Composite fault diagnosis method under low rotating speed working condition based on encoder signal

Abstract

The invention relates to the technical field of rotating machinery fault diagnosis and signal processing analysis, and particularly discloses a composite fault diagnosis method under a low-rotating-speed working condition based on an encoder signal. The method comprises the following steps: firstly, picking up a square wave signal y (t) of an encoder by using an acquisition card, and carrying out band-pass filtering to obtain a sinusoidal signal # imgabs0 #; an instantaneous angular velocity omega (t) is obtained through the steps of Hilbert transformation calculation and the like; performing time domain synchronous average calculation on the instantaneous angular velocity omega (t) to obtain the instantaneous angular velocity # imgabs1 # of the gear, and exposing the gear fault by identifying the time interval T of the instantaneous angular velocity # imgabs1 # of the gear matched with the theoretical fault; and subtracting the gear instantaneous angular speed # imgabs3 # from the instantaneous angular speed # imgabs2 # to obtain the bearing instantaneous angular speed # imgabs4 #, performing envelope analysis on the bearing instantaneous angular speed # imgabs4 #, and realizing the fault diagnosis of the rolling bearing by identifying the frequency matched with the theoretical fault feature of the rolling bearing in an envelope spectrum. According to the method, feature extraction of the rolling bearing and gear compound fault under the low-rotating-speed working condition is achieved, and a basis is provided for fault diagnosis in rotating mechanical equipment through encoder signals.

Description

Technical Field

[0001] The invention relates to the technical field of rotating machinery fault diagnosis and signal processing analysis, and in particular to a method for diagnosing rolling bearing and gear composite faults under low-speed working conditions based on encoder signals. Background Art

[0002] Rolling bearings and gears are key components of rotating mechanical equipment, and their health status plays an important role in the normal operation of the entire equipment. Researching effective rolling bearing and gear fault detection methods is crucial to ensure the normal operation and maintenance of mechanical equipment. At present, the method of extracting fault features of rolling bearings and gears based on vibration signals is widely used. However, vibration signals have the following shortcomings: the signal-to-noise ratio is low under low-speed conditions; the signal is easily attenuated due to the influence of the transmission path; the modulation phenomenon is serious, resulting in complex algorithms for processing signals. Therefore, the development of fault diagnosis methods based on other signal sources is an urgent problem to be solved.

[0003] In recent years, studies have found that when rolling bearings and gears fail, the transient features in the encoder signal contain rich information about the health status of the mechanical equipment. Compared with vibration signals, encoder signals have the following advantages: the signal is not affected by the transmission path and has a high signal-to-noise ratio; the signal can be directly obtained in the device with a built-in encoder without the need to install additional sensors; it has a better low-frequency response capability and is suitable for fault detection under low-speed conditions. Although the prior art involves some methods for extracting fault features of rolling bearings and gears based on encoder signals, the current methods only extract features for single faults of rolling bearings or gears, and the situation where rolling bearings and gears fail at the same time often occurs in engineering. The transient features caused by rolling bearings and gears are coupled with each other and affected by other interference components, making it difficult to directly extract fault features from the instantaneous angular velocity signal calculated by the encoder. Summary of the invention

[0004] In order to solve the problems analyzed above, the present invention provides a composite fault diagnosis method under low-speed conditions based on encoder signals. The method uses encoder signals to estimate instantaneous angular velocity, and separates rolling bearing and gear composite fault signals based on the time domain synchronous averaging method according to the key phase characteristics of the encoder. The method has the characteristics of simple algorithm and accurate fault identification.

[0005] The method comprises the following steps: S1. Data acquisition: Use the acquisition card to pick up the output signal of the encoder coaxial with the gear and rolling bearing to obtain a square wave signal ; S2, data processing: The output signal of the encoder is processed through a rectangular bandpass filter to convert the square wave signal Convert to sinusoidal signal ; S3. Signal analysis: for sinusoidal signals Perform Hilbert transform to obtain analytical signal , by analyzing the signal The polar coordinates of the instantaneous phase , then the instantaneous phase Perform differential calculation to obtain instantaneous angular velocity ; S4. Gear fault feature extraction: According to the key phase characteristics of the encoder signal, the instantaneous angular velocity Perform time domain synchronous averaging to obtain the instantaneous angular velocity of the gear including the gear period component ; S5. Bearing fault feature extraction: instantaneous angular velocity Subtract the instantaneous angular velocity of the gear Obtain the instantaneous angular velocity of the bearing including the rolling bearing fault information ; S6, gear fault diagnosis: by the instantaneous angular velocity of the gear caused by the fault The jitter time interval Theoretical failure time interval with gear Matching is performed to determine gear failure; S7, bearing fault diagnosis: bearing instantaneous angular velocity signal Envelope analysis is performed to identify rolling bearing faults by identifying frequencies in the envelope spectrum that match the theoretical rolling bearing fault characteristics.

[0006] Further, in step S2, the center frequency of the rectangular bandpass filter is , the bandwidth is ,in, is the number of encoder lines, For frequency conversion.

[0007] Further, in step S3, the Hilbert transform process is performed as follows: Calculates a sine signal The analytical signal , ; in, , represents a sinusoidal signal The Hilbert transform of the above analytical signal Get its polar coordinates, ; in, , represents the instantaneous amplitude, , represents the instantaneous phase, and then the instantaneous phase Differentiate to get the instantaneous angular velocity , ; in, dt is the sampling time interval, which is the reciprocal of the sampling frequency.

[0008] Furthermore, in step S4, the instantaneous angular velocity Perform time domain synchronous averaging processing, that is, ; in, M represents the average number of times, Indicates the theoretical failure period.

[0009] The beneficial effects of the present invention are: 1. Compared with vibration signals, encoder signals have better low-frequency response characteristics and higher signal-to-noise ratio, which is conducive to solving the fault diagnosis problem of key components of rotating machinery under low-speed conditions. In addition, encoder signals are not affected by the transmission path and can monitor the operating status of multiple key components on the coaxial axis, which is conducive to reducing sensor layout and reducing costs.

[0010] 2. The frequency demodulation method based on Hilbert transform calculates the instantaneous angular velocity of the encoder signal. Compared with the method of estimating the instantaneous angular velocity by the time elapsed method, the signal acquisition is simpler and overcomes the dependence on the high-speed counter acquisition card. In addition, it has advantages when other analog signals need to be acquired synchronously.

[0011] 3. The rotation of the gear generates a quasi-periodic signal. The rolling bearing has a non-integer number of failures per revolution due to slippage, which is a random signal. The encoder signal has a good key phase characteristic and contains the composite fault signal of the coaxial gear and the rolling bearing. The periodic signal is retained by the time domain synchronous averaging method, and the random signal is obtained by calculating the residual. In this way, the composite fault information of the rolling bearing and the gear is separated, and the composite fault feature extraction is realized. The algorithm is simple and is conducive to application in engineering practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a flow chart of the present invention; Figure 2 It is a structural schematic diagram of the gear and rolling bearing test bench of the present invention; Figure 3 is a schematic diagram of gear wear of the present invention; Figure 4 A schematic diagram of a rolling bearing outer ring failure of the present invention; Figure 5 is the original encoder signal in the embodiment of the present invention; Figure 6The signal is a signal obtained by bandpass filtering the original encoder signal according to the present invention; Figure 7 The angular position signal calculated from the encoder signal of the present invention Figure 8 is the instantaneous angular velocity diagram of the present invention; Fig. 9 It is the instantaneous angular velocity diagram after time-domain synchronous averaging of the present invention (under normal conditions); Fig.10 It is the instantaneous angular velocity diagram after time domain synchronous averaging of the present invention (under fault state); Fig.11 It is the envelope spectrum of the rolling bearing signal of the present invention (under normal state); Fig.12 It is the envelope spectrum of the rolling bearing signal of the present invention (under fault state).

[0013] Description of reference numerals: 1. Driving motor; 2. Coupling; 3. Gearbox; 4. Driving gear; 5. Driven gear; 6. Rolling bearing; 7. Encoder; 8. Magnetic powder brake. DETAILED DESCRIPTION

[0014] The present invention is further described below in conjunction with the accompanying drawings and embodiments. The embodiments described herein are only used to explain and illustrate the relevant invention and are not intended to limit the invention. For ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0015] The rotating mechanical equipment in the present invention is introduced by taking a fixed-axis gearbox and a rolling bearing as an example. The encoder is installed on a shaft coaxial with the driven gear and the rolling bearing. Figure 3 As shown in the figure, the rolling bearing with outer ring failure is as follows Figure 4 shown.

[0016] like Figure 2 As shown, the test bench is composed of a driving motor 1, a coupling 2, a driving gear 4, a driven gear 5, a coupling 2, a rolling bearing 6, an encoder 7 and a magnetic powder brake 8 connected in sequence, and the driving gear 4 and the driven gear 5 are placed in a gear box 3. The number of teeth of the driving gear 4 and the driven gear 5 of the gear box 3 are 30 and 70 respectively. The pitch diameter of the rolling bearing 6 D , rolling element diameter d 、Number of rolling elements n and contact angle α They are 34 mm, 7.97 mm, 11 and 0° respectively. In the embodiment, the output speed of the driving motor 1 is 42 rpm, and the faulty gear is the driven gear 5, with a speed of 18 rpm. Therefore, the theoretical fault cycle of the instantaneous angular velocity jitter caused by the gear fault is 3.3 s, and the theoretical fault characteristic frequency of the rolling bearing 6 is 1.29 Hz.

[0017] like Figure 1 As shown, the composite fault diagnosis method of this embodiment includes the following steps: S1. Data acquisition: Use the acquisition card to pick up the output signal of the encoder coaxial with the gear and rolling bearing to obtain a square wave signal ,like Figure 5 shown.

[0018] S2, data processing: The output signal of the encoder is processed through a rectangular bandpass filter to convert the square wave signal Convert to sinusoidal signal . Among them, the encoder line number N is 5000, and the rotation frequency is calculated based on the encoder signal f r is 18rpm, or 0.3 Hz. Therefore, the center frequency and bandwidth of the rectangular bandpass filter are N·f r =1500Hz and =750Hz; according to this parameter, the square wave signal Band-pass filtering is performed by Figure 6 It can be seen that the encoder signal becomes a sinusoidal signal .

[0019] S3. Signal analysis: for sinusoidal signals Perform Hilbert transform to obtain analytical signal , ; in, , represents a sinusoidal signal The Hilbert transform of the above analytical signal Get its polar coordinates, ; in, , represents the instantaneous amplitude, , represents the instantaneous phase, the instantaneous phase like Figure 7 As shown in the figure, it can be seen that the cumulative angular position component is a slant line, and the transient characteristics caused by the fault cannot be extracted at present; Then the instantaneous phase Differentiate to get the instantaneous angular velocity , ; in, dt is the sampling time interval, i.e. the inverse of the sampling frequency; Instantaneous angular velocity like Figure 8As shown in the figure, due to the mutual coupling between rolling bearings and gears, and the existence of other interference components, the instantaneous angular velocity The cycle matching the theoretical fault cycle of the gear or rolling bearing cannot be identified, and the fault cannot be identified. The present invention further calculates the instantaneous angular velocity of the encoder signal by a frequency demodulation method, which is conducive to separating and extracting the transient characteristics caused by the fault.

[0020] S4. Gear fault feature extraction: According to the key phase characteristics of the encoder signal, the instantaneous angular velocity Perform time domain synchronous averaging to obtain the instantaneous angular velocity of the gear including the gear period component ;Right now ; in, M represents the average number of times, Indicates the theoretical fault cycle. Due to the low speed working condition, in order to ensure a higher average number to achieve a better average effect, this embodiment selects 170s data for 50 averages. f r The theoretical failure cycle is 18 rpm. The average instantaneous angular velocity is 3.3s. like Fig.10 shown.

[0021] S5. Bearing fault feature extraction: The impact caused by rolling bearings is often smaller than that of gears and is easily obliterated. In order to improve the signal-to-noise ratio of rolling bearing signals, the instantaneous angular velocity Subtract the instantaneous angular velocity of the gear Obtain the instantaneous angular velocity of the bearing including the rolling bearing fault information ; S6, gear fault diagnosis: by the instantaneous angular velocity of the gear caused by the fault The jitter time interval Theoretical failure time interval with gear Matching is performed to determine gear failure; from Fig.10 It is easy to identify the theoretical gear failure cycle Matching time intervals T , and corresponding Fig. 9 The instantaneous angular velocity signal under normal conditions shown does not have any fault cycle Matching time intervals. This shows that the synchronous averaging method can effectively extract the gear fault characteristics. The present invention proposes a time domain synchronous averaging method based on the key phase characteristics of the encoder to achieve gear fault feature extraction and provide a basis for accurately using encoder signals to realize gear fault detection.

[0022] S7, bearing fault diagnosis: bearing instantaneous angular velocity signal Envelope analysis is performed to identify rolling bearing faults by identifying frequencies in the envelope spectrum that match the theoretical rolling bearing fault characteristics.

[0023] like Fig.12 As shown in the figure, the frequency (1.29 Hz) and its harmonic components that match the theoretical fault characteristics of rolling bearings can be clearly identified in the envelope spectrum. Fig.11 The envelope spectrum under normal conditions shown does not have a frequency component that matches the fault frequency. Therefore, when the theoretical fault characteristic component can be identified in the envelope spectrum, it can be determined that the rolling bearing is faulty, and the rolling bearing fault diagnosis based on the encoder signal can be realized.

[0024] The main fault diagnosis process of the present invention is to recommend the selection of appropriate center frequency and bandwidth, which can convert the encoder signal (square wave signal) into a sine wave signal, and then accurately calculate the instantaneous angular velocity. Combined with the encoder key phase characteristics, the synchronous averaging algorithm can be accurately used to highlight the instantaneous angular velocity changes caused by gear faults, thereby identifying faults. In addition, the gear and bearing signals have the characteristics of first-order cyclostationarity and second-order cyclostationarity respectively, and the demodulated signal still retains the encoder key phase characteristics. A method of obtaining a signal rich in bearing fault information by subtracting a periodic signal from the original signal is proposed, and the rolling bearing fault signal is separated and identified. This method is not affected by the transmission path, can solve the fault diagnosis problem of different objects under low-speed conditions, and is conducive to reducing detection costs.

[0025] Through the above embodiments, the present invention can realize rolling bearing and gear compound fault diagnosis based on encoder signals, and provide a basis for the application of encoder signals in rotating machinery operating status monitoring. Finally, it should be noted that the above embodiments are only used to illustrate the core idea of ​​the present invention rather than to limit the technical method of the present invention. Although the present invention is described in detail with reference to the above embodiments, a person skilled in the art can modify or replace the present invention, and any modification or partial replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A composite fault diagnosis method under low speed conditions based on encoder signals, characterized in that: The following steps are involved: S1. Data acquisition: Use the acquisition card to pick up the output signal of the encoder coaxial with the gear and rolling bearing to obtain a square wave signal ; S2, data processing: The output signal of the encoder is processed through a rectangular bandpass filter to convert the square wave signal Convert to sinusoidal signal ; S3. Signal analysis: for sinusoidal signals Perform Hilbert transform to obtain analytical signal , by analyzing the signal The polar coordinates of the instantaneous phase , then the instantaneous phase Perform differential calculation to obtain instantaneous angular velocity ; S4. Gear fault feature extraction: According to the key phase characteristics of the encoder signal, the instantaneous angular velocity Perform time domain synchronous averaging to obtain the instantaneous angular velocity of the gear including the gear period component ; S5. Bearing fault feature extraction: instantaneous angular velocity Subtract the instantaneous angular velocity of the gear Obtain the instantaneous angular velocity of the bearing including the rolling bearing fault information ; S6, gear fault diagnosis: by the instantaneous angular velocity of the gear caused by the fault The jitter time interval Theoretical failure time interval with gear Matching is performed to determine gear failure; S7, bearing fault diagnosis: bearing instantaneous angular velocity signal Envelope analysis is performed to identify rolling bearing faults by identifying frequencies in the envelope spectrum that match the theoretical rolling bearing fault characteristics.

2. A composite fault diagnosis method under low speed conditions based on encoder signals according to claim 1, characterized in that: In step S2, the center frequency of the rectangular bandpass filter is , the bandwidth is ,in, is the number of encoder lines, For frequency conversion.

3. A composite fault diagnosis method under low speed conditions based on encoder signals according to claim 2, characterized in that: In step S3, the Hilbert transform process is as follows: Calculates a sine signal The analytical signal , ； in, , represents a sinusoidal signal The Hilbert transform of the above analytical signal Get its polar coordinates, ； in, , represents the instantaneous amplitude, , represents the instantaneous phase, and then the instantaneous phase Differentiate to get the instantaneous angular velocity , ； in, dt is the sampling time interval, which is the reciprocal of the sampling frequency.

4. A composite fault diagnosis method under low speed conditions based on encoder signals according to claim 3, characterized in that: In step S4, the instantaneous angular velocity Perform time domain synchronous averaging processing, that is, ； in, M represents the average number of times, Indicates the theoretical failure cycle.

Images