Wenxiu Lu

Condition monitoring and fault diagnostics of wind turbines

Condition monitoring and fault diagnostics is important for wind turbines to ensure safety and reliability. However, the application in wind turbine is still in initial stage. This paper analyses the frequent faults in wind turbines and determines the sensor configuration of a wind turbine. Architecture of condition monitoring and fault diagnostics system is built to monitor a wind farm. Methodology and algorithms based on vibration, noise and acoustic emission signals have been discussed to diagnose the fault of wind turbines.

Application of AE techniques for the detection of wind turbine using Hilbert-Huang transform

This paper describes acoustic emission(AE) techniques based on Hilbert-Huang transform(HHT) that were recently exercised to characterise the AE signals released from the wind turbine bearing. Acoustic emission that detects elastic stress waves within a structure failure is capable of online monitoring and very sensitive to the fault development. AE wave is a non-stationary stochastic signal. Hilbert-Huang transform is applicable to nonlinear and non-stationary processes. With the Hilbert-Huang transform, instantaneous frequencies based on local properties of the signal can be got as functions of time and energy designated as the Hilbert spectrum that give sharp identifications of imbedded structures. We analyzed the AE signals recording from the wind turbine bearing test using Hilbert-Huang transform. The results show that the AE in the wind turbine bearing can be described in terms of features like frequency and energy, and inferences can be made about kinds of damage processes taking place in the bearing. And thus the HHT analysis method will has a good potential for the acoustic emission signal processing in the field of wind turbines.

Experimental Investigation of Pedestal Looseness in a Rotor-Bearing System

An experimental setup of rotor-bearing system is installed and vibration characteristics of the system with pedestal looseness are investigated. The pretightening bolt between the bearing house and pedestal is adjusted to simulate the pedestal looseness fault. The vibration waveforms, spectra and orbits are used to analyze the nonlinear response of the system with pedestal looseness. Different parameters, including speed, looseness gap, imbalance mass and disk position are changed to observe the nonlinear vibration characteristics. The experiments show that the system motion generally contains the 1/2X fractional harmonic component and multiple harmonic components such as 2X, 3X, etc. Under some special conditions, the pedestal looseness occurs intermittently, that is, occurs in some periods and doesn’t in other periods.

Shaft crack identification based on vibration and AE signals

The shaft crack is one of the main serious malfunctions that often occur in rotating machinery. However, it is difficult to locate the crack and determine the depth of the crack. In this paper, the acoustic emission (AE) signal and vibration response are used to diagnose the crack. The wavelet transform is applied to AE signal to decompose into a series of time-domain signals, each of which covers a specific octave frequency band. Then an improved union method based on threshold and cross-correlation method is applied to detect the location of the shaft crack. T he finite element method is used to build the model of the crack ed rotor, and the crack depth is identified by comparing the vibr ation response of experiment and simulation. The experimental results show that the AE signal is effective and convenient to locate the shaft crack, and the vibration signal is feasibl e to determine the depth of shaft crack.

Fault Diagnostics Based on Pattern Spectrum Entropy and Proximal Support Vector Machine

Based on pattern spectrum entropy and proximal support vector machine (PSVM), a motor rolling bearing fault diagnosis method is proposed in this paper. It is very difficult to filter the fault vibration signals from the strong noise background because the roller bearing fault diagnosis is a problem of multi-class classification of inner ring fault, outer ring fault and ball fault. Firstly, vibration signals are processed by the pattern spectrum. Secondly, the morphological pattern spectrum entropy, and pattern spectrum values are utilized to identify the fault features of input parameters of PSVM classifiers. The experiment results demonstrate that the pattern spectrum quantifies various aspects of the shape-size content of a signal, and PSVM costs a little time and has better efficiency than the standard SVM.

Lateral Vibration of Hydroelectric Generating Set with Different Supporting Condition of Thrust Pad

The variations of the supporting condition, which change the stiffness of tilting pad thrust bearing, may alter the dynamic behavior of the rotor system. The effects of supporting condition of thrust pad on the lateral vibration of a hydroelectric generating set are investigated in this paper. The action of a thrust bearing is described as moments acting on the thrust collar, and the tilting stiffness coefficients of thrust bearing are calculated. A model based on typical beam finite element method is established to calculate the dynamic response, and the effects of supporting conditions such as elastic oil tank support, different heights of the thrust pads with rigid support are discussed. The results reveal that the influence of thrust bearing is small when the elastic oil tanks work normally. When the supporting conditions turn to be rigid due to the oil leakage, the differences of thrust pad heights have evident influence on the load distribution of the thrust pads; while the effects on the tilting stiffness of the thrust bearing and the amplitude of the lateral shaft vibration is small when the maximum load on thrust pads is smaller than the allowable value.

Internet/Intranet Based Remote Condition Monitoring and Fault Diagnosis Scheme and System for Steam Turboset

The steam turboset is the key equipment of the electric power system. Thus, it is very important and necessary to monitor and diagnose the running condition and the faults of the steam turboset for the safe and normal running of the electric power system. In this paper, the Internet/Intranet based remote condition monitoring and fault diagnosis scheme is proposed. The corresponding technique and methods are discussed in detail. And a real application system is developed for the 300MW steam turboset. In this scheme, the system is built on the Internet/Intranet and the Client/Server construction and Web/Server model are adopted. The proposed scheme can guarantee real-time data acquisition and on-line condition analysis simultaneously. And especially, the remote condition monitoring and fault diagnosis can be implemented effectively. The developed system has been installed in a power plant of China. And the plant has obtained great economic benefits from it.

Condition Monitoring of Shaft Crack with Acoustic Emission

A cracked shaft is one of the most catastrophic failures in any rotating machine such as the axle of a railway vehicle with increasing speed which is usually working in a hazardous environment. This paper describes an experimental investigation in which Acoustic Emission technology was applied to monitor the presence of a naturally fatigued crack in an operating high speed shaft. A test rig was built for generating natural rotating fatigue crack on a shaft. The experiments show that the classical AE parameters, such as energy, hit, and amplitude can indicate the cracked shaft failure process.

Genetic Algorithms Based Rubbing Location Identification in a Rotor System

The rubbing fault is a very serious and frequent malfunction in rotating machinery, and the determination of the rubbing location is very important in actual fault diagnosis. In this paper, a method based on genetic algorithms to detect the rubbing location is presented. The finite element model of the rubbing rotor is established with the rubbing location, the stator stiffness, the clearance between stator and rotor, the damping coefficient and the friction coefficient as the fault parameters, and the rubbing location determination is transferred into the parameter identification problem. The genetic algorithm is then utilized to search the solution. Using genetic algorithms avoids some of the weaknesses of traditional parameter identification methods such as local minimum problem in nonlinear system identification. The experimental results suggest that the rubbing location can be effectively determined when the rubbing occurs.

Influence of bearing pre-load coefficient on shaft vibration and pad temperature in a hydroturbine generator unit: A case study

From data collected by an online condition monitoring and fault diagnosis system, a higher pad temperature at the upper guide bearing in a pumped storage power generator unit installed in Guangdong province(GPSPS), China, was found. And also a relatively intensive shaft vibration occurred at the lower guide bearing. By calculating the Reynolds equation and viscosity-temperature equation of the lubricant, a curve between the pre-load coefficient and the increment of pad temperature is obtained, which shows that the larger, the pre-load coefficient, the bigger, the increment of pad temperature. For a practical unit in GPSPS, by employing Transfer matrix method and Wilson-θ method to analyze shaft vibration at different pre-load coefficients of the whole bearing or pad pair bearings, the results show that the larger the pre-load coefficient is, the smaller the vibration amplitude is, the shorter the time for vibration to become steady is. And an uneven pre-load coefficient of the pad pair bearings will cause shaft orbit from a circle to an ellipse whose long axes is at the direction of the pad pair with the lowest pre-load coefficient. Finally, reasons of higher pad temperature of the upper guide bearing and larger shaft vibration at the lower guide bearing are due to the inconsistent relation of bearing assembling clearance or pre-load coefficient of the upper and lower guide bearing, and also due to the too small, uneven pre-load coefficient of pad pair bearings. After a scheme for adjusting the bearing clearance is given, data measured show that the analysis and simulation methods are correct and the adjustment scheme to the assembling clearance of the upper and lower guide bearings is feasible and can be used to guide the field maintenance conveniently.