Zhengjia He

A new approach to intelligent fault diagnosis of rotating machinery

This paper presents a new approach to intelligent fault diagnosis based on statistics analysis, an improved distance evaluation technique and adaptive neuro-fuzzy inference system (ANFIS). The approach consists of three stages. First, different features, including time-domain statistical characteristics, frequency-domain statistical characteristics and empirical mode decomposition (EMD) energy entropies, are extracted to acquire more fault characteristic information. Second, an improved distance evaluation technique is proposed, and with it, the most superior features are selected from the original feature set. Finally, the most superior features are fed into ANFIS to identify different abnormal cases. The proposed approach is applied to fault diagnosis of rolling element bearings, and testing results show that the proposed approach can reliably recognise different fault categories and severities. Moreover, the effectiveness of the proposed feature selection method is also demonstrated by the testing results.

EEMD method and WNN for fault diagnosis of locomotive roller bearings

Research highlights? EEMD and WNN are combined to propose an automated fault diagnosis method. ? Features are extracted from the sensitive IMF of EEMD in this method. ? The features are fed into WNN to identify the bearing health conditions. ? The method can identify the fault severities and the compound faults. The ensemble empirical mode decomposition (EEMD) can overcome the mode mixing problem of the empirical mode decomposition (EMD) and therefore provide more precise decomposition results. Wavelet neural network (WNN) possesses the advantages of both wavelet transform and artificial neural networks. This paper combines the merits of EEMD and WNN to propose an automated and effective fault diagnosis method of locomotive roller bearings. First, the vibration signals captured from the locomotive roller bearings are preprocessed by EEMD method and intrinsic mode functions (IMFs) are produced. Second, a kurtosis based method is presented and used to select the sensitive IMF. Third, time- and frequency-domain features are extracted from the sensitive IMF, its frequency spectrum and its envelope spectrum. Finally, these features are fed into WNN to identify the bearing health conditions. The diagnosis results show that the proposed method enables the identification of the single faults in the bearings and at the same time the recognition of the fault severities and the compound faults.

Application of an intelligent classification method to mechanical fault diagnosis

A new method for intelligent fault diagnosis of rotating machinery based on wavelet packet transform (WPT), empirical mode decomposition (EMD), dimensionless parameters, a distance evaluation technique and radial basis function (RBF) network is proposed in this paper. In this method, WPT and EMD are, respectively, used to preprocess vibration signals to mine fault characteristic information more accurately. Then, dimensionless parameters in time domain are extracted from each of the original vibration signals and preprocessed signals to form a combined feature set. Moreover, the distance evaluation technique is utilised to calculate evaluation factors of the combined feature set. Finally, according to the evaluation factors, the corresponding sensitive features are selected and input into the RBF network to automatically identify different machine operation conditions. An experiment of rolling element bearings is carried out to test the performance of the proposed method. The experimental result demonstrates that the method combining WPT, EMD, the distance evaluation technique and the RBF network may accurately extract fault information and select sensitive features, and therefore it may correctly diagnose the different fault categories occurring in the bearings. Furthermore, this method is applied to slight rub fault diagnosis of a heavy oil catalytic cracking unit, the actual result shows the method may be applied to fault diagnosis of rotating machinery effectively.

A multidimensional hybrid intelligent method for gear fault diagnosis

Identifying gear damage categories, especially for early faults and combined faults, is a challenging task in gear fault diagnosis. This paper proposes a new multidimensional hybrid intelligent diagnosis method to identify different categories and levels of gear damage automatically. In this method, Hilbert transform, wavelet packet transform (WPT) and empirical mode decomposition (EMD) are performed on gear vibration signals to extract additional fault characteristic information. Then, multidimensional feature sets including time-domain, frequency-domain and time-frequency-domain features are generated to reveal gear health conditions. Multiple classifiers based on several classification algorithms and input features are combined with genetic algorithm (GA). Because of the use of multidimensional features and the combination of multiple classifiers, more accurate diagnosis results are expected with the proposed method. Experiments with different gear damage categories and damage levels were conducted, and the vibration signals were captured under different loads and motor speeds. The proposed method is applied to the collected signals to identify the gear damage categories and damage levels. The diagnosis results show it can reliably recognize single damage modes, combined damage modes, and damage levels.

Multi-fault classification based on wavelet SVM with PSO algorithm to analyze vibration signals from rolling element bearings

Condition monitoring and fault diagnosis of rolling element bearings timely and accurately is very important to ensure the reliable operation of rotating machinery. In this paper, a multi-fault classification model based on the kernel method of support vector machines (SVM) and wavelet frame, wavelet basis were introduced to construct the kernel function of SVM, and wavelet support vector machine (WSVM) is presented. To seek the optimal parameters of WSVM, particle swarm optimization (PSO) is applied to optimize unknown parameters of WSVM. In this work, the vibration signals measured from rolling element bearings are preprocessed using empirical model decomposition (EMD). Moreover, a distance evaluation technique is performed to remove the redundant and irrelevant information and select the salient features for the classification process. Hence, a relatively new hybrid intelligent fault detection and classification method based on EMD, distance evaluation technique and WSVM with PSO is proposed. This method is validated on a rolling element bearing test bench and then applied to the bearing fault diagnosis for electric locomotives. Compared with the commonly used SVM, the WSVM can achieve a greater accuracy. The results indicated that the proposed method can reliably identify different fault patterns of rolling element bearings based on the vibration signals.

An ACO-based algorithm for parameter optimization of support vector machines

One of the significant research problems in support vector machines (SVM) is the selection of optimal parameters that can establish an efficient SVM so as to attain desired output with an acceptable level of accuracy. The present study adopts ant colony optimization (ACO) algorithm to develop a novel ACO-SVM model to solve this problem. The proposed algorithm is applied on some real world benchmark datasets to validate the feasibility and efficiency, which shows that the new ACO-SVM model can yield promising results.

A Comparative Study on the Local Mean Decomposition and Empirical Mode Decomposition and Their Applications to Rotating Machinery Health Diagnosis

Health diagnosis of the rotating machinery can identify potential failure at its early stage and reduce severe machine damage and costly machine downtime. In recent years, the adaptive decomposition methods have attracted many researchers’ attention, due to less influences of human operators in the practical application. This paper compares two adaptive methods: local mean decomposition (LMD) and empirical mode decomposition (EMD) from four aspects, i.e., local mean, decomposed components, instantaneous frequency, and the waveletlike filtering characteristic through numerical simulation. The comparative results manifest that more accurate instantaneous frequency and more meaningful interpretation of the signals can be acquired by LMD than by EMD. Then LMD and EMD are both exploited in the health diagnosis of two actual industrial rotating machines with rub-impact and steam-excited vibration faults, respectively. The results reveal that LMD seems to be more suitable and have better performance than EMD for the incipient fault detection. LMD is thus proved to have potential to become a powerful tool for the surveillance and diagnosis of rotating machinery.

Matching Demodulation Transform and SynchroSqueezing in Time-Frequency Analysis

The authors introduce an iterative algorithm, called matching demodulation transform (MDT), to generate a time-frequency (TF) representation with satisfactory energy concentration. As opposed to conventional TF analysis methods, this algorithm does not have to devise ad-hoc parametric TF dictionary. Assuming the FM law of a signal can be well characterized by a determined mathematical model with reasonable accuracy, the MDT algorithm can adopt a partial demodulation and stepwise refinement strategy for investigating TF properties of the signal. The practical implementation of the MDT involves an iterative procedure that gradually matches the true instantaneous frequency (IF) of the signal. Theoretical analysis of the MDTs performance is provided, including quantitative analysis of the IF estimation error and the convergence condition. Moreover, the MDT-based synchrosqueezing algorithm is described to further enhance the concentration and reduce the diffusion of the curved IF profile in the TF representation of original synchrosqueezing transform. The validity and practical utility of the proposed method are demonstrated by simulated as well as real signal.

A demodulation method based on improved local mean decomposition and its application in rub-impact fault diagnosis

Demodulation is an available method for mechanical diagnoses, and a demodulation technique based on improved local mean decomposition (LMD) is proposed in this paper. A method of boundary process and a strategy for determining the step size of moving average are presented to improve the LMD algorithm. Instantaneous amplitude (IA) and instantaneous frequency (IF) of the signal can be computed independently of Hilbert transform using the improved LMD method. A well-constructed description of the derived IA and IF is given in the form of instantaneous time–frequency spectrum (ITFS) which preserves both the time and frequency information simultaneously. Results of three synthetic signals indicate that this proposed method is the best demodulation approach to extracting the all-round carrier and modulated components as well as the accurate IF, compared with Hilbert–Huang transform and stationary wavelet transform. The validity of the technique is then demonstrated on a real rotor system of a gas turbine with rub-impact fault. Due to the opposite friction during operation, the transient fluctuations of the IF of the fundamental harmonic component are successfully identified in the ITFS. In addition, we find that the proposed technique is more effective and sensitive than other methods in detecting sub-harmonics and FM components contained in the rub-impact signals. Thus the present method is powerful in the analysis of modulated signals and is an effective tool for the detection of rub-impact faults.

Dynamic Modeling and Vibration Response Simulation for High Speed Rolling Ball Bearings With Localized Surface Defects in Raceways

A dynamic model is developed to investigate vibrations of high speed rolling ball bearings with localized surface defects on raceways. In this model, each bearing component (i.e., inner raceway, outer raceway and rolling ball) has six degrees of freedom (DOFs) to completely describe its dynamic characteristics in three-dimensional space. Gyroscopic moment, centrifugal force, lubrication traction/slip between bearing component are included owing to high speed effects. Moreover, local defects are modeled accurately and completely with consideration of additional deflection due to material absence, changes of Hertzian contact coefficient and changes of contact force directions due to raceway curvature variations. The obtained equations of motion are solved numerically using the fourth order Runge–Kutta–Fehlberg scheme with step-changing criterion. Vibration responses of a defective bearing with localized surface defects are simulated and analyzed in both time domain and frequency domain, and the effectiveness of fault feature extraction techniques is also discussed. An experiment is carried out on an aerospace bearing test rig. By comparing the simulation results with experiments, it is confirmed that the proposed model is capable of predicting vibration responses of defective high speed rolling ball bearings effectively.