Shaft end vibration monitoring method for running state of low-speed variable-load wind driven generator

Abstract

The invention relates to the technical field of generator monitoring, and discloses a low-speed variable-load wind driven generator operation state shaft end vibration monitoring method, which is characterized in that when a wind turbine generator blade and a main shaft bearing break down, a fan main shaft can generate a corresponding vibration signal in a rotation process; the ORS acquisition device is arranged at the shaft end of a main shaft of the wind driven generator or in the radial direction of a main bearing, acquires vibration signals of the main shaft of the wind driven generator through an MEMS (Micro Electro Mechanical System) accelerometer in the ORS acquisition device and sends the vibration signals to the upper computer in a Bluetooth wireless transmission mode, and the upper computer displays and processes the vibration signals; according to the equal-angle resampling method, non-stationary signals sampled at equal time intervals can be converted into quasi-stationary signals sampled at equal-angle intervals, and fault feature extraction of the non-stationary signals can be achieved. Rotating speed and vibration signals can be measured through a single ORS sensor device, and state monitoring of important components such as a fan main shaft bearing and blades is achieved by fusing the rotating speed and the vibration signals.

Description

Technical Field

[0001] The invention relates to the technical field of generator monitoring, and in particular to a method for monitoring shaft end vibration in the operating state of a low-speed variable-load wind turbine generator. Background Art

[0002] At present, the condition monitoring of wind turbines is mainly completed through vibration analysis, acoustic emission monitoring, current signal monitoring and other means. In comparison, vibration analysis has the advantages of non-invasiveness, low cost, and good real-time performance, and is still the mainstream means of condition monitoring and fault diagnosis of wind turbine transmission systems. The conventional monitoring idea is to obtain wide-band vibration information by arranging acceleration sensors in the bearing base, gearbox housing, generator housing and other parts, and combine advanced signal processing algorithms to achieve effective extraction of fault characteristics. It is worth noting that the speed of large wind turbines is very slow, and the main shaft rotation frequency at the low-speed end is often less than 0.3Hz; in addition, affected by heavy loads, irregular wind force and the asymmetric structure of their own blades, wind turbines have both low speed and non-stationary characteristics, which makes many technologies that perform well under high speed and medium load conditions unreliable in wind turbine main shaft bearing fault detection.

[0003] Domestic and foreign scholars have conducted a lot of research on the condition monitoring of wind turbines, and a variety of measurement methods have emerged. Methods based on vibration measurement and acoustic measurement have a wide measurement range and convenient data processing. Ulriksen et al. proposed a damage location method based on extended random dynamic damage location vector, which realized the location of cracks in wind turbine blades and was able to preliminarily identify the size of the damage; Fremmelev et al. combined power spectral density and principal component analysis to extract features from vibration signals, and used Mahalanobis distance as a damage indicator to monitor the health of blades; Beale et al. installed speakers inside the blades and used microphone arrays and beamforming methods to generate acoustic images. The acoustic images can reflect the specific damage location of the blades and their degree of damage. Chen et al. applied the stator current analysis method based on the dual spectrum of modulated signals to the bearing fault diagnosis of doubly fed wind turbines, proposed overlapping segmentation to improve the calculation accuracy under limited data, and experimentally proved that faults can be detected by identifying the spectral components of the fundamental frequency and characteristic frequency.

[0004] The current monitoring methods often place sensors on the bearing base, gearbox housing, generator housing and other parts, which cannot directly measure the vibration signal of the transmission shaft, and the collected signal quality is poor. Under the influence of random wind speed, the speed of the wind turbine changes greatly, which is not conducive to signal monitoring. The status signals of multiple components are coupled with each other. If multiple sets of sensor signals are used, the correlation between them is poor, the cost is high, the installation is difficult, and the system robustness is relatively poor, making it difficult to evaluate the operating status; in view of this, we propose a method for monitoring the shaft end vibration of low-speed variable-load wind turbines in operation. Summary of the invention

[0005] The object of the present invention is to provide a method for monitoring the shaft end vibration of a low-speed variable-load wind turbine in operation, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for monitoring the shaft end vibration of a low-speed variable-load wind turbine in operation, the monitoring method comprising the following steps:

[0007] S1, ORS signal acquisition device collects three-dimensional acceleration;

[0008] S2, passing the collected three-axis acceleration signals of the main shaft through a bandpass filter to extract the vibration signals in the relevant frequency range and filter the noise signals;

[0009] S3, the time point of equal-angle resampling can be determined by the obtained speed curve;

[0010] S4, using piecewise linear interpolation to remap the irregular sampling signal to a uniformly spaced angular domain to obtain a quasi-stationary signal;

[0011] S5, performing Hilbert transform on the resampled angular domain stationary signal to obtain an envelope signal in the time domain;

[0012] S6. Obtain the envelope spectrum using fast Fourier transform;

[0013] S7, extracting the fault characteristics of the fan from the envelope spectrum;

[0014] S8. Test experiment simulating bearing failure and blade deflection.

[0015] Optionally, the ORS signal acquisition device is installed at the shaft end of the wind turbine generator set, and the ORS signal acquisition device acquires the vibration signal of the wind turbine main shaft through a MEMS accelerometer.

[0016] Optionally, the ORS signal acquisition device sends the collected signals to a host computer via Bluetooth wireless transmission.

[0017] Optionally, the vibration signal after bandpass filtering in S2 indirectly estimates the wind turbine speed through the zero crossing point, and the phenomenon caused by the change of gravity acceleration sensed by the ORS sensor during the rotation process is used to estimate the speed of the wind turbine using the periodically changing signal.

[0018] Optionally, in S2, the generator speed is controlled by controlling the load of the wind turbine generator.

[0019] Optionally, the S8 includes:

[0020] A bearing with an outer ring pitting fault was installed on a wind turbine, and an ORS sensor was installed on the end of the main shaft to collect its three-axis acceleration signals at different speeds. According to the instantaneous speed estimated by the ORS sensor, the acceleration signal was resampled in the angular domain, and then the envelope spectrum was calculated and the outer ring fault signal was detected.

[0021] Optionally, the S8 further includes:

[0022] By adjusting the deflection angle of a single blade, the failure of blade imbalance caused by pitch failure is simulated;

[0023] Estimate the average speed of wind turbines through ORS sensors;

[0024] At the same wind speed, the fan speed increases with the increase of load; under the same load, the fan speed under blade imbalance failure is lower than the speed when the blades are balanced.

[0025] Compared with the prior art, the present invention provides a method for monitoring the shaft end vibration of a low-speed variable-load wind turbine in operation, which has the following beneficial effects:

[0026] The method for monitoring the shaft end vibration of the low-speed variable-load wind turbine in operation can convert the non-stationary signal sampled at equal time intervals into a quasi-stationary signal sampled at equal angle intervals by equal angle resampling method, and can realize the fault feature extraction of the non-stationary signal. The speed and vibration signal can be measured by a single ORS sensor device, and the status monitoring of the wind turbine main shaft bearing and blade important components can be realized by fusing the two. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a schematic diagram of the process of the present invention;

[0028] Figure 2 It is a schematic diagram of the three-axis acceleration signal collected by the ORS sensor of the present invention, the low-frequency spectrum of the three-axis acceleration signal, and the broadband spectrum of the three-axis acceleration signal;

[0029] Figure 3 It is a schematic diagram of the angular domain envelope spectrum obtained by the ORS sensor for bearing faults at different speeds according to the present invention;

[0030] Figure 4 This is a schematic diagram of the change in fan speed with load at the same wind speed when simulating a blade imbalance fault in the present invention. DETAILED DESCRIPTION

[0031] 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.

[0032] like Figure 1-Figure 4 As shown, the present invention provides a technical solution: a method for monitoring the shaft end vibration of a low-speed variable-load wind turbine in operation, the monitoring method comprising the following steps:

[0033] S1. ORS signal acquisition device collects three-dimensional acceleration. The ORS signal acquisition device is installed at the shaft end of the wind turbine generator set. The ORS signal acquisition device collects the vibration signal of the wind turbine main shaft through the MEMS accelerometer. The ORS signal acquisition device sends the collected signal to the host computer through Bluetooth wireless transmission.

[0034] S2, passing the collected three-axis acceleration signals of the main shaft through a bandpass filter to extract the vibration signals in the relevant frequency range and filter the noise signals;

[0035] The vibration signal after bandpass filtering indirectly estimates the wind turbine speed through the zero crossing point. The ORS sensor senses the phenomenon caused by the change in gravity acceleration during the rotation process, and uses this periodically changing signal to estimate the speed of the wind turbine. The speed of the generator is controlled by controlling the load of the wind turbine.

[0036] S3, the time point of equal angle resampling can be determined by the obtained speed curve; Figure 2 As shown in the time domain waveform, the U-axis and V-axis signals show obvious periodic changes; the U-axis, V-axis and W-axis are the three-axis acceleration signals of the main axis.

[0037] The ORS sensor estimates the wind turbine speed to be 0.95 Hz when the load is 25 ohms.

[0038] S4. Use piecewise linear interpolation to remap the irregular sampling signal to the evenly spaced angular domain to obtain a quasi-stationary signal.

[0039] S5. Perform Hilbert transform on the resampled angular domain stationary signal to obtain an envelope signal in the time domain.

[0040] S6. Use fast Fourier transform to obtain the envelope spectrum.

[0041] S7. Extract the fault characteristics of the fan from the envelope spectrum.

[0042] S8. Test experiment simulating bearing failure and blade deflection.

[0043] A bearing with an outer ring pitting fault was installed on a wind turbine, and an ORS sensor was installed on the end of the main shaft to collect its three-axis acceleration signals at different speeds. According to the instantaneous speed estimated by the ORS sensor, the acceleration signal was resampled in the angular domain, and then the envelope spectrum was calculated and the outer ring fault signal was detected.

[0044] By adjusting the deflection angle of a single blade, the blade imbalance caused by the pitch failure is simulated; the average speed of the wind turbine is estimated by the ORS sensor; at the same wind speed, the wind turbine speed increases with the increase of load; at the same load, the wind turbine speed under the blade imbalance failure is lower than the speed when the blade is balanced. Figure 3 As shown, Figure 3 The angular domain envelope spectrum of (a) is 1.17 Hz, the angular domain envelope spectrum of (b) is 0.98 Hz, the angular domain envelope spectrum of (c) is 0.73 Hz, and the angular domain envelope spectrum of (d) is 0.45 Hz.

[0045] As an application of this embodiment:

[0046] Since the fan is in a low-speed non-stationary working condition, the collected signal is a non-stationary signal, so it is difficult to extract the fault characteristics of the fan by conventional signal processing methods. The equal-angle resampling method can convert the non-stationary signal sampled at equal time intervals into a quasi-stationary signal sampled at equal angle intervals, which can realize the extraction of fault characteristics of non-stationary signals. The speed and vibration signals can be measured by a single ORS sensor device, and by fusing the two, the status monitoring of important components of the fan main shaft bearing and blades can be realized.

[0047] The above generally describes the present invention in detail, but it is obvious to a person skilled in the art that some modifications or improvements can be made to the present invention. Therefore, modifications or improvements that do not depart from the spirit of the present invention are within the scope of protection of the present invention.

Claims

1. A method for monitoring shaft end vibration in the running state of a low-speed variable-load wind turbine, characterized in that: The monitoring method comprises the following steps: S1, ORS signal acquisition device collects three-dimensional acceleration; S2, passing the collected three-axis acceleration signals of the main shaft through a bandpass filter to extract the vibration signals in the relevant frequency range and filter the noise signals; S3, the time point of equal-angle resampling can be determined by the obtained speed curve; S4, using piecewise linear interpolation to remap the irregular sampling signal to a uniformly spaced angular domain to obtain a quasi-stationary signal; S5, performing Hilbert transform on the resampled angular domain stationary signal to obtain an envelope signal in the time domain; S6. Obtain the envelope spectrum using fast Fourier transform; S7, extracting the fault characteristics of the fan from the envelope spectrum; S8. Test experiment simulating bearing failure and blade deflection.

2. A method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 1, characterized in that: The ORS signal acquisition device is installed at the shaft end of the wind turbine generator set, and the ORS signal acquisition device acquires the vibration signal of the wind turbine main shaft through a MEMS accelerometer.

3. A method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 1, characterized in that: The ORS signal acquisition device sends the collected signals to the host computer via Bluetooth wireless transmission.

4. A method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 1, characterized in that: The vibration signal after bandpass filtering in S2 indirectly estimates the wind turbine speed through the zero crossing point. The ORS sensor senses the phenomenon caused by the change in gravity acceleration during the rotation process, and uses the periodically changing signal to estimate the speed of the wind turbine.

5. The method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 1, characterized in that: In S2, the generator speed is controlled by controlling the load of the wind turbine generator.

6. A method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 1, characterized in that: The S8 includes: A bearing with an outer ring pitting fault was installed on a wind turbine, and an ORS sensor was installed on the end of the main shaft to collect its three-axis acceleration signals at different speeds. According to the instantaneous speed estimated by the ORS sensor, the acceleration signal was resampled in the angular domain, and then the envelope spectrum was calculated and the outer ring fault signal was detected.

7. A method for monitoring shaft end vibration of a low-speed variable-load wind turbine in operation according to claim 6, characterized in that: The S8 further comprises: By adjusting the deflection angle of a single blade, the failure of blade imbalance caused by pitch failure is simulated; Estimate the average speed of wind turbines through ORS sensors; At the same wind speed, the fan speed increases with the increase of load; at the same load, the fan speed under blade imbalance failure is lower than the speed when the blades are balanced.

Images