Yimin Shao

Fatigue life prediction of rear axle using time series model

The rear axle is one of the key parts of the automobile, lots failure of rear axle resulted from fatigue failure of the spiral bevel gears. A new method is proposed to solve the problem of accurately predicting the fatigue life of spiral bevel gears in rear axle. The method uses the recurrence tracing and difference method to improve the autoregressive moving average (ARMA) model prediction accuracy, which uses variables determined from on-line measurements to characterize the state of the deterioration rear axle. The experimental results show the proposed method has relatively high prediction accuracy.

Fault Prognosis and Diagnosis of an Automotive Rear Axle Gear Using a RBF-BP Neural Network

The rear axle gear is one of the key parts of transmission system for automobiles. Its healthy state directly influences the security and reliability of the automotives. However, non-stationary and nonlinear characteristics of gear vibration due to load and speed fluctuations, makes it difficult to detect and diagnosis the faults from the transmission gear. To solve this problem a fault prognosis and diagnosis method based on a combination of radial basis function(RBF) and back-propagation (BP) neural networks is proposed in this paper. Firstly, a moving average pretreatment is used to suppress the time series fluctuation of vibration characteristic parameter tie series and reduce the interference of random noise. Then, the RBF network is applied to the pretreated parameter sequences for fault prognosis. Furthermore, based on self-learning ability of neural networks, characteristic parameters for different common faults are learned by a BP network. Then the trained BP neural network is utilized for fault diagnosis of the rear axle gear. The results show that the proposed method has a good performance in prognosing and diagnosing different faults from the rear axle gear.

A simulation study of defects in a rolling element bearing using FEA

An internal impact usually happens when there is a small defect in one part of rolling bearings. The Fault signal from this impact is always masked by different noises such as strong vibrations from other parts and the random noise of instrumentation, which makes it difficult to extract an accurate feature signal for early fault diagnosis. In this paper, a simulation study is conducted using the method of finite element analysis (FEA) to understand the vibration characteristics from the small impact. The vibration responses have been modelled based on a typical bearing assembly. Common faults including outer ring defect, inner ring defect and rolling ball defect are simulated and their vibration responses are compared between different faults and at different locations in the bearing housing. The results obtained have shown that under the same defect size, the vibration from the outer ring is the highest whereas that from the rolling ball is the smallest. In addition the vibration close to the mounting hole attenuates considerably compared to that close to outer ring. These findings provide fundamental information to place vibration sensors and to analyse vibration signals.

Gear Fault Signal Detection based on an Adaptive Fractional Fourier Transform Filter

Vibration-based fault diagnosis is widely used for gearbox monitoring. However, it often needs considerable effort to extract effective diagnostic feature signal from noisy vibration signals because of rich signal components contained in a complex gear transmission system. In this paper, an adaptive fractional Fourier transform filter is proposed to suppress noise in gear vibration signals and hence to highlight signal components originated from gear fault dynamic characteristics. The approach relies on the use of adaptive filters in the fractional Fourier transform domain with the optimised fractional transform order and the filter parameters, while the transform orders are selected when the signal have the highest energy gathering and the filter parameters are determined by evolutionary rules. The results from the simulation and experiments have verified the performance of the proposed algorithm in extracting the gear failure signal components from the noisy signals based on a multistage gearbox system.

Application of an improved dynamic time synchronous averaging method for fault diagnosis in conditions of speed fluctuation and no tachometer

Time synchronous averaging has been widely used for machinery fault diagnosis. However, it cannot reveal signal characteristics accurately in conditions of speed fluctuation and no tachometer due to the phase accumulation error. In this paper, an improved dynamic-time synchronous averaging method is proposed to extract the periodic feature signal from the fluctuated vibration signal for fault detection when no tachometer signal is available. In this method, empirical mode decomposition, dynamic time warping, and time synchronous averaging are performed on gear vibration signals to detect fault characteristic information. First, empirical mode decomposition is performed on the vibration signal and a series of intrinsic mode functions are produced. The sensitive intrinsic mode functions providing fault-related information are selected and reconstructed and the corresponding envelop signals are equal-space intercepted. Then, the phase accumulation error among the envelop signal segments is estimated by the dynamic time warping, which is further used to compensate the phase accumulation error between the intrinsic mode function segments of the reconstructed signal. Finally, the compensated intrinsic mode function segments are averaged to obtain the feature signal. Simulation analysis shows the advantages of the proposed method in extracting faulty feature signal from speed fluctuation signal without tachometer and identifying gear fault. Experiments with both normal and faulty gear were conducted and the vibration signals were captured. The proposed method is applied to identify the gear damage and the diagnosis results demonstrate its superiority than other methods.

Fault diagnosis system based on smart bearing

According to statistics, a lot of the rotating machinery faults are caused by the bearings, so the smart bearing technique is important for ensuring safety of them. For the smart bearing, a representative definition is that sensing devices of the different use are integrated into the traditional bearing in order to realize self-diagnosis. For condition of the variable speed, variable load and heavy load, the diagnostic technology cannot satisfy requirement of the fault feature extraction at present. This paper presents a new smart bearing of the multi-parameters including of two vibration acceleration sensing devices, two speed sensing devices, and two temperature sensing devices. In addition, the heavy noise can be decreased for extraction of the weak fault signals by the embedded integrated mode of bearing and sensor.

Planetary gear fault diagnosis based on instantaneous angular speed analysis

Planetary Gears (PG) are widely used in many important transmission systems such as helicopters and wind turbines due to its advantages of high power-weight ratio, self-centering and high transmission ratio. Vibration based condition monitoring of PG has received extensive researches for ensuring safe operations of these critical systems. However, due to the moving mesh gears and noise influences, the diagnostics of planet gear faults by conventional vibration measurements needs intensive signal processing but provides less satisfactory performance. This study investigates Instantaneous Angular Speed (IAS) based diagnostics which associates more directly with gear dynamics and is not influenced by the moving mesh gears. A pure torsional dynamic model of a PG is developed to gain the characteristics of IAS under different fault cases. Then experiments are performed to evaluate this IAS based diagnostics. Particularly, IAS signatures obtained by demodulating the frequency modulated pulse trains produced by two in-house made encoder wheels mounted at both the input and output of the PG. In addition, order spectrum analysis is applied to IAS signals to highlight fault components. IAS order spectra exhibit clear changes in the spectral amplitudes associating with different fault frequencies, showing consistent and efficient diagnostics. Besides, both the measurement system and signal processing computation for IAS based monitoring are more cost-effective and easier to be implemented online, compared with conventional vibration based methods.

Detection of bearing fault using hybrid digital filter

The signal-to-noise (S/N) ratio is lower and the impulse signal of bearing is difficult to extract when the bearing is operating under heavy environmental conditions. For this reason, a new method is proposed for improving the signal to noise ratio from measurements of bearing vibration by the hybrid digital filter (HDF). Experimental result has shown that these techniques can be made more effective only after HDF has reduced the background noise from the diagnostic signals.

A new algorithm of time domain synchronous averaging analysis

The speed of rotary machine varies with the fluctuation of voltage and load. And the method of period-based time domain synchronous averaging can not process the case of fluctuation of rotation speed because of the phase errors accumulation from the errors of round-off and frequency estimation. In order to solve the problem, a modified method of period-based time domain synchronous averaging is proposed in this paper. The idea is to get the frequency of per rotation by using FFT-FS (fast Fourier transform-Fourier series). This method can improve the effectiveness of the phase error accumulation. Simulation and experiment results have shown that the new algorithm is excellent on extracting the periodic of signal from speed fluctuation signal.