A transmission system fault detection method for oil pumping unit maintenance

Through synchronous compression wavelet transformation and time frequency matrix analysis, the problem of weak fault signals of the pumping engine reducer is solved, more efficient fault detection is achieved, and the prediction accuracy and safety of the pumping engine transmission system faults are improved.

CN118641200BActive Publication Date: 2024-11-19DAQING GUANGXIN PETROLEUM MACHINERY EQUIP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410888275.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2024-11-19
Estimated Expiration
2044-07-04

AI Technical Summary

Technical Problem

The signal of the oil pump gearbox failure is weak in the early stage of the failure and is difficult to detect through traditional methods, which makes it difficult to detect the fault, affecting oil production and may cause safety accidents.

Method used

Synchronous compression wavelet transformation is used to obtain the time frequency matrix of the vibration signal, divide the transmission system modulation frequency band, calculate the frequency band characteristic value and fault value, combine the time frequency characteristic matrix and local warning value to determine the gearbox fault.

Benefits of technology

It improves the sensitivity and accuracy of fault detection of oil pump transmission system, can identify weak faults in the early stage, and reduces equipment damage and safety risks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118641200B_ABST
    Figure CN118641200B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of fault prediction, and specifically to a transmission system fault detection method for oil pump maintenance, the method comprising: simultaneously obtaining vibration signals of the reducer when the reducer is working on both sides of the oil pump to be detected, and dividing them into local vibration signals; obtaining the time-frequency matrix of each local vibration signal, dividing the frequency of each local vibration signal into each transmission system modulation band; determining the frequency band characteristic value of each transmission system modulation band; determining the transmission system fault value and time-frequency characteristic matrix of each local vibration signal; determining the local warning value of the oil pump reducer in the time period corresponding to each local vibration signal; based on the distribution of local warning values ​​of the oil pump reducer in normal operation and the oil pump reducer to be detected in multiple time periods, determining the fault condition of the oil pump reducer. The present application aims to improve the sensitivity and accuracy of the prediction of oil pump transmission system faults.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of fault prediction, and in particular to a transmission system fault detection method for oil pump maintenance. Background Art

[0002] Oil pumps are the main oil production equipment in my country today, but most oil pumps work in field environments with complex and varying geographical locations. Some areas are affected by extreme weather such as high temperature, high cold, high precipitation, and sandstorms, resulting in a high incidence of oil pump failures.

[0003] The reduction box is an important part of the transmission system of the oil pump, and its working condition directly affects the normal operation of the oil pump. The oil pump reduction box realizes the deceleration movement by rotating the meshing gears installed in the box, reducing the motor from the high speed to the normal operating speed required by the oil pump. The sealed box of the reduction box makes it impossible to use image analysis and other means for internal detection, which increases the difficulty of detecting internal faults. At present, the time-frequency analysis method is usually used to detect the vibration signal of the oil pump reduction box to determine whether there is a fault inside the reduction box. However, in the early stage of the fault, the fault signal is weak, causing it to be submerged in the normal signal and noise, making it difficult to detect. Often when the fault is discovered, the transmission system has been severely damaged, which will not only affect the oil production, but also may cause safety accidents in serious cases. Therefore, timely and accurate prediction of the fault of the oil pump reduction box and repair it is a key issue in the field of oil extraction. Summary of the invention

[0004] In view of the above, it is necessary to provide a transmission system fault detection method for oil pumping unit maintenance, which improves the sensitivity and accuracy of oil pumping unit transmission system fault prediction compared with traditional oil pumping unit transmission system fault detection methods.

[0005] A transmission system fault detection method for oil pumping unit maintenance in this application adopts the following technical solution:

[0006] An embodiment of the present application provides a transmission system fault detection method for oil pumping unit maintenance, the method comprising the following steps:

[0007] The vibration signals of the reduction box when working are obtained at both sides of the reduction box of the oil pumping unit to be tested, and are evenly divided into local vibration signals;

[0008] The time-frequency matrix of each local vibration signal is obtained by synchronous compression wavelet transform, and the frequency of each local vibration signal is divided into modulation frequency bands of each transmission system based on the distribution of elements in the time-frequency matrix of each local vibration signal;

[0009] Determine the frequency band characteristic value of the modulation frequency band of each transmission system based on the stability of the frequency amplitude within the modulation frequency band of each transmission system, as well as the average level and dispersion of the frequency amplitude;

[0010] Determine the transmission system fault value of each local vibration signal based on the distribution of the frequency amplitude of each local vibration signal and a time-frequency matrix of multiple adjacent local vibration signals at the same position in combination with the frequency band eigenvalue;

[0011] Combining the time-frequency matrix of each local vibration signal and the frequency band characteristic value, a time-frequency characteristic matrix of each local vibration signal is obtained; based on the distribution of transmission system fault values ​​of local vibration signals at the same position of different vibration signals, and the distance between the time-frequency characteristic matrices, the local warning value of the pumping unit reducer in the time period corresponding to each local vibration signal is determined;

[0012] Based on the distribution of local warning values ​​of the normally operating oil pumping unit reducer and the oil pumping unit reducer to be detected in multiple time periods, the fault condition of the oil pumping unit reducer is determined.

[0013] In one embodiment, the process of determining the modulation frequency band of the transmission system is as follows:

[0014] The accumulated results of each row of elements in each time-frequency matrix are recorded as accumulated amplitudes, and are used to form accumulated amplitude spectra respectively;

[0015] Smoothing each accumulated amplitude spectrum to obtain each ideal accumulated amplitude spectrum;

[0016] The frequency of each local vibration signal is divided into modulation frequency bands of each transmission system using the frequencies of the minimum points of each ideal cumulative amplitude spectrum as the segmentation boundaries.

[0017] In one embodiment, the process of determining the frequency band characteristic value is:

[0018] Calculate the distance between the accumulated amplitude in the accumulated amplitude spectrum and the corresponding ideal accumulated amplitude spectrum of the frequency in the modulation frequency band of each transmission system;

[0019] Calculate the mean and dispersion of the accumulated amplitude of the frequency in the modulation frequency band of each transmission system in the accumulated amplitude spectrum;

[0020] Based on the distance, the mean value and the dispersion degree of each transmission system modulation frequency band, a frequency band characteristic value of each transmission system modulation frequency band is determined.

[0021] In one embodiment, the frequency band characteristic value of each transmission system modulation frequency band is the product of the distance, the mean value and the dispersion degree of each transmission system modulation frequency band.

[0022] In one embodiment, the process of determining the transmission system fault value is as follows:

[0023] Taking any local vibration signal as the starting frame, taking a preset number of local vibration signals backward to form a transmission system signal window;

[0024] Calculate the variance of the frequency amplitudes at the same positions of the time-frequency matrix of all local vibration signals in the transmission system signal window, and calculate the mean of the normalized values ​​of the frequency band eigenvalues ​​of all transmission system modulation frequency bands at each frequency, which is recorded as the normalized mean;

[0025] Calculate the product of each variance and the normalized mean at the same position, and use the fusion result of all the products of any local vibration signal at each moment as the fault value of any local vibration signal at each moment;

[0026] Based on the fault values ​​of any local vibration signal at all times, a transmission system fault value of any local vibration signal is determined.

[0027] In one embodiment, the transmission system fault value of any local vibration signal is the accumulated value of the fault values ​​of any local vibration signal at all times.

[0028] In one embodiment, the process of determining the time-frequency feature matrix is ​​as follows:

[0029] For each local vibration signal of any vibration signal, the product of each element of the time-frequency matrix of the local vibration signal and the normalized value of the frequency band eigenvalue of the transmission system modulation frequency band where the frequency corresponding to the element is located is calculated to obtain the time-frequency characteristic matrix.

[0030] In one embodiment, the process of determining the local warning value is:

[0031] The transmission system fault values ​​of the local vibration signals in each transmission system signal window are normalized, and the average of the normalized values ​​of the transmission system fault values ​​of the local vibration signals at any same position of the two groups of vibration signals collected on both sides of the reduction box is recorded as the fault mean value;

[0032] Calculating the distance between the time-frequency feature matrices of the local vibration signals at any same position of the two groups of vibration signals, recorded as the feature distance;

[0033] Based on the fault mean and the characteristic distance between the local vibration signals at any same position, a local warning value of the local vibration signal of the oil pumping unit reduction box at any same position corresponding to the time period is determined.

[0034] In one embodiment, the local warning value of the local vibration signal of the oil pumping unit reduction box at any same position in the corresponding time period is: the product of the fault mean and the characteristic distance between the local vibration signals at any same position.

[0035] In one embodiment, the process of determining the fault condition of the oil pumping unit reduction box is as follows:

[0036] Respectively obtain the maximum values ​​of the local warning values ​​in a preset number of transmission system signal windows of the pumping unit reduction box in normal operation, and calculate the average value of all the maximum values, which is recorded as the warning degree threshold;

[0037] Calculate the average of the local warning values ​​in the current transmission system signal window of the reduction box of the pumping unit to be detected, and record it as the average value to be detected;

[0038] When the to-be-detected average value is greater than the warning threshold, it is determined that the pumping unit reduction box has a fault; otherwise, the pumping unit reduction box does not have a fault.

[0039] This application has at least the following beneficial effects:

[0040] According to the weakness of the pumping unit reducer fault information, the present application performs synchronous compression wavelet transform on the two sets of collected vibration signals, and performs denoising on the output time-frequency matrix, so as to facilitate the subsequent extraction of weak fault features in the vibration signal; through the frequency spectrum distribution of the pumping unit reducer vibration signal and the difference of fault information contained at different frequencies, the frequency band characteristic value is calculated to improve the pertinence of the fault-prone frequency, strengthen the subsequent analysis of the frequency interval rich in fault information, and improve the detection accuracy of the pumping unit reducer fault;

[0041] Furthermore, the frequency band characteristic value is used as the weight, and according to the randomness of the failure of the pumping unit reducer and the time-varying nature of the failure frequency, the transmission system failure value is calculated to enhance the recognition ability of minor failure information and improve the detection sensitivity of the pumping unit reducer failure.

[0042] Furthermore, by utilizing the symmetry of the internal structure of the pumping unit reducer, the local warning value is calculated according to the frequency band characteristic value and the transmission system fault value, so as to further amplify the fault information of the pumping unit reducer and enhance the recognition capability of the pumping unit reducer fault.

[0043] Furthermore, a warning threshold is set according to the local warning value of each local vibration signal obtained. When the average local warning value of the transmission system signal window exceeds the warning threshold, an early warning is issued, thereby improving the data utilization rate of the vibration signal, amplifying the fault information in a targeted manner, and enhancing the ability to identify weak fault frequencies, thereby improving the sensitivity and accuracy of the prediction of pumping unit transmission system faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A flowchart of the steps of a transmission system fault detection method for oil pumping unit maintenance provided in this application;

[0046] Figure 2 A schematic diagram of the process of obtaining the modulation frequency band of the transmission system;

[0047] Figure 3 is a schematic diagram of the process of determining the frequency band characteristic value;

[0048] Figure 4 It is a schematic diagram of the process of determining the fault value of the transmission system;

[0049] Figure 5 Schematic diagram of the process of determining the local warning value. DETAILED DESCRIPTION

[0050] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. It should be understood that, unless otherwise specified, " / " means or.

[0052] It should also be noted that the terms "first" and "second" in the present application are used to distinguish similar objects rather than to describe a specific order or sequence.

[0053] The following is a detailed description of a specific scheme of a transmission system fault detection method for oil pump maintenance provided by the present application in conjunction with the accompanying drawings.

[0054] An embodiment of the present application provides a transmission system fault detection method for oil pumping unit maintenance. Specifically, the following transmission system fault detection method for oil pumping unit maintenance is provided. Figure 1 , the method comprises the following steps:

[0055] Step 1: simultaneously obtain vibration signals of the reduction box when it is working on both sides of the reduction box of the oil pumping unit to be tested, and divide them into local vibration signals.

[0056] The same type of acceleration sensors are installed at the same position on both sides of the oil pumping unit reducer to collect the vibration signals of the reducer when the oil pumping unit is working. The two sets of vibration signals collected are subjected to the same frame division and preprocessing, and each frame of the vibration signal is recorded as a local vibration signal, where the preprocessing includes windowing, pre-emphasis, etc., and the number of signal points per frame is N.

[0057] In one embodiment of the present application, the value of N is 1024. The value of N is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0058] Step 2: Use synchronous compression wavelet transform to obtain the time-frequency matrix of each local vibration signal, and divide the frequency of each local vibration signal into modulation frequency bands of each transmission system based on the distribution of elements in the time-frequency matrix of each local vibration signal.

[0059] The synchronous compression wavelet transform is used to obtain the time-frequency matrix of each local vibration signal of each group of vibration signals. The synchronous compression wavelet transform is used to obtain the time-frequency matrix, which is a well-known technology and will not be described in detail in this application.

[0060] Due to the mixing of various components in the pumping unit reducer and the complex transmission path, the fault signal is submerged in the noise. Especially in the early stage of the fault, the fault signal is weak and easily covered by noise. Therefore, the K-SVD algorithm is used to reduce the noise of the time-frequency matrix.

[0061] The vibration signal of the oil pump reduction box is modulated by the meshing frequency of the gears at each level in the box, the rotation frequency of the shaft, and the characteristic frequency of the bearing, resulting in the vibration signal often having complex modulation characteristics. The modulation frequency band contains rich fault characteristic information, and identifying the fault characteristic frequency band near the modulation frequency is the key to predicting the gear failure of the reduction box. However, due to the interference of various factors such as noise, it is difficult to extract the modulation frequency band from the time-frequency matrix through traditional time-frequency analysis methods.

[0062] First, the cumulative result of each row of elements in each time-frequency matrix is ​​recorded as the cumulative amplitude, and each cumulative amplitude spectrum is arranged in order from small to large frequency. The cumulative amplitude spectrum can reflect the change of the amplitude of the vibration signal with frequency over the entire time scale.

[0063] Secondly, due to the failure of the oil pump reducer, its cumulative amplitude spectrum is rich in components and may contain multiple modulation bands. Therefore, it is necessary to effectively segment the cumulative amplitude spectrum to improve the accuracy of subsequent fault detection. The cumulative amplitude spectrum is smoothed to obtain an ideal cumulative amplitude spectrum, which not only retains the larger main components but also shields the noise and other interference components, which is conducive to the subsequent configuration of different weights for different modulation bands.

[0064] Finally, the frequency of each local vibration signal is divided into each transmission system modulation frequency band using the frequency of each minimum point of each ideal cumulative amplitude spectrum as the segmentation boundary. The schematic diagram of the acquisition process of the transmission system modulation frequency band is shown in Figure 2 shown.

[0065] Among them, arranging the accumulated amplitudes in order from small to large frequency to form an accumulated amplitude spectrum is only one embodiment of the present application. The implementer can also arrange the accumulated amplitudes in order from large to small frequency to form an accumulated amplitude spectrum.

[0066] In one embodiment of the present application, an exponential moving average algorithm is used to smooth the accumulated amplitude spectrum. As other implementation methods, on the basis of being able to smooth the accumulated amplitude spectrum, the implementer may use other existing technologies to smooth the accumulated amplitude spectrum, such as the Kalman filtering method, the moving average method, etc., and the present application does not impose any special restrictions.

[0067] Step 3: Determine the frequency band characteristic value of each transmission system modulation band based on the stability of the frequency amplitude within each transmission system modulation band, as well as the average level and dispersion of the frequency amplitude.

[0068] Since the fault information of the oil pumping unit reducer is mainly concentrated in the modulation frequency band of the vibration signal, the frequency band characteristic values ​​of the modulation frequency band of each transmission system are determined by the characteristics of the modulation frequency band. The specific process is as follows:

[0069] Calculate the distance between the accumulated amplitude in the accumulated amplitude spectrum and the corresponding ideal accumulated amplitude spectrum of the frequency in the modulation frequency band of each transmission system;

[0070] Calculate the mean and dispersion of the accumulated amplitude of the frequency in the modulation frequency band of each transmission system in the accumulated amplitude spectrum;

[0071] The product of the distance, the mean value and the discrete degree of the transmission system modulation frequency band is used as the frequency band characteristic value of the transmission system modulation frequency band.

[0072] In one embodiment of the present application, the distance between the accumulated amplitudes is the DTW distance. As other implementation methods, on the basis of being able to measure the distance between the accumulated amplitudes, the implementer may use other existing technologies to measure the distance between the accumulated amplitudes, such as Euclidean distance, Manhattan distance, etc., and the present application does not impose any special restrictions.

[0073] In one embodiment of the present application, the degree of dispersion is the standard deviation. The degree of dispersion may also be a statistic such as variance or coefficient of variation that describes the degree of unevenness of the accumulated amplitude.

[0074] It should be noted that the modulation frequency band of the oil pump reducer has the largest energy in its vibration signal. Therefore, the larger the mean value of the cumulative amplitude of the transmission system modulation frequency band, the greater the corresponding energy, the greater the possibility that it is in the actual modulation frequency band, and the larger the frequency band characteristic value; when a fault occurs, the dispersion degree of the cumulative amplitude in the cumulative amplitude spectrum corresponding to the transmission system modulation frequency band is large, the more the transmission system modulation frequency band deviates from the actual modulation frequency band, the smaller the dispersion degree, the less fault information, and the smaller the frequency band characteristic value. The flow chart of determining the frequency band characteristic value is as follows Figure 3 shown.

[0075] The frequency band characteristic value can reflect the richness of fault information contained in the modulation frequency band, which is convenient for subsequent strengthening of the analysis of the frequency rich in fault information to improve the detection accuracy of the pumping unit reducer fault.

[0076] Step 4: Determine the transmission system fault value of each local vibration signal based on the distribution of the frequency amplitude of each local vibration signal and a plurality of adjacent local vibration signals at the same position in combination with the frequency band eigenvalue.

[0077] In the early stage of the oil pumping unit reduction box failure, the probability of the failure frequency is very small, and its periodicity is not obvious. In order to enhance the ability to capture fault information, any local vibration signal is used as the starting frame, and a preset first number of local vibration signals are taken backward to form a transmission system signal window.

[0078] In one embodiment of the present application, the value of the preset first number is 100. The value of the preset first number is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0079] Calculate the variance of the frequency amplitudes at the same positions of the time-frequency matrix of all local vibration signals in the transmission system signal window, and calculate the mean of the normalized values ​​of the frequency band eigenvalues ​​of all transmission system modulation frequency bands at each frequency, which is recorded as the normalized mean;

[0080] Calculate the product of each variance and the normalized mean at the same position, and use the fusion result of all the products of any local vibration signal at each moment as the fault value of any local vibration signal at each moment;

[0081] The transmission system fault value of any local vibration signal is the accumulated value of the fault values ​​of any local vibration signal at all times.

[0082] It should be understood that fusion refers to combining multiple independent variables in a way that enhances the overall effect, such as additive relationship, multiplicative relationship, etc. Implementers can define it according to actual conditions.

[0083] In one embodiment of the present application, the sum of all the products of the local vibration signal at each moment is used as the fault value of the local vibration signal at each moment.

[0084] It is important to understand that: the larger the variance, the greater the change in the distribution of the frequency components of the vibration signal, the greater the probability of the fault frequency and the amplitude of the fault frequency, and the greater the fault value; when the fault value is larger, the greater the impact of the fault frequency on the distribution of the vibration signal frequency, the greater the corresponding transmission system fault value, the greater the probability of the pumping unit reduction box failure, and the greater the transmission system fault value. The schematic diagram of the transmission system fault value determination process is as follows Figure 4 shown.

[0085] At this point, the transmission system fault value of each local vibration signal is calculated, and the time-varying characteristics of the frequency distribution of the vibration signal of the oil pumping unit reducer are obtained, thereby enhancing the ability to identify minor fault information and improving the detection sensitivity of the oil pumping unit reducer fault.

[0086] Step 5, combining the time-frequency matrix of each local vibration signal and the frequency band characteristic value to obtain the time-frequency characteristic matrix of each local vibration signal; based on the distribution of transmission system fault values ​​of local vibration signals at the same position of different vibration signals, and the distance between the time-frequency characteristic matrices, determine the local warning value of the oil pump reducer in the time period corresponding to each local vibration signal.

[0087] The internal structure of the oil pumping unit reduction box usually has good symmetry. Under normal operation, the vibration signals detected on both sides are very similar. When the gear wear, gear eccentricity, shaft imbalance and other faults occur in the reduction box, the symmetrical structure of the reduction box is destroyed, and the vibration signals on both sides will show obvious differences, and the similarity will be seriously affected.

[0088] The process of determining the local warning value is as follows:

[0089] First, since the fault information of the oil pumping unit reducer is mostly reflected in the modulation frequency band of the vibration signal, the element with the larger frequency band eigenvalue in the time-frequency matrix has a larger change when the fault occurs. In order to enhance the amplification of fault information, for each local vibration signal of any vibration signal, the product of each element of the time-frequency matrix of the local vibration signal and the normalized value of the frequency band eigenvalue of the transmission system modulation frequency band where the element's corresponding frequency is located is calculated to obtain the time-frequency feature matrix.

[0090] Secondly, in the early stage of the failure of the oil pump reducer, the periodicity of its fault information is not obvious, and the fault information contained in each local vibration signal in the transmission system signal window is highly random.

[0091] The transmission system fault values ​​of the local vibration signals in each transmission system signal window are normalized, and the average of the normalized values ​​of the transmission system fault values ​​of the local vibration signals at any same position of the two groups of vibration signals is recorded as the fault mean value.

[0092] Finally, the distance between the time-frequency feature matrices of the local vibration signals at any same position of the two groups of vibration signals is calculated and recorded as the feature distance. The fusion result of the fault mean and the feature distance between the local vibration signals at any same position is used as the local warning value of the local vibration signal of the oil pump reducer at any same position in the corresponding time period.

[0093] In one embodiment of the present application, the product of the fault mean and the characteristic distance between the local vibration signals at any same position is used as the local warning value of the local vibration signal of the oil pump reduction box at any same position in the corresponding time period.

[0094] In one embodiment of the present application, the distance between matrices is the Euclidean distance. As other implementations, based on the calculable distance between matrices, the implementer may use other existing technologies to obtain the distance between matrices, such as the Manhattan distance, etc. This application does not impose any special restrictions.

[0095] It should be noted that: when a pumping unit reducer fails, its internal symmetry is destroyed. The worse the similarity between the time-frequency feature matrices of the local vibration signals at the same position of the two sets of vibration signals, the greater the distance between the two sets of vibration signals. At the same time, the greater the degree of change of the frequency distribution of the current local vibration signal over time, the more fault information it contains, the larger the corresponding fault mean, and the larger the local warning value. The local warning value reflects the impact of the internal fault of the pumping unit reducer on its symmetrical structure, further amplifies the fault information of the pumping unit reducer, and enhances the ability to identify the pumping unit reducer fault. The schematic diagram of the process of determining the local warning value is shown in the figure. Figure 5 shown.

[0096] Step 6: determine the fault condition of the oil pumping unit reducer based on the distribution of the local warning values ​​of the oil pumping unit reducer in normal operation and the oil pumping unit reducer to be detected in multiple time periods.

[0097] Respectively obtain the maximum values ​​of the local warning values ​​in the preset second number of transmission system signal windows of the pumping unit reduction box in normal operation, and calculate the average value of all the maximum values, which is recorded as the warning degree threshold;

[0098] Calculate the average of the local warning values ​​in the current transmission system signal window of the reduction box of the pumping unit to be detected, and record it as the average value to be detected;

[0099] When the average value to be detected is greater than the warning threshold, it is determined that there is a fault in the reduction box of the oil pumping unit and an early warning is issued; otherwise, there is no fault in the reduction box of the oil pumping unit.

[0100] In one embodiment of the present application, the preset value of the second number is 500. The value of the preset second number is preset manually and can be set by the implementer. This application does not impose any special restrictions.

[0101] In summary, according to the weakness of the pumping unit reducer fault information, the present application performs synchronous compression wavelet transform on the two sets of collected vibration signals, and performs denoising on the output time-frequency matrix, so as to facilitate the subsequent extraction of weak fault features in the vibration signal; through the frequency spectrum distribution of the pumping unit reducer vibration signal and the difference in fault information contained at different frequencies, the frequency band characteristic value is calculated to improve the pertinence of the fault-prone frequency, strengthen the subsequent analysis of the frequency interval rich in fault information, and improve the detection accuracy of the pumping unit reducer fault;

[0102] Furthermore, the frequency band characteristic value is used as the weight, and according to the randomness of the failure of the pumping unit reducer and the time-varying nature of the failure frequency, the transmission system failure value is calculated to enhance the recognition ability of minor failure information and improve the detection sensitivity of the pumping unit reducer failure.

[0103] Furthermore, by utilizing the symmetry of the internal structure of the pumping unit reducer, the local warning value is calculated according to the frequency band characteristic value and the transmission system fault value, so as to further amplify the fault information of the pumping unit reducer and enhance the recognition capability of the pumping unit reducer fault.

[0104] Furthermore, a warning threshold is set according to the local warning value of each local vibration signal obtained. When the average local warning value of the transmission system signal window exceeds the warning threshold, an early warning is issued, thereby improving the data utilization rate of the vibration signal, amplifying the fault information in a targeted manner, and enhancing the ability to identify weak fault frequencies, thereby improving the sensitivity and accuracy of the prediction of pumping unit transmission system faults.

[0105] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0106] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the basic features of the present application. Therefore, no matter from which point of view, the above embodiments of the present application should be regarded as exemplary and non-restrictive.

Claims

1. A transmission system fault detection method for oil pumping unit maintenance, It is characterized in that The method comprises the following steps: The vibration signals of the reduction box when working are obtained at both sides of the reduction box of the oil pumping unit to be tested, and are evenly divided into local vibration signals; The time-frequency matrix of each local vibration signal is obtained by synchronous compression wavelet transform, and the frequency of each local vibration signal is divided into modulation frequency bands of each transmission system based on the distribution of elements in the time-frequency matrix of each local vibration signal; Based on the stability of the frequency amplitude within the modulation frequency band of each transmission system, as well as the average level and dispersion of the frequency amplitude, the frequency band characteristic value of the modulation frequency band of each transmission system is determined; Determine the transmission system fault value of each local vibration signal based on the distribution of the frequency amplitude of each local vibration signal and a time-frequency matrix of multiple adjacent local vibration signals at the same position in combination with the frequency band eigenvalue; Combining the time-frequency matrix of each local vibration signal and the frequency band characteristic value, a time-frequency characteristic matrix of each local vibration signal is obtained; based on the distribution of transmission system fault values ​​of local vibration signals at the same position of different vibration signals, and the distance between the time-frequency characteristic matrices, the local warning value of the pumping unit reducer in the time period corresponding to each local vibration signal is determined; Based on the distribution of local warning values ​​of the normally operating pumping unit reducer and the pumping unit reducer to be tested in multiple time periods, the fault condition of the pumping unit reducer is determined; The process of determining the transmission system fault value is as follows: Taking any local vibration signal as the starting frame, taking a preset number of local vibration signals backward to form a transmission system signal window; Calculate the variance of the frequency amplitudes at the same positions of the time-frequency matrix of all local vibration signals in the transmission system signal window, and calculate the mean of the normalized values ​​of the frequency band eigenvalues ​​of all transmission system modulation frequency bands at each frequency, which is recorded as the normalized mean; Calculate the product of each variance and the normalized mean at the same position, and use the fusion result of all the products of any local vibration signal at each moment as the fault value of any local vibration signal at each moment; Based on the fault values ​​of any local vibration signal at all times, a transmission system fault value of any local vibration signal is determined.

2. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 1, It is characterized in that The process of determining the modulation frequency band of the transmission system is as follows: The accumulated results of each row of elements in each time-frequency matrix are recorded as accumulated amplitudes, and are used to form accumulated amplitude spectra respectively; Smoothing each accumulated amplitude spectrum to obtain each ideal accumulated amplitude spectrum; The frequency of each local vibration signal is divided into modulation frequency bands of each transmission system using the frequencies of the minimum points of each ideal cumulative amplitude spectrum as the segmentation boundaries.

3. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 2, It is characterized in that The process of determining the frequency band characteristic value is as follows: Calculate the distance between the accumulated amplitude in the accumulated amplitude spectrum and the corresponding ideal accumulated amplitude spectrum of the frequency in the modulation frequency band of each transmission system; Calculate the mean and dispersion of the accumulated amplitude of the frequency in the modulation frequency band of each transmission system in the accumulated amplitude spectrum; Based on the distance, the mean value and the dispersion degree of each transmission system modulation frequency band, a frequency band characteristic value of each transmission system modulation frequency band is determined.

4. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 3, It is characterized in that The frequency band characteristic value of each transmission system modulation frequency band is the product of the distance, the mean value and the dispersion degree of each transmission system modulation frequency band.

5. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 1, It is characterized in that The transmission system fault value of any local vibration signal is the accumulated value of the fault values ​​of any local vibration signal at all times.

6. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 1, It is characterized in that The determination process of the time-frequency feature matrix is ​​as follows: For each local vibration signal of any vibration signal, the product of each element of the time-frequency matrix of the local vibration signal and the normalized value of the frequency band eigenvalue of the transmission system modulation frequency band where the frequency corresponding to the element is located is calculated to obtain the time-frequency characteristic matrix.

7. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 1, It is characterized in that The process of determining the local warning value is as follows: The transmission system fault values ​​of the local vibration signals in each transmission system signal window are normalized, and the average of the normalized values ​​of the transmission system fault values ​​of the local vibration signals at any same position of the two groups of vibration signals collected on both sides of the reduction box is recorded as the fault mean value; Calculating the distance between the time-frequency feature matrices of the local vibration signals at any same position of the two groups of vibration signals, recorded as the feature distance; Based on the fault mean and the characteristic distance between the local vibration signals at any same position, a local warning value of the local vibration signal of the oil pumping unit reduction box at any same position corresponding to the time period is determined.

8. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 7, It is characterized in that The local warning value of the local vibration signal of the oil pumping unit reduction box at any same position in the corresponding time period is: the product of the fault mean value and the characteristic distance between the local vibration signals at any same position.

9. A transmission system fault detection method for oil pumping unit maintenance as claimed in claim 1, It is characterized in that The process of determining the fault condition of the pumping unit reduction box is as follows: Respectively obtain the maximum values ​​of the local warning values ​​in a preset number of transmission system signal windows of the pumping unit reduction box in normal operation, and calculate the average value of all the maximum values, which is recorded as the warning degree threshold; Calculate the average of the local warning values ​​in the current transmission system signal window of the reduction box of the pumping unit to be detected, and record it as the average value to be detected; When the to-be-detected average value is greater than the warning threshold, it is determined that the pumping unit reduction box has a fault; otherwise, the pumping unit reduction box does not have a fault.