flow samples of the planetary gears in different states during the operation of the coal mining machine, and obtain infrared thermal image samples of the planetary gears, wherein the different states include normal states and various fault states;

[0100] Step 2, performing image denoising processing on the infrared thermal image samples to obtain denoised planetary gear thermal image samples;

[0101] Step 3, segmenting the denoised planetary gear thermal image samples and performing edge refinement to obtain refined planetary gear high temperature area samples;

[0102] Step 4, marking the planetary gear high temperature area samples onto the infrared thermal image samples, obtaining the infrared thermal image samples marked with the planetary gear high temperature area, and obtaining the planetary gear fault diagnosis data set;

[0103] Step 5, dividing the planetary gear fault diagnosis data set into a training set, a validation set, and a test set according to a preset ratio;

[0104] Step 6: Input the training set and the validation set into the VGG16 convolutional neural network for training. After multiple rounds of iterations, the final network parameters are determined until the iteration stop condition is met. The network parameters of the VGG16 convolutional neural network are updated to the final network parameters to obtain a preliminarily trained fault diagnosis model.

[0105] Step 7, input the test set into the preliminarily trained fault diagnosis model to test the accuracy of the preliminarily trained fault diagnosis model. If the accuracy meets the requirement, the trained fault diagnosis model is obtained. If the accuracy does not meet the requirement, return to step 1 until the accuracy meets the requirement and the trained fault diagnosis model is obtained.

[0106] Among them, by collecting temperature flow samples of planetary gears in different states during the operation of the coal mining machine, a planetary gear fault diagnosis data set is created to train the fault diagnosis model and realize accurate identification of planetary gear faults.

[0107] Among them, the infrared thermal image samples marked with the high temperature area of ​​the planetary gear are classified according to different states, such as normal state, pitting fault state, broken tooth fault state, bonding fault state and plastic deformation fault state. LabelImg is used to annotate the thermal images of the planetary gear running in 5 states to obtain an xml file. The infrared thermal image samples with the planetary gear running state label and marked with the high temperature area of ​​the planetary gear constitute the planetary gear fault diagnosis data set.

[0108] In one embodiment, the high temperature area can be retained, useless background information and low-value low temperature areas can be removed, and then the data set can be produced, which is beneficial to reducing training costs and improving fault diagnosis accuracy.

[0109] Wherein, the planetary gear fault diagnosis data set is divided into a training set, a validation set and a test set according to a certain ratio; in this embodiment, the planetary gear fault diagnosis data set is divided into a training set, a validation set and a test set according to a ratio of 8:1:1;

[0110] The training set and validation set are input into the VGG16 convolutional neural network for training. After multiple rounds of iterations, the weights of the rounds with the minimum training loss and the highest validation accuracy are saved to obtain a json weight file. The weight file and the test set are loaded into the test program to count the recognition accuracy of the four types of fault states, such as Figure 7 As shown, Figure 7 The 0 in the horizontal axis represents the pitting failure state. Figure 7 The 1 in the horizontal axis represents the broken tooth fault state. Figure 7 The 2 in the horizontal axis represents the glue failure state, and Figure 7 The 3 in the horizontal axis represents the plastic deformation fault state. Finally, the autonomous identification of different fault types of planetary gears is achieved.

[0111] Compared with the traditional vibration-based fault diagnosis technology, the above-mentioned fault diagnosis method for the planetary gear of the rocker arm of the coal mining machine not only has the advantages of non-contact, large detection area, high recognition rate, intuitiveness and reliability, but also can obtain additional visual temperature information that is easy for humans to understand. The k-means algorithm is used to extract the high-temperature area of ​​the planetary gear with higher value, so as to calculate the temperature data of the high-temperature area of ​​the planetary gear, and effectively utilize all the information reflected by the thermal imaging image, so that the calculation result has a high reference value. This application classifies the thermal image of the planetary gear high-temperature area after segmentation according to the fault type, removes useless background information and interference from the temperature information of the low-value low-temperature area, and uses the VGG16 convolutional neural network to extract features from the infrared thermal image marked with the high-temperature area of ​​the planetary gear, and finally realizes the autonomous recognition of different fault types of the planetary gear.

[0112] Real-time monitoring and accurate detection of the movement of the planetary gears can help optimize the working efficiency of the coal mining machine. Operators can adjust and optimize according to the movement of the planetary gears to improve the working efficiency and production capacity of the coal mining machine.

[0113] In one embodiment, the fault diagnosis method of the planetary gear of the rocker arm of the coal mining machine further includes:

[0114] The temperature flow of the planetary gears during the operation of the coal mining machine is collected in real time by the planetary gear temperature monitoring device installed on the planetary frame inside the rocker arm of the coal mining machine, and the full-screen temperature matrix of the infrared thermal image is obtained;

[0115] According to the corresponding functional relationship between the predetermined infrared thermal image and the full-screen temperature matrix of the infrared thermal image, the temperature value of each pixel point in the high-temperature area of ​​the refined planetary gear is determined;

[0116] A temperature cloud map is drawn according to the refined temperature values ​​of each pixel point in the high-temperature area of ​​the planetary gear to visualize the temperature information of the planetary gear during the operation of the coal mining machine.

[0117] The infrared thermal image full-screen temperature matrix is ​​a matrix that records the temperature value of each pixel point of the infrared thermal image of the planetary gear.

[0118] Among them, the temperature value of each point in the high-temperature area of ​​the planetary gear is calculated by using the corresponding functional relationship between the temperature value and the grayscale value (i.e., the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image), and the maximum temperature, minimum temperature and average temperature can be obtained but are not limited to, and a temperature cloud map is drawn based on the temperature information of the high-temperature area of ​​the planetary gear (i.e., the maximum temperature, minimum temperature and average temperature).

[0119] Specifically, firstly, the high temperature area of ​​the planetary gear on the infrared thermal image of the planetary gear is grayed, the grayed image is converted into an array form, the index of the non-zero element in the array is found, and the corresponding function relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image is calculated for the non-zero element to obtain the temperature value of the non-zero area, i.e., the high temperature area, which can be obtained but not limited to the maximum temperature, the minimum temperature and the temperature cloud map, such as Figure 8 As shown. The image obtained by the k-means algorithm after the high temperature area is refined is converted into an array form, the number of non-zero pixels is counted, the temperature values ​​of the non-zero area, i.e. the high temperature area, are accumulated and summed, and the sum is divided by the number of non-zero pixels to obtain the average temperature of the high temperature area. The highest temperature, the lowest temperature and the average temperature can all be compared with the data measured by the infrared sensor supporting software (such as Fig. 9 ) confirms that the average temperature in the high-temperature area should be slightly lower than the maximum temperature.

[0120] In one embodiment, the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image is determined as follows:

[0121] Through the planetary gear temperature monitoring device installed on the planetary frame inside the rocker arm of the coal mining machine, the temperature flow samples of the planetary gears in different states during the operation of the coal mining machine are collected, and the infrared thermal image full-screen temperature matrix samples corresponding to the infrared thermal image samples of the planetary gears are obtained; the denoised planetary gear thermal image samples are converted into grayscale image samples; all points of integer temperature values ​​are screened out in the infrared thermal image full-screen temperature matrix samples, and the coordinate information of the points of integer temperature values ​​is recorded to obtain a coordinate information set; the grayscale values ​​corresponding to each coordinate information in the coordinate information set are used to obtain all grayscale values ​​corresponding to each integer temperature value; the number of occurrences of each grayscale value in all grayscale values ​​corresponding to each integer temperature value is counted, and the grayscale value with the largest number of occurrences is used as the grayscale value corresponding to the integer temperature value, and each pair of integer temperature value-grayscale value fitting samples is determined; linear regression fitting is applied to each pair of integer temperature value-grayscale value fitting samples to obtain the corresponding functional relationship between the infrared thermal image and the infrared thermal image full-screen temperature matrix.

[0122] In one embodiment, the method for converting the denoised planetary gear thermal image samples into grayscale image samples is to calculate the values ​​of the three color channels in the color image by weighting, and the calculation formula is:

[0123] G gray (i,j)=0.299R(i,j)+0.587G(i,j)+0.114B(i,j)

[0124] Among them, R(i,j), G(i,j), and B(i,j) are the red channel pixel value, green channel pixel value, and blue channel pixel value of the (i,j) position respectively. gray(i,j) is the grayscale value after the conversion of position (i,j).

[0125] In one embodiment, the expression of the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image is:

[0126]

[0127] in, is the temperature value, G is the gray value, ω is the linear coefficient, and b is the constant term.

[0128] It should be understood that by collecting the infrared thermal image of the planetary gear and its full-screen temperature matrix of the infrared thermal image, the corresponding functional relationship between the pixel value in the infrared thermal image of the planetary gear and the full-screen temperature matrix of the thermal image is calculated, and the thermal image of the planetary gear is segmented, and the temperature information of the high-temperature area of ​​the planetary gear after segmentation is extracted using the corresponding functional relationship, and visualized. In this way, the temperature of the high-temperature area of ​​the planetary gear can be accurately grasped, and the effective analysis of the operating status of the planetary gear can be achieved.

[0129] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0130] In one embodiment, Figure 10-11 As shown, a planetary gear temperature monitoring device used in the fault diagnosis method of the planetary gear of the rocker arm of the coal mining machine, the planetary gear temperature monitoring device comprises: an infrared sensor 1, an outer conductive slip ring mover 2, an outer conductive slip ring stator 3, an inner conductive slip ring mover 4, an inner conductive slip ring stator 5 and a temperature flow receiving device;

[0131] An annular groove is opened on the surface of the planetary carrier 6 inside the rocker arm of the coal mining machine near the planetary gear and the sun gear, and the infrared sensor 1 is fixed in the annular groove, so that the infrared sensor 1 can monitor the synchronous movement of the planetary gear and the sun gear in real time and take temperature flow photos; the inner hole of the planetary carrier 6 is expanded according to the size of the outer conductive slip ring mover 2, so that there is a gap between the outer conductive slip ring mover 2 and the inner hole wall of the planetary carrier 6, and the infrared sensor 1 is electrically connected to one end of the outer conductive slip ring mover 2 through the cable 11, so that the outer conductive slip ring mover 2 rotates synchronously with the planetary gear and the planetary carrier 6; the outer conductive slip ring stator 3 is sleeved in the ring of the outer conductive slip ring mover 2, and the outer conductive slip ring stator 3 is fixed on the outer wall of the output shaft 7 of the rocker arm of the coal mining machine, so that the outer conductive slip ring stator 3 rotates synchronously with the output shaft 7; the inner conductive slip ring mover According to the size of the bottom end of the output shaft 7 of the rocker arm of the coal mining machine, a flange is extended into a cover-shaped structure and fixedly installed on the bottom end of the output shaft 7 of the rocker arm of the coal mining machine, a through hole is opened on the output shaft 7 near the installation of the outer conductive slip ring stator 3, and a cable 11 is passed through the through hole to electrically connect one end of the outer conductive slip ring stator 3 with one end of the inner conductive slip ring mover 4, and the inner conductive slip ring mover 4 rotates synchronously with the outer conductive slip ring stator 3 and the output shaft 7; the inner conductive slip ring stator 5 is sleeved in the inner conductive slip ring mover 4, and one end of the inner conductive slip ring stator 5 away from the inner conductive slip ring mover 4 is connected to the cable 11, and the temperature flow receiving device or the underground switch is connected through the cable 11. When the cable 11 is connected to the underground switch, the temperature flow receiving device is connected to the temperature flow receiving device network.

[0132] The outer conductive slip ring mover 2 and the outer conductive slip ring stator 3 form an outer conductive slip ring.

[0133] The inner conductive slip ring mover 4 and the inner conductive slip ring stator 5 constitute an inner conductive slip ring.

[0134] Among them, the infrared sensor 1 is arranged on the planetary carrier 6 inside the coal mining machine rocker arm. The infrared sensor 1 rotates synchronously with the planetary wheel and the planetary carrier 6, and can obtain infrared temperature data on the meshing tooth surface of the planetary wheel and the sun wheel inside the coal mining machine rocker arm in real time.

[0135] The planetary gear is mounted on the planetary gear bearing 8 .

[0136] The sun gear is mounted on the sun gear shaft 9 .

[0137] Among them, the inner conductive slip ring stator 5 is close to the end cover 10 inside the coal mining machine rocker arm, and the cable 11 connecting the inner conductive slip ring stator 5 passes through the end cover 10 (the cable 11 can pass through the end cover 10 by drilling a hole in the end cover 10) to connect to the temperature flow receiving equipment or the underground switch.

[0138] The cable also has a power supply function.

[0139] Among them, an annular groove is opened on the planetary carrier 6 inside the coal mining machine rocker arm to expand the installation space of the infrared sensor 1, and the inner hole of the planetary carrier 6 is expanded according to the size of the selected outer conductive slip ring, so that there is a gap between the outer conductive slip ring mover 2 and the inner hole wall of the planetary carrier 6.

[0140] A through hole is opened on the output shaft 7 near the outer conductive slip ring stator 3 , and the size of the through hole is determined according to the diameter of the cable 11 . The diameter of the through hole can be greater than or equal to the diameter of the cable 11 .

[0141] The cable 11 has sufficient tensile strength, and the cable 11 connecting the outer conductive slip ring mover 2 and the infrared sensor 1 can drag the outer conductive slip ring mover 2 to rotate.

[0142] The data collected by the infrared sensor 1 is transmitted to the underground switch through the transmission system, and the data is transmitted to the surface equipment (temperature flow receiving equipment) through the underground communication system. According to the data provided by the infrared sensor 1, the temperature value and distribution of the planetary gear tooth surface can be analyzed to accurately detect the health status of the gear. If the working efficiency of the coal mining machine needs to be adjusted or optimized, the operator can operate according to the test results.

[0143] Among them, infrared sensor 1: In order to adapt to the internal environment of the rocker arm of the coal mining machine, an infrared sensor 1 is independently developed. The infrared movement adopts the PICO384 infrared module, which is small in size and has excellent performance, with a resolution of 632×384 and a frame rate of up to 60Hz. According to the oily environment inside the rocker arm, the surface of the infrared sensor 1 is treated with oil repellent. The infrared sensor 1 is installed on the planetary carrier 6 to better obtain the infrared temperature image on the tooth surface of the planetary gear.

[0144] The above planetary gear temperature monitoring device uses the gap between the planetary gears to minimize the impact of the modification on the structure of the coal mining machine. At the same time, the synchronous movement of the planetary gear and the sun gear can be monitored in real time, and the infrared temperature of the meshing tooth surface can be photographed. Compared with other infrared sensors installed externally, the internal infrared sensor can provide more intuitive and more accurate data to ensure accurate detection of the working conditions of the internal gears of the rocker gearbox. Due to the reasonable design of the infrared sensor and the transmission system (the transmission system composed of the outer conductive slip ring mover, the outer conductive slip ring stator, the inner conductive slip ring mover and the inner conductive slip ring stator), it has high reliability. The data of the infrared sensor is output to the outer conductive slip ring mover on the outer wall of the shaft, and is output to the conductive slip ring mover on the inner wall of the output shaft through the external conductive slip ring stator. This design can realize the communication routing of the transmission line of the infrared sensor between the rotating parts with different speeds. The cable used in the transmission process is transmitted to the outside of the coal mining machine through the hollow part of the output shaft until it is transmitted to the temperature flow receiving device, which effectively avoids the problem that the transmission line is easily damaged or broken, and improves the stability and reliability of the system. Compared with wireless communication systems, the planetary gear temperature monitoring device has a larger communication data volume and faster rate, and there will be no data delay and packet loss. With line power supply, the infrared sensor does not need to consider the energy consumption part, and the developed infrared sensor has better performance. The structure of the planetary gear temperature monitoring device is relatively simple and easy to maintain. If the infrared sensor or part of the transmission system needs to be replaced or repaired, it can be operated relatively easily, reducing the difficulty of maintenance and maintenance.

[0145] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A fault diagnosis method for a planetary gear of a rocker arm of a coal mining machine, characterized in that: The method comprises: The temperature flow of the planetary gears during the operation of the coal mining machine is collected in real time by a planetary gear temperature monitoring device installed on the planetary carrier inside the rocker arm of the coal mining machine, and an infrared thermal image of the planetary gears is obtained; Performing image denoising processing on the infrared thermal image to obtain a denoised thermal image of the planetary gear; Segmenting the denoised planetary gear thermal image and performing edge thinning to obtain a high-temperature region thinned image; According to the refined image of the high temperature area, the high temperature area is marked on the infrared thermal image to obtain an infrared thermal image with the high temperature area of ​​the planetary gear marked; Inputting the infrared thermal image marked with the high temperature area of ​​the planetary gear into the fault diagnosis model trained based on the VGG16 convolutional neural network to perform fault diagnosis, and determining the fault diagnosis result of the planetary gear; The performing image denoising processing on the infrared thermal image to obtain a denoised thermal image of the planetary gear includes: The infrared thermal image is input into a bilateral filter for bilateral filtering denoising, and the edge information of the image is maintained while removing the noise, so as to obtain a denoised planetary gear thermal image, wherein the expression of the bilateral filtering is: in, is the updated pixel value of the center pixel of the box, W p is the normalization factor, is the spatial domain kernel, S is the filter window, is the image pixel domain kernel, I(p) is the value of the center pixel point p of the filter window, and I(q) is the value of the pixel point q of the infrared thermal image; The step of segmenting the denoised planetary gear thermal image and performing edge thinning to obtain a thinned image of a high temperature area includes: Using a k-means algorithm to classify pixels with similar features in the denoised planetary gear thermal image, segmenting a high temperature area and a low temperature area in the denoised planetary gear thermal image to obtain a segmented image; Performing an opening operation on the high temperature region in the segmented image to achieve edge thinning, thereby obtaining an image after the high temperature region is thinned; The method of marking the high temperature area on the infrared thermal image according to the image after the high temperature area is refined to obtain the infrared thermal image marked with the high temperature area of ​​the planetary gear includes: Binarizing the image after the high temperature area is refined and then converting it into data of the same data type as the infrared thermal image to obtain converted data; The converted data is multiplied by the infrared thermal image to achieve a masking effect on the high-temperature area of ​​the planetary gear in the infrared thermal image, so as to obtain an infrared thermal image marked with the high-temperature area of ​​the planetary gear.

2. The method according to claim 1, characterized in that The expression of the opening operation is: Where A is the high temperature area in the segmented image, B is the structural element, ⊙ is the corrosion operation on the high temperature area A in the segmented image, z is the pixel coordinate in the structural element, (B) z is the position of the structural element on the segmented image, To dilate the high temperature area A in the segmented image, Is an empty set.

3. The method according to claim 1, characterized in that: The training method of the fault diagnosis model is: Step 1, by means of a planetary gear temperature monitoring device installed on a planetary carrier inside a rocker arm of a coal mining machine, temperature flow samples of the planetary gears in different states during the operation of the coal mining machine are collected to obtain infrared thermal image samples of the planetary gears, wherein the different states include a normal state and various fault states; Step 2, performing image denoising processing on the infrared thermal image sample to obtain a denoised planetary gear thermal image sample; Step 3, segmenting the denoised planetary gear thermal image samples and performing edge thinning to obtain thinned planetary gear high temperature area samples; Step 4, marking the planetary gear high temperature area samples onto the infrared thermal image samples, obtaining infrared thermal image samples marked with the planetary gear high temperature areas, and obtaining a planetary gear fault diagnosis data set; Step 5, dividing the planetary gear fault diagnosis data set into a training set, a validation set and a test set according to a preset ratio; Step 6, inputting the training set and the validation set into the VGG16 convolutional neural network for training, after multiple rounds of iterations, until the iteration stop condition is met, determining the final network parameters, updating the network parameters of the VGG16 convolutional neural network to the final network parameters, and obtaining a preliminarily trained fault diagnosis model; Step 7, input the test set into the preliminarily trained fault diagnosis model to test the accuracy of the preliminarily trained fault diagnosis model. If the accuracy meets the requirement, the trained fault diagnosis model is obtained. If the accuracy does not meet the requirement, return to execute step 1 until the accuracy meets the requirement and the trained fault diagnosis model is obtained.

4. The method according to claim 3, characterized in that The method further comprises: The temperature flow of the planetary gears during the operation of the coal mining machine is collected in real time by the planetary gear temperature monitoring device installed on the planetary frame inside the rocker arm of the coal mining machine, and the full-screen temperature matrix of the infrared thermal image is obtained; Determine the temperature value of each pixel point in the high-temperature area of ​​the refined planetary gear according to a predetermined corresponding function relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image; A temperature cloud diagram is drawn according to the refined temperature values ​​of each pixel point in the high-temperature area of ​​the planetary gear, so as to visualize the temperature information of the planetary gear during the operation of the coal mining machine.

5. The method according to claim 4, characterized in that The corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image is determined as follows: The planetary gear temperature monitoring device installed on the planetary carrier inside the rocker arm of the coal mining machine is used to collect temperature flow samples of the planetary gears in different states during the operation of the coal mining machine, and obtain infrared thermal image full-screen temperature matrix samples corresponding to the infrared thermal image samples of the planetary gears; Converting the denoised planetary gear thermal image samples into grayscale image samples; Screening out all points with integer temperature values ​​in the full-screen temperature matrix sample of the infrared thermal image, and recording coordinate information of the points with integer temperature values ​​to obtain a coordinate information set; Using the grayscale value corresponding to each coordinate information in the coordinate information set, all grayscale values ​​corresponding to each integer temperature value are obtained; Count the number of occurrences of each grayscale value among all grayscale values ​​corresponding to each integer temperature value, take the grayscale value with the largest number of occurrences as the grayscale value corresponding to the integer temperature value, and determine each pair of integer temperature value-grayscale value fitting samples; Linear regression fitting is applied to each pair of integer temperature value-grayscale value fitting samples to obtain the corresponding functional relationship between the infrared thermal image and the full-screen temperature matrix of the infrared thermal image.

6. A planetary gear temperature monitoring device in a planetary gear fault diagnosis method for a rocker arm of a coal mining machine according to any one of claims 1 to 5, characterized in that: The planetary gear temperature monitoring device comprises: an infrared sensor, an outer conductive slip ring mover, an outer conductive slip ring stator, an inner conductive slip ring mover, an inner conductive slip ring stator and a temperature flow receiving device; An annular groove is formed on the surface of the planetary frame inside the rocker arm of the coal mining machine near the planetary gear and the sun gear, and the infrared sensor is fixed in the annular groove, so that the infrared sensor monitors the synchronous movement of the planetary gear and the sun gear in real time and takes temperature flow pictures; According to the size of the outer conductive slip ring mover, the inner hole of the planet carrier is expanded so that there is a gap between the outer conductive slip ring mover and the inner hole wall of the planet carrier, and the infrared sensor is electrically connected to one end of the outer conductive slip ring mover through a cable, so that the outer conductive slip ring mover rotates synchronously with the planetary gear and the planet carrier; The outer conductive slip ring stator is sleeved in the ring of the outer conductive slip ring mover, and the outer conductive slip ring stator is fixed on the outer wall of the output shaft of the coal mining machine rocker arm, so that the outer conductive slip ring stator rotates synchronously with the output shaft; The inner conductive slip ring mover is fixedly installed at the bottom end of the output shaft of the coal mining machine rocker arm in a flange-like structure extending into a cover-like structure according to the size of the bottom end of the output shaft of the coal mining machine rocker arm, and a through hole is opened on the output shaft near the installation of the outer conductive slip ring stator, so that a cable passes through the through hole to electrically connect one end of the outer conductive slip ring stator with one end of the inner conductive slip ring mover, and the inner conductive slip ring mover rotates synchronously with the outer conductive slip ring stator and the output shaft; The inner conductive slip ring stator is sleeved in the inner conductive slip ring mover, and one end of the inner conductive slip ring stator away from the inner conductive slip ring mover is connected to a cable, through which the temperature flow receiving device or the downhole switch is connected. When the cable is connected to the downhole switch, the temperature flow receiving device is connected to the temperature flow receiving device network.

Images