Niaoqing Hu

A bearing fault diagnosis method based on the low-dimensional compressed vibration signal

The traditional bearing fault diagnosis method is achieved often by sampling the bearing vibration data under the Shannon sampling theorem. Then, the information of the bearing state can be extracted from the vibration data, which will be used in fault diagnosis. A long-term and continuous monitoring needs to sample and store large amounts of raw vibration signals, which will burden the data storage and transmission greatly. For this problem, a new bearing fault diagnosis method based on compressed sensing is presented, which just needs to sample and store a small amount of compressed observation data and uses these data directly to achieve the fault diagnosis. First, several over-complete dictionaries are trained by dictionary learning method using the historical operating data of the bearings. Each of these dictionaries can be effective in signal sparse decomposition for a particular state, while the signals corresponding to other states cannot be decomposed sparsely. According to this difference, the bearing states can be identified finally. The fault diagnosis results of the proposed method with different parameters are analyzed. The effectiveness of the method is validated by experimental tests.

Detection and diagnosis of motor stator faults using electric signals from variable speed drives

Motor current signature analysis has been investigated widely for diagnosing faults of induction motors. However, most of these studies are based on open loop drives. This paper examines the performance of diagnosing motor stator faults under both open and closed loop operation modes. It examines the effectiveness of conventional diagnosis features in both motor current and voltage signals using spectrum analysis. Evaluation results show that the stator fault causes an increase in the sideband amplitude of motor current signature only when the motor is under the open loop control. However, the increase in sidebands can be observed in both the current and voltage signals under the sensorless control mode, showing that it is more promising in diagnosing the stator faults under the sensorless control operation.

Modeling the Relationship between Vibration Features and Condition Parameters Using Relevance Vector Machines for Health Monitoring of Rolling Element Bearings under Varying Operation Conditions

Rotational speed and load usually change when rotating machinery works. Both this kind of changing operational conditions and machine fault could make the mechanical vibration characteristics change. Therefore, effective health monitoring method for rotating machinery must be able to adjust during the change of operational conditions. This paper presents an adaptive threshold model for the health monitoring of bearings under changing operational conditions. Relevance vector machines (RVMs) are used for regression of the relationships between the adaptive parameters of the threshold model and the statistical characteristics of vibration features. The adaptive threshold model is constructed based on these relationships. The health status of bearings can be indicated via detecting whether vibration features exceed the adaptive threshold. This method is validated on bearings running at changing speeds. The monitoring results show that this method is effective as long as the rotational speed is higher than a relative small value.

Application of Phase Space Warping on Damage Tracking for Bearing Fault

Nowadays, the significance of keeping equipment function properly each time is obvious. If equipment fails during its use, it may have disastrous consequences. Estimating remaining useful life (RUL) of equipment is a key to prevent such calamities, improve its reliability, provide security and reduce unnecessary maintenance and operational cost. The evolution and tracking of damage is the foundation of RUL predicting, and also is one of the most important content of mechanical fault diagnosis. Slow-time variable process of mechanical damage would lead the phase space reconstructed by fast-time variable vibrate signals warping. Search the dynamics characteristic law of damage evolution analysis in the phase space, and build the relationship between fast-time variable signals and slow-time variable damage, and then damage evolution tracking is possible. To validate the theory, simulation model of bearing damage evolution is built, the outer-race fault evolution signals is obtained, and the trend of evolution of degradation of bearing fault is described with Phase Space Warping (PSW) theory and Smooth Orthogonal Decomposition (SOD). The results proved the feasibility of the methodology of PSW in damage evolution tracking.

Crack Level Estimation Approach for Planetary Gear Sets Based on Simulation Signal and GRA

The planetary gearbox is a critical mechanism in helicopter transmission systems. Tooth failures in planetary gear sets will cause great risk to helicopter operations. A crack level estimation methodology has been devised in this paper by integrating a physical model for simulation signal generation and a grey relational analysis (GRA) algorithm for damage level estimation. The proposed method was calibrated firstly with fault seeded test data and then validated with the data of other tests from a helicopter transmission test rig. The estimation results of test data coincide with the actual test records, showing the effectiveness and accuracy of the method in providing a novel way to hybrid model based methods and signal analysis methods for more accurate health monitoring and condition prediction.

Enhancement detection of characteristic signal using stochastic resonance by adding a harmonic excitation

For a bistable nonlinear system, deterministic and stochastic excitations play equivalent roles in promotion of chaos according to qualitative results of Melnikov theory. When a bistable system maintains the state of stochastic resonance (SR), the output of system is chaotic, and the most effective spectral shape is obtained when the output power is distributed closet to the frequency of the Melnikov scales peak. In classical SR, improvement of the signal-to-noise ratio (SNR) is achieved by increasing the noise intensity, but this approach may be unwieldy. Instead of it in this paper, the more effective SNR enhancement is achieved by adding a harmonic excitation with frequency based on the systems Melnikov scale factor to the system while the noise is left unchanged. The effectiveness of this method is confirmed and replicated by numerical simulations. Combined with the strategy of scale transform, the method cab be used to detect weak periodic signal with arbitrary frequency buried in the heavy noise. At last, the method for enhancement detection of machinery fault characteristic signal is discussed via a case data.

A condition indicator performance assessment method based on signal detection theory

Condition indicators are significant elements in condition monitoring systems. A condition indicator (CI) performance assessment method would help users to select effective CIs and improve fault diagnosis performance of condition monitoring system (CMS). Investigating of measurement influence factors shows that CIs obey normal distribution. Taking advantage of signal detection theory, a CI assessment method is proposed; Discriminability Index is used to evaluate ability of CIs to distinguish fault from health. Finally, the method is validated with experimental signals of bearing and gear, the result show that this method is effective to assess CI performance, and it could help to select CIs in CMS system.

Feature optimization for bearing fault diagnosis

This paper presents methods of feature optimization for bearing fault diagnosis. These methods optimize statistical features in time domain and frequency domain. These optimization methods mainly consist of dimensionless processing and evaluation. Dimensionless processing method is used to avoid the influence of dimension and magnitude to the sensitivity. Fault sensitivity and discrete degree of features are evaluated. And features are selected according to the evaluation results. Analysis results of vibration signals of normal bearings, bearings with outer ring fault, bearings with inner ring fault and bearings with rolling element fault are presented. The results show that these methods are efficient to improve the separability of features.

Application of novelty detection methods to health monitoring and typical fault diagnosis of a turbopump

Novelty detection is the identification of deviations from a training set. It is suitable for monitoring the health of mechanical systems where it usually is impossible to know every potential fault. In this paper, two novelty detectors are presented. The first detector which integrates One-Class Support Vector Machine (OCSVM) with an incremental clustering algorithm is designed for health monitoring of the turbopump, while the second one which is trained on sensor fault samples is designed to recognize faults from sensors and faults actually from the turbopump. Analysis results showed that these two detectors are both sensitive and efficient for the health monitoring of the turbopump.

Enhanced Fault Detection of Rolling Element Bearing Based on Cepstrum Editing and Stochastic Resonance

By signal pre-whitening based on cepstrum editing,the envelope analysis can be done over the full bandwidth of the pre-whitened signal, and this enhances the bearing characteristic frequencies. The bearing faults detection could be enhanced without knowledge of the optimum frequency bands to demodulate, however, envelope analysis over full bandwidth brings more noise interference. Stochastic resonance (SR), which is now often used in weak signal detection, is an important nonlinear effect. By normalized scale transform, SR can be applied in weak signal detection of machinery system. In this paper, signal pre-whitening based on cepstrum editing and SR theory are combined to enhance the detection of bearing fault. The envelope spectrum kurtosis of bearing fault characteristic components is used as indicators of bearing faults. Detection results of planted bearing inner race faults on a test rig show the enhanced detecting effects of the proposed method. And the indicators of bearing inner race faults enhanced by SR are compared to the ones without enhancement to validate the proposed method.