A method and system for diagnosing oil seal leakage fault of a mining reducer

Abstract

The present invention relates to the technical field of sealing test, and in particular to a method and system for diagnosing oil seal leakage faults for mining reducers. The method jointly analyzes multidimensional state parameters in a detection time period, divides the detection time period into multiple sub-time periods, determines the temperature rise amount according to the data change characteristics in the temperature dimension, and screens out the suspected abnormal sub-time periods, and further obtains the degree of underlubrication to determine whether an abnormality has occurred in the detection time period. The temperature-sound synchronization is obtained, and the anomalies in the speed dimension are statistically analyzed according to the data outliers, and finally the change characteristics of the three state parameter dimensions are combined to determine whether the oil seal is leaking. The present invention conducts a joint analysis of the three state parameter dimensions through time series analysis combined with the obvious characteristics generated when the oil seal leaks, and accurately and effectively diagnoses the oil seal leakage fault of the mining reducer.

Description

Technical Field

[0001] The invention relates to the technical field of sealing test, and in particular to a method and system for diagnosing leakage fault of an oil seal of a mining reducer. Background Art

[0002] Mining reducers usually refer to reducers used in mining equipment. This type of reducer is mostly used in heavy equipment such as mining machinery, conveying equipment, mining hoists, etc., and undertakes the functions of deceleration, torque increase and power transmission. Mining equipment usually faces high loads, heavy loads and complex working environments during operation, so high reliability and stability requirements are placed on the reducer. Oil seals are a sealing device that is usually installed on the bearing part or connection of the reducer to prevent the leakage of lubricating oil or grease and to prevent external dust, impurities and moisture from entering the reducer. Oil seals are generally made of rubber or polymer materials and have good elasticity and wear resistance. Its structure usually includes lip parts, spring reinforcement rings, sealing rings and other components, which can fit tightly between the shaft and the bearing to form an effective seal.

[0003] During the operation of the reducer, it is necessary to detect the oil seal leakage status in real time. The existing technology can analyze the reducer temperature, noise decibels generated by friction and other data, and set thresholds to feedback leakage warning signals. However, for the reducer, the data changes caused by these state parameters are not only caused by oil seal leakage. Aging of the reducer or wear of internal components will cause the reducer to have no obvious lubrication effect, which will also show abnormal values ​​of state parameters such as the temperature of the reducer. This situation does not belong to oil seal leakage, but normal wear, aging or the influence of force majeure of the environment. The existing technology only compares the state parameters with the threshold, which is easy to cause misjudgment of oil seal leakage. Summary of the invention

[0004] In order to solve the technical problem that the prior art only compares the state parameters with the threshold value, which easily causes the misjudgment of oil seal leakage, the purpose of the present invention is to provide a method and system for diagnosing the oil seal leakage fault of a mining reducer. The technical scheme adopted is as follows:

[0005] The present invention proposes a method for diagnosing a leakage fault of an oil seal of a mining reducer, the method comprising:

[0006] Acquire multi-dimensional state parameters of the reducer in real time during a preset detection time period, wherein the state parameters include: temperature, sound intensity, and rotation speed;

[0007] The detection time period is divided into multiple sub-time periods in terms of time series; in each sub-time period, in the temperature dimension, the temperature rise amount of each sub-time period is obtained according to the change amplitude and change trend of the data in the temperature sequence; and the suspected abnormal sub-time period is screened out according to the temperature rise amount;

[0008] Count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period to obtain the degree of underlubrication in each state parameter dimension; determine whether the current detection time period is abnormal based on the degree of underlubrication in all state parameter dimensions;

[0009] If an abnormality occurs in the current detection time period, the temperature-sound synchronization is obtained based on the difference in the degree of delubrication between the temperature dimension and the sound intensity dimension, as well as the difference in the change amplitude in the same sub-time period; the outliers of the data in the speed dimension are statistically analyzed, and whether the oil seal of the mining reducer is leaking is determined based on the outliers and the temperature-sound synchronization.

[0010] Furthermore, the method for obtaining the temperature rise includes:

[0011] The overall change amplitude is obtained according to the data difference between the first element and the last element in the temperature sequence in the sub-time period; the overall change trend is obtained according to the difference between adjacent elements in the temperature sequence; and the temperature rise is obtained according to the overall change amplitude and the overall change trend.

[0012] Furthermore, the method for obtaining the overall change amplitude includes:

[0013] The element difference between the last element and the first element is obtained, and the ratio of the element difference to the number of elements in the temperature sequence is used as the overall change amplitude.

[0014] Furthermore, the method for obtaining the overall change trend includes:

[0015] For each pair of adjacent elements in the temperature sequence, the difference between the latter element and the former element is calculated to obtain a difference sequence, and the accumulated value of the elements in the difference sequence is used as the overall change trend.

[0016] Furthermore, the method for obtaining the degree of underlubrication includes:

[0017] For the temperature and sound intensity dimensions, the overall change amplitude of the data in each suspected abnormal sub-time period is obtained, and the overall change amplitudes in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the two state parameter dimensions;

[0018] For the speed dimension, the overall change amplitude of the data in each suspected abnormal sub-time period is negatively correlated to obtain the speed reduction degree. The speed reduction degrees in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the speed dimension.

[0019] Further, judging whether an abnormality occurs in the current detection time period according to the underlubrication degree under all state parameter dimensions includes:

[0020] After normalizing the underlubrication degree in each state parameter dimension, the average underlubrication degree in all state parameter dimensions is calculated. If the average underlubrication degree is greater than a preset underlubrication degree threshold, it is determined that an abnormality has occurred in the current detection time period.

[0021] Furthermore, the method for obtaining the temperature-sound synchronization includes:

[0022] Obtaining the difference in underlubrication between the temperature dimension and the sound intensity dimension;

[0023] Obtain the sound intensity rise in each sub-time period in the sound intensity dimension according to the temperature rise acquisition method; obtain the difference between the sound intensity rise and the temperature rise in the same sub-time period; and calculate the average rise difference in all sub-time periods;

[0024] The temperature-sound synchronization is obtained by multiplying the difference in underlubrication and the difference in average rise, performing negative correlation mapping and normalizing.

[0025] Furthermore, the method for obtaining the outlier includes:

[0026] The outlier factor of each element in the speed sequence in the detection time period is obtained by the outlier algorithm, the average outlier factor in the sub-time period is used as the overall outlier factor in the sub-time period, and the difference between the average overall outlier factor of all sub-time periods and the preset outlier threshold is used as the outlier.

[0027] Further, judging whether the oil seal of a mining reducer is leaking according to the outlier property and the temperature-sound synchronization includes:

[0028] The product of the outlier and the temperature-sound synchronization is used as the oil seal leakage degree. If the oil seal leakage degree is greater than a preset leakage degree threshold, it is determined that the mining reducer oil seal is leaking.

[0029] The present invention also proposes a mining reducer oil seal leakage fault diagnosis system, the system comprising:

[0030] A data acquisition unit, used to obtain multi-dimensional state parameters of the reducer in real time during a preset detection time period, wherein the state parameters include: temperature, sound intensity, and rotation speed;

[0031] A first abnormality analysis unit is used to divide the detection time period into multiple sub-time periods in time series; in each sub-time period, in the temperature dimension, according to the change amplitude and change trend of the data in the temperature sequence, obtain the temperature rise amount of each sub-time period; according to the temperature rise amount, select the suspected abnormal sub-time period;

[0032] The second abnormality analysis unit is used to count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period, and obtain the degree of underlubrication in each state parameter dimension; according to the degree of underlubrication in all state parameter dimensions, determine whether the current detection time period is abnormal;

[0033] The oil seal leakage judgment unit is used to obtain the temperature-sound synchronization according to the difference in the degree of delubrication between the temperature dimension and the sound intensity dimension, and the difference in the change amplitude in the same sub-time period if an abnormality occurs in the current detection time period; the outlier of the data in the speed dimension is statistically analyzed, and whether the oil seal of the mining reducer is leaking is judged according to the outlier and the temperature-sound synchronization.

[0034] The present invention has the following beneficial effects:

[0035] The present invention takes into account that the change of state parameters when the reducer oil seal leaks should be a continuous process, so the multi-dimensional state parameters are jointly analyzed in a detection time period. In order to effectively analyze the change of state parameters in time series, the embodiment of the present invention divides the detection time period into multiple sub-time periods. Considering that the temperature parameter has a more intuitive performance for leakage anomalies, the temperature rise amount is first determined according to the data change characteristics in the temperature dimension and the suspected abnormal sub-time period is screened out. The more suspected abnormal sub-time periods, the more likely the oil seal leakage anomaly is in the detection time period, so the underlubrication degree is further obtained to determine whether the detection time period has an abnormality. Further considering that the state abnormality in the detection time period may not be caused by oil seal leakage, the abnormality caused by oil seal leakage will cause the sound and speed to be abnormal at the same time, so the embodiment of the present invention further obtains temperature-sound synchronization, and according to the data outlier statistics of the abnormality in the speed dimension, finally combines the change characteristics of the three state parameter dimensions to determine whether the oil seal is leaking. The present invention analyzes the three state parameter dimensions through time series analysis, combined with the obvious characteristics generated when the oil seal leaks, and accurately and effectively diagnoses the oil seal leakage fault of the mining reducer. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 A flow chart of a method for diagnosing oil seal leakage faults of a mining reducer provided by one embodiment of the present invention;

[0038] Figure 2 A module structure diagram of a mining reducer oil seal leakage fault diagnosis device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method and system for diagnosing oil seal leakage faults of a mining reducer proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The specific scheme of a method and system for diagnosing oil seal leakage fault of a mining reducer provided by the present invention is described in detail below in conjunction with the accompanying drawings.

[0042] The embodiment of the present invention firstly proposes a method for diagnosing oil seal leakage fault of a mining reducer. Figure 1 , which shows a flow chart of a method for diagnosing oil seal leakage fault of a mining reducer provided by an embodiment of the present invention, the method comprising:

[0043] Step S1: acquiring multi-dimensional state parameters of the reducer in real time during a preset detection time period, wherein the state parameters include temperature, sound intensity and rotation speed.

[0044] If the oil seal of a mining reducer leaks, it will lead to insufficient lubrication, and then produce abnormal high temperature, noise and speed. These three state parameters are more intuitive for oil seal leakage. Therefore, the embodiment of the present invention uses three state parameters of temperature, sound intensity and speed to analyze the working state of the reducer. The state parameters of the reducer in multiple dimensions are obtained in real time during the preset detection time period. It should be noted that because the data is obtained in real time in a time series, the data obtained for each state parameter during the detection time period is in the form of a sequence, and a set of data can be expressed as a data matrix formed by three sequences.

[0045] In one embodiment of the present invention, considering the influence of noise in the data collection and transmission process, and the different dimensions between different state parameters, it is necessary to perform denoising and de-dimensionalization processing on the data after obtaining it. In the embodiment of the present invention, only the numerical value of each state parameter is retained, and then the sequence composed of the numerical values ​​is cleaned using a moving average algorithm to achieve denoising and de-dimensionalization processing of the data, and eliminate local changes caused by noise.

[0046] It should be noted that, because the state parameters are collected in real time in the embodiment of the present invention, subsequent operations need to be performed after the collection is completed in the entire detection time period. If detection is still ongoing during the detection time period, it is considered that the amount of data is insufficient for subsequent algorithm processing.

[0047] Step S2: Divide the detection time period into multiple sub-time periods in time series; in each sub-time period, in the temperature dimension, obtain the temperature rise of each sub-time period according to the change amplitude and change trend of the data in the temperature sequence; and screen out suspected abnormal sub-time periods according to the temperature rise.

[0048] If the reducer oil seal leaks, lubrication fails, and friction increases, the temperature rise of the reducer will be most obvious. Therefore, the embodiment of the present invention obtains the temperature rise. The higher the temperature rise, the higher the abnormality of the components in the mining reducer. The more likely it is that the reducer abnormality is due to oil seal leakage, insufficient lubricating oil, or aging of components. Considering that the temperature data is continuous time series data within the detection time period, in order to avoid the loss of local information caused by the overall analysis of data changes within the detection time period, the embodiment of the present invention first divides the detection time period into multiple sub-time periods, that is, analyzes the data features within the local time series range in each sub-time period to avoid the loss of local information.

[0049] In each sub-time period, the greater the change amplitude of the data in the temperature sequence and the stronger the change trend, the greater the temperature rise in the sub-time period. Therefore, the embodiment of the present invention can obtain the temperature rise in each sub-time period according to the change amplitude and change trend of the data in the temperature sequence. The greater the temperature rise, the more likely it is that the reducer lubrication abnormality has occurred in the sub-time period.

[0050] Preferably, in an embodiment of the present invention, the method for obtaining the temperature rise amount includes:

[0051] The overall change amplitude is obtained based on the data difference between the first element and the last element in the temperature sequence in the sub-time period. The data difference between the first and last elements represents the overall data difference in the time series range of the sub-time period. Based on this, the overall change amplitude can be obtained to characterize the data change of the entire sub-time period. Similarly, the overall change trend is obtained based on the difference between adjacent elements in the temperature sequence.

[0052] It should be noted that the overall characteristics represented here are the overall characteristics within the sub-time period, and the overall characteristics of different sub-time periods respectively represent the characteristics of each local time series range of the detection time period. The temperature rise of each sub-time period is obtained according to the overall change amplitude and the overall change trend. In the embodiment of the present invention, the product of the overall change amplitude and the overall change trend is used as the temperature rise.

[0053] Furthermore, in one embodiment of the present invention, the method for obtaining the overall change amplitude includes: obtaining the element difference between the last element and the first element, and taking the ratio of the element difference to the number of elements in the temperature sequence as the overall change amplitude. The purpose of dividing the element difference by the number of elements is to characterize the overall change amplitude in the entire sub-time period based on the principle of averaging. The larger the positive tangent of the element difference is, the greater the temperature rise change has occurred in the sub-time period, the greater the overall change amplitude, and the corresponding temperature rise should also be greater.

[0054] Furthermore, in one embodiment of the present invention, the method for obtaining the overall change trend includes:

[0055] For each pair of adjacent elements in the temperature sequence, the difference between the latter element and the previous element is calculated to obtain the difference sequence, and the accumulated value of the elements in the difference sequence is used as the overall change trend. The more positive the elements in the difference sequence are and the larger the value is, the greater the trend of temperature rise is. The larger the accumulated value is, the greater the temperature is in an obvious upward trend throughout the sub-time period. The greater the overall change trend is, the greater the corresponding temperature rise is.

[0056] After obtaining the temperature rise, the suspected abnormal sub-time period can be screened out. It should be noted that because the sub-time period is a local time series range in the detection time period, normal state parameter fluctuations may occur in the detection time period, and this fluctuation will not last for a long time. Therefore, the embodiment of the present invention first screens out the suspected abnormal sub-time period, and then confirms whether the detection time period has obvious abnormalities based on the characteristics of the suspected abnormal sub-time period in subsequent steps. The larger the temperature rise, the more likely it is a suspected abnormal sub-time period. In the embodiment of the present invention, after the temperature rise is normalized, the rise threshold is set to 0.8. If the temperature rise is greater than the rise threshold, the sub-time period is identified as a suspected abnormal sub-time period.

[0057] Step S3: Count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period to obtain the degree of underlubrication in each state parameter dimension; determine whether the current detection time period is abnormal based on the degree of underlubrication in all state parameter dimensions.

[0058] For the detection time period, the more suspected abnormal sub-time periods there are in the entire time series range, it means that the abnormal fluctuations in the detection time period are not normal fluctuations, but abnormalities caused by problems such as component aging or oil seal leakage. Similarly, if the change amplitude of each state parameter temperature in the suspected abnormal sub-time period has a large change, it means that the detection time period is more likely to have abnormalities caused by problems such as component aging or oil seal leakage. Therefore, the embodiment of the present invention counts the number of suspected abnormal sub-time periods and the change amplitude of the data in each state parameter dimension under the suspected abnormal sub-time period to obtain the degree of underlubrication in each state parameter dimension. It should be noted that lubrication anomalies reflect the characteristics of temperature increase, sound intensity increase and speed decrease. Therefore, when calculating the degree of underlubrication, the degree of underlubrication in each state parameter dimension can be obtained based on this correlation. According to the degree of underlubrication in all state parameter dimensions, it can be jointly judged whether the current detection time period has an abnormality.

[0059] Preferably, in one embodiment of the present invention, considering that the overall change amplitude of the suspected abnormal sub-time period can be obtained according to the change amplitude of the data, but the overall change amplitude represents a rising amplitude, the underlubrication degree of the temperature and sound intensity dimensions can be obtained by the same method, while the rotation speed dimension needs to be obtained after a certain negative correlation mapping, specifically including:

[0060] For the temperature and sound intensity dimensions, the overall change amplitude of the data in each suspected abnormal sub-time period is obtained, and the overall change amplitudes in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the two state parameter dimensions.

[0061] For the speed dimension, the overall change amplitude of the data in each suspected abnormal sub-time period is negatively correlated to obtain the speed reduction degree. The speed reduction degrees in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the speed dimension.

[0062] It should be noted that the negative correlation mapping in the embodiment of the present invention adopts the inverse form, and the inverse of the data after adding the positive integer 1 is used as the result of the negative correlation mapping. The purpose of adding the positive integer 1 is to prevent the denominator from being zero.

[0063] Preferably, in one embodiment of the present invention, judging whether an abnormality occurs in the current detection time period according to the degree of underlubrication in all state parameter dimensions includes:

[0064] Because the data range under each state parameter dimension may be different, it is necessary to normalize the degree of underlubrication under each state parameter dimension separately. Then calculate the average degree of underlubrication under all state parameter dimensions. If the average degree of underlubrication is greater than the preset underlubrication threshold, it is determined that an abnormality has occurred in the current detection time period. It should be noted that the normalization method used in the embodiment of the present invention can use range normalization or function mapping method, which are technical means well known to those skilled in the art and will not be described in detail here. In the embodiment of the present invention, the underlubrication threshold is set to 0.9.

[0065] Step S4: If an abnormality occurs in the current detection time period, the temperature-sound synchronization is obtained based on the difference in the degree of delubrication between the temperature dimension and the sound intensity dimension, and the difference in the change amplitude in the same sub-time period; the outliers of the data in the speed dimension are statistically analyzed, and whether the oil seal of the mining reducer is leaking is determined based on the outliers and the temperature-sound synchronization.

[0066] Although step S3 can determine whether there is an abnormality in the detection time period, it still cannot determine whether the abnormality is caused by oil seal leakage. The embodiment of the present invention aims to accurately identify the abnormality caused by oil seal leakage, so further analysis is required after confirming that the abnormality occurs in the detection time period.

[0067] The abnormality caused by oil seal leakage is mainly manifested as a sharp increase or cumulative increase in the local temperature value of the reducer, and in this process, it will be accompanied by an increase in noise, because the lack of lubricating oil at this time leads to an increase in operating friction, resulting in temperature rise and noise. If the abnormality is caused by aging of reducer components, the two state parameters will produce abnormalities, but their synchronization is not high; for the abnormality caused by insufficient lubricating oil due to oil seal leakage, the two state parameters will produce abnormalities at the same time, and because it is a lubricating oil leakage, compared with the normal time period, the friction is increased and the power is fixed, and the speed value will drop sharply when the abnormality occurs. Therefore, the embodiment of the present invention first obtains the temperature-sound synchronization based on the difference in the degree of underlubrication between the temperature dimension and the sound intensity dimension, and the difference in the change amplitude in the same sub-time period. That is, the smaller the two differences, the higher the synchronization of the two dimensions of temperature and sound intensity, and the higher the temperature-sound synchronization, the more likely it is that the abnormality is caused by oil seal leakage. Further statistically analyze the outliers of the data under the speed dimension. The greater the outliers, the more likely it is that the speed has undergone a sudden and rapid change, and the more likely it is that the abnormality is caused by oil seal leakage. Therefore, the outlier property and temperature-sound synchronization can be combined to determine whether the oil seal of the mining reducer is leaking.

[0068] Preferably, in one embodiment of the present invention, the method for obtaining temperature-sound synchronization includes:

[0069] The difference in the degree of underlubrication between the temperature dimension and the sound intensity dimension is obtained.

[0070] The sound intensity rise in each sub-time period is obtained according to the method for obtaining the temperature rise; the difference between the sound intensity rise and the temperature rise in the same sub-time period is obtained; and the average rise difference in all sub-time periods is calculated.

[0071] The smaller the difference in underlubrication and the smaller the difference in average rise, the greater the synchronization of the two dimensions. Therefore, the underlubrication difference and the average rise difference are multiplied and then negatively correlated and normalized to obtain temperature-sound synchronization. In an embodiment of the present invention, the method of negatively correlated mapping and normalization is to use the opposite number of the data as the power of an exponential function with a natural constant as the base, and the function output result is the result after negatively correlated mapping and normalization.

[0072] Preferably, in one embodiment of the present invention, the method for obtaining outliers includes:

[0073] The outlier factor of each element in the speed sequence in the detection time period is obtained by the outlier algorithm. The outlier algorithm in the embodiment of the present invention can select a density-based outlier detection algorithm, and those skilled in the art can select other outlier algorithms to calculate the outlier factor, which will not be described in detail here.

[0074] The average outlier factor in the sub-time period is used as the overall outlier factor in the sub-time period, and the difference between the average overall outlier factor of all sub-time periods and the preset outlier threshold is used as the outlier. The average overall outlier factor characterizes the data characteristics in the entire detection time period. The larger it is relative to the preset outlier threshold, the more obvious the change in speed has occurred in the detection time period. In the embodiment of the present invention, the outlier factor is normalized data, so the average overall outlier factor is also data with a value range between 0 and 1, and the outlier threshold is set to 0.7.

[0075] Preferably, in one embodiment of the present invention, judging whether the oil seal of a mining reducer is leaking according to the outlier property and the temperature-sound synchronization includes:

[0076] The product of outlier and temperature-sound synchronization is taken as the oil seal leakage. If the oil seal leakage is greater than the preset leakage threshold, it is judged that the oil seal of the mining reducer is leaking. Otherwise, the abnormality in the detection time period is caused by other reasons. In the embodiment of the present invention, after the oil seal leakage is normalized, the leakage threshold is set to 0.7. It should be noted that after judging that the reducer has an oil seal leakage in the detection time period, a warning signal can be fed back to the staff; if it is judged that the abnormality is not caused by oil seal leakage, it is also necessary to feed back an abnormal warning signal to remind the staff to investigate the cause of the abnormality.

[0077] Based on the same inventive concept, the present invention also proposes a mining reducer oil seal leakage fault diagnosis system, the system comprising:

[0078] The data acquisition unit is used to obtain multi-dimensional state parameters of the reducer in real time during a preset detection time period. The state parameters include: temperature, sound intensity and rotation speed.

[0079] The first abnormality analysis unit is used to divide the detection time period into multiple sub-time periods in time series. In each sub-time period, in the temperature dimension, the temperature rise of each sub-time period is obtained according to the change amplitude and change trend of the data in the temperature sequence. The suspected abnormal sub-time period is screened out according to the temperature rise.

[0080] The second abnormality analysis unit is used to count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period to obtain the degree of underlubrication in each state parameter dimension. According to the degree of underlubrication in all state parameter dimensions, it is judged whether the current detection time period is abnormal.

[0081] The oil seal leakage judgment unit is used to obtain the temperature-sound synchronization according to the difference in the degree of underlubrication between the temperature dimension and the sound intensity dimension, and the difference in the change amplitude in the same sub-time period if an abnormality occurs in the current detection time period. The outliers of the data in the speed dimension are counted, and whether the oil seal of the mining reducer is leaking is judged based on the outliers and temperature-sound synchronization.

[0082] The present invention also proposes a mining reducer oil seal leakage fault diagnosis device, please refer to Figure 2 , which shows a module structure diagram of a mining reducer oil seal leakage fault diagnosis device provided by an embodiment of the present invention, the device is composed of 5 modules:

[0083] (1) Module 001: This module is a temperature detection sensor. The sensor model is a PT100 platinum resistance temperature sensor, which is suitable for high-precision temperature measurement and is used to monitor the temperature data of the reducer. Its temperature measurement range is -50 degrees to 250 degrees, with an accuracy of 0.5 degrees. The installation location can be selected near the key heat source area of ​​the reducer housing, the bearing seat, the gear box, etc., and fixed with high-temperature resistant glue.

[0084] (2) Module 002: This module is an airborne acoustic conduction decibel sensor. The sensor model is the GY-MAX4466 microphone module. The detection range is 30 decibels to 120 decibels, and the frequency response range is 20 Hz to 20,000 Hz. The installation location is outside the reducer housing and away from the mechanical vibration source to avoid interference from background noise, and is equipped with an acoustic shielding cover to avoid interference from the external environment.

[0085] (3) Module 003: This module is a speed sensor. The sensor model is A3144 Hall Effect Speed ​​Sensor. The speed measurement range is 0 to 10,000 RPM, and the output is a digital pulse signal. It is installed on the shaft end of the reducer main shaft or related rotating parts, and the speed change is measured by a magnetic marker. A gap of 2-3 mm is maintained between the magnetic marker and the sensor to ensure signal stability.

[0086] (4) Module 004: This module is a single-chip microcomputer, model STM32F103C8T6, which has good computing performance and low power consumption. The single-chip microcomputer also includes a communication module, which is composed of an HC-05 Bluetooth module that supports the Bluetooth 2.0 protocol and has a communication range of 10 meters. In the single-chip microcomputer, the analog signal output by the temperature sensor is collected through the ADC module, and the digital signals of the air acoustic conduction decibel sensor and the speed sensor are connected through the GPIO port. The DMA mode is configured to improve data transmission efficiency and avoid CPU blocking. The baud rate of Bluetooth communication is configured to be 115200, and UART communication is used for data transmission. The check bit is 1, the stop bit is 1, and there is no parity check. The FIFO cache is configured in the single-chip microcomputer to reserve temporary storage capacity when data overflows. In addition, the HC-05 module needs to be set to slave mode through AT commands, and an independent Bluetooth name and pairing password are set to ensure data independence and security.

[0087] (5) Module 005: This module is a data analysis and fault diagnosis module, which is implemented through the single-chip microcomputer in module 004. The signals of the above-mentioned status parameters are read through the specific interface of the single-chip microcomputer. The single-chip microcomputer integrates the data in real time and packages them according to the timestamp to form a data packet. When transmitting data, a standard data format, JSON or CSV format, is used for easy parsing; each frame of data contains a timestamp, temperature value, decibel value, speed value, and device ID. A double buffer structure is involved on the single-chip microcomputer. When one buffer is full, it immediately switches to another buffer and starts data transmission at the same time to avoid data loss. The algorithm program included in the above-mentioned method for diagnosing the oil seal leakage fault of a mining reducer is integrated in the single-chip microcomputer, and the judgment result of the oil seal leakage is obtained by running the program.

[0088] Among them, the above-mentioned sensor and the single-chip computer module are connected by a high-temperature resistant and shielded cable. The sensor interface area needs a dowry dust cover and a waterproof seal ring, and the composite IP67 level protection. It should be noted that there are two buffer storage areas in the single-chip computer, and the storage area size is 64Kb. One data transmission and reception package contains time, temperature, decibel and speed value, where the time is clk clock data, and the others are int type data. One data occupies four bytes, so the amount of data occupied is 16bit. The collection, transmission and cache of n times of data are marked as a group. One group of data participates in the analysis of the possibility of oil seal leakage, that is, the group is the data collected in the detection time period, and the byte storage capacity that needs to be cached is 16×n bytes. In the embodiment of the present invention, the acquisition frequency of the above-mentioned module is set to twice per second, the amount of data in one second is 32bit, and the detection time period is set to 10 minutes. The amount of data for each group of data is 19200bit, occupying 18.75Kb, which is less than the storage size of each buffer, and can participate in caching and calculation normally.

[0089] It should be noted that the algorithm program included in the fault diagnosis method of oil seal leakage of a mining reducer can be written and burned into the core of the single-chip microcomputer, including but not limited to outlier algorithm, data denoising, etc. Finally, the obtained oil warning signal can be transmitted to the computer via Bluetooth protocol for easy viewing by the staff.

[0090] In summary, the embodiment of the present invention jointly analyzes multi-dimensional state parameters in a detection time period, divides the detection time period into multiple sub-time periods, determines the temperature rise according to the data change characteristics in the temperature dimension and screens out the suspected abnormal sub-time periods, and further obtains the degree of underlubrication to determine whether an abnormality has occurred in the detection time period. The temperature-sound synchronization is obtained, and the anomalies in the speed dimension are statistically analyzed based on the data outliers, and finally the change characteristics of the three state parameter dimensions are combined to determine whether the oil seal is leaking. The present invention conducts a joint analysis of the three state parameter dimensions through time series analysis combined with the obvious characteristics generated when the oil seal leaks, and accurately and effectively diagnoses the oil seal leakage fault of the mining reducer.

[0091] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0092] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for diagnosing oil seal leakage fault of a mining reducer, characterized in that: The method comprises: Acquire multi-dimensional state parameters of the reducer in real time during a preset detection time period, wherein the state parameters include: temperature, sound intensity, and rotation speed; The detection time period is divided into multiple sub-time periods in terms of time series; in each sub-time period, in the temperature dimension, the temperature rise amount of each sub-time period is obtained according to the change amplitude and change trend of the data in the temperature sequence; and the suspected abnormal sub-time period is screened out according to the temperature rise amount; Count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period to obtain the degree of underlubrication in each state parameter dimension; determine whether the current detection time period is abnormal based on the degree of underlubrication in all state parameter dimensions; If an abnormality occurs in the current detection time period, the temperature-sound synchronization is obtained according to the difference in the degree of delubrication between the temperature dimension and the sound intensity dimension, as well as the difference in the amplitude of change in the same sub-time period; the outlier of the data in the speed dimension is statistically analyzed, and the product of the outlier and the temperature-sound synchronization is used as the oil seal leakage degree. If the oil seal leakage degree is greater than the preset leakage degree threshold, it is determined that the oil seal of the mining reducer is leaking; The method for obtaining the degree of underlubrication comprises: For the temperature and sound intensity dimensions, the overall change amplitude of the data in each suspected abnormal sub-time period is obtained, and the overall change amplitudes in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the two state parameter dimensions; For the speed dimension, the overall change amplitude of the data in each suspected abnormal sub-time period is negatively correlated to obtain the speed reduction degree. The speed reduction degrees in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the speed dimension.

2. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 1, characterized in that: The method for obtaining the temperature rise comprises: The overall change amplitude is obtained according to the data difference between the first element and the last element in the temperature sequence in the sub-time period; the overall change trend is obtained according to the difference between adjacent elements in the temperature sequence; and the temperature rise is obtained according to the overall change amplitude and the overall change trend.

3. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 2, characterized in that: The method for obtaining the overall change amplitude includes: The element difference between the last element and the first element is obtained, and the ratio of the element difference to the number of elements in the temperature sequence is used as the overall change amplitude.

4. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 2, characterized in that: The method for obtaining the overall change trend includes: For each pair of adjacent elements in the temperature sequence, the difference between the latter element and the former element is calculated to obtain a difference sequence, and the accumulated value of the elements in the difference sequence is used as the overall change trend.

5. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 1, characterized in that: The determining whether an abnormality occurs in the current detection time period according to the underlubrication degree under all state parameter dimensions includes: After normalizing the underlubrication degree in each state parameter dimension, the average underlubrication degree in all state parameter dimensions is calculated. If the average underlubrication degree is greater than a preset underlubrication degree threshold, it is determined that an abnormality has occurred in the current detection time period.

6. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 1, characterized in that: The method for obtaining the temperature-sound synchronization comprises: Obtaining the difference in underlubrication between the temperature dimension and the sound intensity dimension; Obtain the sound intensity rise in each sub-time period in the sound intensity dimension according to the temperature rise acquisition method; obtain the difference between the sound intensity rise and the temperature rise in the same sub-time period; and calculate the average rise difference in all sub-time periods; The temperature-sound synchronization is obtained by multiplying the difference in underlubrication and the difference in average rise, performing negative correlation mapping and normalizing.

7. A method for diagnosing oil seal leakage fault of a mining reducer according to claim 1, characterized in that: The method for obtaining the outlier includes: The outlier factor of each element in the speed sequence in the detection time period is obtained by the outlier algorithm, the average outlier factor in the sub-time period is used as the overall outlier factor in the sub-time period, and the difference between the average overall outlier factor of all sub-time periods and the preset outlier threshold is used as the outlier.

8. A mining reducer oil seal leakage fault diagnosis system, characterized in that: The system comprises: A data acquisition unit, used to obtain multi-dimensional state parameters of the reducer in real time during a preset detection time period, wherein the state parameters include: temperature, sound intensity, and rotation speed; A first abnormality analysis unit is used to divide the detection time period into multiple sub-time periods in time series; in each sub-time period, in the temperature dimension, according to the change amplitude and change trend of the data in the temperature sequence, obtain the temperature rise amount of each sub-time period; according to the temperature rise amount, select the suspected abnormal sub-time period; The second abnormality analysis unit is used to count the number of suspected abnormal sub-time periods and the change range of data in each state parameter dimension in the suspected abnormal sub-time period, and obtain the degree of underlubrication in each state parameter dimension; according to the degree of underlubrication in all state parameter dimensions, determine whether the current detection time period is abnormal; The oil seal leakage judgment unit is used to obtain the temperature-sound synchronization according to the difference in the degree of underlubrication between the temperature dimension and the sound intensity dimension, and the difference in the change amplitude in the same sub-time period if an abnormality occurs in the current detection time period; the outlier of the data in the speed dimension is counted, and the product of the outlier and the temperature-sound synchronization is used as the oil seal leakage degree. If the oil seal leakage degree is greater than a preset leakage degree threshold, it is judged that the oil seal of the mining reducer is leaking; The method for obtaining the degree of underlubrication comprises: For the temperature and sound intensity dimensions, the overall change amplitude of the data in each suspected abnormal sub-time period is obtained, and the overall change amplitudes in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the two state parameter dimensions; For the speed dimension, the overall change amplitude of the data in each suspected abnormal sub-time period is negatively correlated to obtain the speed reduction degree. The speed reduction degrees in all suspected abnormal sub-time periods are accumulated and multiplied by the number of suspected abnormal sub-time periods to obtain the degree of underlubrication in the speed dimension.

Images