Method for automatically diagnosing fault of top drive gearbox based on current

Abstract

The invention discloses a current-based method for automatically diagnosing faults of a top drive gearbox. The method comprises the following steps: S1, mounting a broadband current sensor on a power supply cable of a top drive motor of a top drive electric control room to collect and collect current data for 4 seconds; s2, performing low-pass filtering on the current data, wherein the low-pass filtering is specifically 0-4000 Hz; s3, performing short-time Fourier transform (STFT) on the current to determine a steady state; s4, fast Fourier transform (FFT) is carried out on the current, and the rotating speed of the motor is estimated; s5, calculating the characteristic frequency of the gearbox; s6, amplitude demodulation is carried out on the current data in the step 2, so that the interference of the power supply frequency can be eliminated; s7, performing discrete wavelet transform (DWT) on the demodulated data, wherein the discrete wavelet transform is specifically four-level decomposition; s8, carrying out FFT (Fast Fourier Transform) on the detail 3 and the approximate4; s9, performing frequency search on the spectrogram after FFT, and obtaining a judgment index E; and S10, performing threshold judgment on the E to judge the health degree of the top drive gearbox. According to the invention, the health state of the top drive gearbox can be judged by using a simple method.

Description

Technical Field

[0001] The present invention relates to the field of equipment fault diagnosis, and more specifically to a method for automatically diagnosing top drive gearbox faults based on current. Background Art

[0002] The top drive of drilling is the core equipment of drilling operation, and there is no backup device. Therefore, the top drive plays a vital role in drilling operation and underground safety. Due to the unpredictability of drilling conditions and the complexity of the top drive's own structure, the top drive will inevitably fail during operation, which may affect production efficiency and the quality of drilling projects, or may cause drilling accidents and casualties. Among them, the biggest threat to life and property is the sudden catastrophic failure leading to downtime. How to effectively avoid sudden downtime events, let users perceive the health status of the equipment in advance, take targeted maintenance measures, and flexibly and conveniently arrange the progress of drilling operations and maintenance time and spare parts in advance, greatly reduce safety risks, improve operation efficiency, and reduce maintenance time and money costs, which is of great significance to the development of drilling staff and companies, and even the development of the entire drilling industry. And with the continuous updating and development of science and technology, the top drive of drilling has begun to develop towards automation and intelligence, making the top drive drilling system more complex, and its sudden downtime will bring greater risks. Therefore, equipment status monitoring and fault diagnosis are the cornerstones of automation and intelligence, and are necessary conditions for the development and progress of the drilling industry.

[0003] Top drive failures often occur in the bearings and gears of the gearbox, which is a two-stage reduction gearbox. The transmission is the meshing of the motor pinion and the second shaft to achieve the first-stage reduction, and the second shaft and the main shaft are connected through gears to achieve the second-stage reduction. The main shaft drives the drill pipe to achieve the drilling function. The gears and bearings involved will have an early wear stage before a catastrophic failure occurs. At present, the vibration monitoring method is generally used in the industry to monitor the bearings in the gearbox, and the health status is analyzed by collecting vibration signals, but it has two main disadvantages: first, it is difficult to install, and the vibration sensor needs to be close to the monitoring target, and a high-precision installation direction is required, which brings great inconvenience in practical applications; second, due to the complex structure of the gearbox, with multiple bearings, shafts and gears, the sensor can only be installed on the outer casing, which makes it difficult to locate the specific fault part of the gearbox. In addition, there are also shortcomings such as being susceptible to interference and high cost. In the top drive application, it is particularly restricted by the inconvenience of installation. Because the top drive is high above the ground during operation and moves up and down for tens of meters, it is difficult to apply vibration in the health monitoring of the top drive gearbox and the effect is not good.

[0004] The current method has been applied in the health monitoring and fault diagnosis of motors and their loads. Its biggest advantage is that it is very convenient to install. It only needs to install sensors and digital acquisition equipment in the electric control room. The environment of the electric control room is stable and mild, so the accuracy of data acquisition is rarely affected by environmental interference factors, so that the signal can accurately restore the original signal, which is very important for result judgment. The principle of health diagnosis for gearboxes is that the deterioration of the gearbox will cause the vibration of the gearbox, and its vibration will be fed back to the motor along the power transmission system to cause the rotor to vibrate, causing the air gap to change, and then reflected in the current. Therefore, the characteristic frequency of the gearbox in the current can be used to diagnose the health status of the gearbox of the top drive, but the signal generated by the fault characteristics of the gearbox in the stator current is very small, and the top drive often has load fluctuations during drilling, and the top drive motor is driven by a frequency converter, which will generate a large amount of harmonic interference, which will mask the fault signal of the gearbox.

[0005] Therefore, how to provide a method for automatically diagnosing top drive gearbox faults based on current is a problem that those skilled in the art urgently need to solve. Summary of the invention

[0006] In view of this, the present invention provides a method for automatically diagnosing top drive gearbox faults based on current.

[0007] The present invention discloses a method for automatically diagnosing a top drive gearbox fault based on current, comprising the following steps:

[0008] S1. Install a broadband current sensor on the power supply cable of the top drive motor in the top drive electric control room to collect and collect current data for a specific time of 4 consecutive seconds;

[0009] S2, low-pass filter the current data, specifically 0-4000Hz;

[0010] S3, perform short-time Fourier transform (STFT) on the current to determine the steady state;

[0011] S4, perform fast Fourier transform (FFT) on the current to estimate the motor speed;

[0012] S5. Calculate the characteristic frequency of the gearbox;

[0013] S6, performing amplitude demodulation on the current data of step S2 to remove interference from the power supply frequency;

[0014] S7, performing discrete wavelet transform (DWT) on the demodulated data, specifically 4-level decomposition;

[0015] S8, perform FFT on detail 3 and approximate 4;

[0016] S9, perform frequency search on the spectrum after FFT and obtain the judgment index E;

[0017] S10, making a threshold determination on E to judge the health of the top drive gearbox.

[0018] Furthermore, the current signal is determined by using the short-time Fourier transform (STFT) method with a window length of 0.5s. When the main frequency (i.e., power supply frequency) of each STFT transformation does not exceed 1Hz, it can be considered as a stable signal and the next step of analysis is performed. Otherwise, it is re-collected until the conditions are met.

[0019] Furthermore, the characteristic frequency of the gearbox includes the gear shaft rotation frequency and the gear meshing frequency. The rotation speed of each shaft and the gear meshing frequency are calculated as shown in formula (1):

[0020]

[0021] Furthermore, the actual speed of the motor is estimated by calculating the actual speed of the motor as shown in formula (2):

[0022]

[0023] Furthermore, the shaft frequency and meshing frequency of the gear box are calculated by the calculation formula in formula (1) and the motor speed value obtained in formula (2).

[0024] Furthermore, the demodulation method is to obtain the amplitude, frequency and phase of the power supply current through the fast Fourier transform (FFT) in step S4, and obtain the demodulated signal by multiplying the original signal with the modulated signal.

[0025] Furthermore, the method for automatic diagnosis of top drive gearbox fault based on current is characterized in that the discrete wavelet transform (DWT) transform is defined as formula (3):

[0026]

[0027] Furthermore, the number of decompositions is called the level, and its mathematical expression is as shown in formula (4):

[0028]

[0029] Furthermore, the expression for performing frequency search on the spectrum after FFT in step 8 is expressed by equation (5):

[0030]

[0031] Through the above technical solutions, it can be known that compared with the prior art, the method for diagnosing the health status of the gearbox based on the current method disclosed in the present invention, compared with the vibration method commonly used at present, makes it unnecessary to install sensors on the top drive body during the diagnosis process, and does not need to be interfered with by the complex outdoor environmental factors on the sensor data, so that a simple method can be used to judge the health status of the top drive gearbox. In addition, the current-based gearbox diagnosis method is mainly aimed at the frequency characteristics of the top drive gearbox and the high interference data under complex working conditions, and a set of methods are designed from the selection of sensors for data collection to the result judgment. In the process of online automated diagnosis, the influence of interference data on the result judgment can be automatically avoided, and the frequency band can be processed separately through wavelet transform processing, which can highlight the difficulty of finding the frequency of the gearbox due to the small energy value of the characteristic frequency. At the same time, this algorithm can be stored and used on any computer or embedded device. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0033] Figure 1 The accompanying drawing is a flow chart of a method for automatic diagnosis of top drive gearbox fault based on current in the present invention.

[0034] Figure 2 The accompanying drawing is a schematic diagram of the gearbox structure of the top drive in the present invention.

[0035] Figure 3 The attached figure is a 4-level decomposition tree diagram of discrete wavelet transform (DWT) in the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] Example 1

[0038] Please refer to the instruction manual Figure 1-3

[0039] The present invention discloses a method for automatically diagnosing a top drive gearbox fault based on current, comprising the following steps:

[0040] S1. Install a broadband current sensor on the power supply cable of the top drive motor in the top drive electric control room to collect and collect current data for a specific time of 4 consecutive seconds;

[0041] In step S1, since the top drive is generally driven by a frequency converter and has high-frequency components, and the characteristic signal generated by the fault has high-frequency components, a broadband current sensor must be selected when collecting the current signal.

[0042] S2, low-pass filter the current data, specifically 0-4000Hz;

[0043] In step S2, the signal collected in step S1 is low-pass filtered. This is because the meshing frequency of the gearbox and its multiples do not exceed 4000 Hz. Therefore, high-frequency interference in the signal is prevented. After the current signal is collected, the signal is low-pass filtered with a pass frequency of (0-4000 Hz).

[0044] S3, performing short-time Fourier transform (STFT) on the current to determine a stable signal;

[0045] The Fourier transform used in step S3 is a conventional means of frequency domain analysis, but it is only applicable to time-invariant signals. Since the top drive is driven by a frequency converter, when the frequency converter working mode is torque mode, its speed will often change (i.e., the power supply frequency). When FFT is performed on the signal with speed fluctuation, spectrum leakage will occur, resulting in extremely poor resolution. Therefore, before further frequency domain analysis, it is necessary to determine that the signal is a time-invariant signal, so the determination method is to use the short-time Fourier transform (STFT) method with a window length of 0.5s. When the main frequency (i.e., the power supply frequency) of each STFT transformation does not exceed 1Hz, it can be considered as a stable signal and the next step of analysis is performed. Otherwise, re-collect until the conditions are met.

[0046] S4, perform fast Fourier transform (FFT) on the current to estimate the motor speed;

[0047] The power of the top drive in the actual drilling process is provided by an asynchronous motor 1, and its main shaft speed generally does not exceed 100rpm. The gearbox reduction ratios of different models of top drives are different, but are all between 10-14. Figure 2 Assume that the number of teeth on the motor pinion is 2, the number of teeth on the second axis gear is 3, the number of teeth on the second axis pinion is 4, and the number of teeth on the main axis gear 5 is N1, N2, N3, and N4 respectively. The motor shaft speed is N0 (rps, revolutions per second),

[0048] In step S4, the speed of each shaft and the gear meshing frequency are calculated as shown in formula (1):

[0049]

[0050] In step S4, the actual speed of the motor is calculated as shown in formula (2):

[0051]

[0052] The fault signal of the gearbox will modulate the amplitude and frequency of the current signal, so it is necessary to determine the characteristic frequency of the gearbox.

[0053] Furthermore, the characteristic frequency of the gearbox includes the gear shaft rotation frequency and the gear meshing frequency, which is calculated as shown in formula (1). It can be seen from formula (1) that the prerequisite for calculating the characteristic frequency of the gearbox is to know the motor shaft rotation speed.

[0054] Furthermore, since the top drive often uses an encoderless mode during actual operation, it is necessary to estimate the motor speed.

[0055] Furthermore, the estimation algorithm principle is as follows: during the manufacturing process of the top drive, the rotor has an eccentricity within the tolerance range, and during use, the load fluctuates greatly, which will increase the degree of eccentricity after a period of time, so the eccentricity frequency is used to estimate the actual speed of the motor. The actual speed of the motor is calculated as shown in formula (2). The synchronous speed is first calculated by the current frequency. According to the principle of asynchronous motor, its rotor will have a slip s between the synchronous speed and the synchronous speed. S will increase with the increase of load. During the actual normal drilling process of the top drive, its load generally does not exceed the rated load. The actual slip is between (0-smax). By performing FFT on the obtained stable signal, the spectrum is obtained, and the maximum value of fl±fsyn to fl±fsyn(1-smax) in the spectrum is obtained as the actual speed fr value. That is, the N0 value in formula (1).

[0056] S5. Calculate the characteristic frequency of the gearbox;

[0057] The shaft frequency and meshing frequency of the gear box are calculated by the calculation formula in equation (1) and the motor speed value obtained in step S4.

[0058] S6, performing amplitude demodulation on the current data of step 2 to remove interference from the power supply frequency;

[0059] The gearbox fault has been proven to produce amplitude modulation (AM) and phase modulation (PM) of the stator current. Since the power supply frequency is the absolute main frequency in the stator current spectrum, and the gearbox fault signal is its sideband, it is easily submerged by its main frequency and its harmonics. Therefore, in order to highlight its sideband, the signal in step (2) is amplitude demodulated for further analysis.

[0060] Specifically, the demodulation method is to obtain the amplitude, frequency and phase of the power supply current through the fast Fourier transform (FFT) in step S4, and obtain the demodulated signal by multiplying the original signal with the modulated signal.

[0061] S7, performing discrete wavelet transform (DWT) on the demodulated data, specifically 4-level decomposition;

[0062] In step S7, the discrete wavelet transform (DWT) is defined as follows:

[0063]

[0064] In step S7, the number of decompositions is called the level, and its mathematical expression is as shown in formula (4):

[0065]

[0066] By performing discrete wavelet transform (DWT) analysis on the demodulated signal, the advantage of wavelet transform is that the bandwidth of each layer will not be interfered by other layers. In addition, a better signal-to-noise ratio can be obtained. DWT is a unified orthogonal transform, which is calculated by moving and expanding the wavelet kernel and associating it with the input signal. The transform definition is as shown in formula (3). DWT is also called multi-resolution analysis, which uses orthogonal wavelets, such as the Daubechies series of wavelets, to decompose the signal into different frequency bandwidths. This is equivalent to passing the signal through a high-pass filter, and then downsampling the result to produce a high-frequency signal called details, and passing it through a low-pass filter and downsampling to produce a low-frequency signal called an approximation. The approximation can be decomposed again into another approximation and details. The number of decompositions is called a level, and the mathematical expression is expressed by formula (4). Here we choose the "db8" wavelet and decompose the signal into 4 layers. The tree diagram is shown in the attached figure. Figure 3

[0067] S8, perform FFT on detail 3 and approximate 4;

[0068] See the instruction manual Figure 3

[0069] In step S8, FFT is performed on approximate4 and detail3 after wavelet decomposition in step S7 to obtain fine frequency domain images in the range of 0-250 Hz and 500-1000 Hz.

[0070] S9, perform frequency search on the spectrum after FFT and obtain the judgment index E;

[0071] Perform frequency search on the spectra in the 0 - 250 Hz and 500 - 1000 Hz intervals after FFT in step 8. The specific search frequencies are the shaft frequency, meshing frequency, and their harmonic frequencies in Equation (1).

[0072] Specifically, it is represented by Equation (5):

[0073]

[0074] In the search process, peak search should be performed on the spectrogram first. Specifically, a sliding window is used to search for the local maximum value higher than the threshold in the spectrum. The search range for each time is the window, the window length is set to 4, and the threshold is set to -70 dB. After detecting the peak, compare it with fr and fm in Equation (5), and calculate the difference between the maximum amplitudes of fr and fm respectively and -70 as the judgment index E.

[0075] S10. Perform threshold judgment on E to determine the health condition of the top drive gearbox;

[0076] Judge the health condition through the threshold of the judgment index E. Specifically:

[0077] E <= 5, healthy;

[0078] 5 < E <= 8, with slight wear;

[0079] 8 < E <= 12, severe wear or broken teeth;

[0080] E > 12, with broken teeth;

[0081] The method for diagnosing the health state of a gearbox based on the current method disclosed in the present invention, compared with the currently commonly used vibration method, makes its diagnosis process not require installing sensors on the top drive body and not be affected by the interference of complex outdoor environmental factors on sensor data, enabling a simple method to judge the health state of the top drive gearbox. In addition, the gearbox diagnosis method based on current mainly aims at the frequency characteristics of the top drive gearbox and the highly interfering data under complex working conditions, and designs a set of methods specifically from the selection of sensors for data collection to result judgment. In the online automatic diagnosis process, it can automatically avoid the influence of interfering data on result judgment, and through wavelet transform processing, separate processing in frequency bands can highlight the difficulty of finding the frequency due to the too small energy value of the gearbox characteristic frequency. At the same time, this algorithm can be stored and used on any computer or embedded device.

[0082] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The present invention discloses a method for automatically diagnosing top drive gearbox faults based on current, characterized in that: The following steps are involved: S1. Install a broadband current sensor on the power supply cable of the top drive motor in the top drive electric control room to collect and collect current data for a specific time of 4 consecutive seconds; S2, low-pass filter the current data, specifically 0-4000Hz; S3, perform short-time Fourier transform (STFT) on the current to determine the steady state; S4, perform fast Fourier transform (FFT) on the current to estimate the motor speed; S5. Calculate the characteristic frequency of the gearbox; S6, performing amplitude demodulation on the current data of step S2 to remove interference from the power supply frequency; S7, performing discrete wavelet transform (DWT) on the demodulated data, specifically 4-level decomposition; S8, perform FFT on detail 3 and approximate 4; S9, perform frequency search on the spectrum after FFT and obtain the judgment index E; S10, making a threshold determination on E to judge the health of the top drive gearbox.

2. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The current signal determination method adopts the short-time Fourier transform (STFT) method with a window length of 0.5s. When the main frequency (i.e., power supply frequency) of each STFT transformation does not exceed 1Hz, it can be considered as a stable signal and the next step of analysis is carried out. Otherwise, re-collection is performed until the conditions are met.

3. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The characteristic frequency of the gearbox includes the gear shaft rotation frequency and the gear meshing frequency. The rotation speed of each shaft and the gear meshing frequency are calculated as shown in formula (1):

4. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The actual speed of the motor is calculated as shown in formula (2): f fr --Dynamic eccentricity frequency in current signal fl--current frequency, p--number of motor pole pairs fr--motor shaft speed s--slip s max --Slip at rated load f ratio --Rated frequency V rpm -Rated speed.

5. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 3 or 4, characterized in that: The shaft frequency and meshing frequency of the gear box are calculated by the calculation formula in formula (1) and the motor speed value obtained in formula (2).

6. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The demodulation method is to obtain the amplitude, frequency and phase of the power supply current through the fast Fourier transform (FFT) in step S4, and obtain the demodulated signal by multiplying the original signal with the modulated signal.

7. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The discrete wavelet transform (DWT) is defined as follows: Ψ(n)-wavelet kernel function j-- telescopic parameters k--Translation parameter.

8. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The number of decompositions is called the level, and its mathematical expression is as shown in formula (4): α--Number of layers x[n]--input signal h[n]--high pass filter g[n]--Low-pass filter.

9. The method for automatic diagnosis of top drive gearbox fault based on current according to claim 1, characterized in that: The expression for performing frequency search on the spectrum after FFT in step 8 is expressed by equation (5):

Images