Qianjin Guo

A hybrid PSO-GD based intelligent method for machine diagnosis

This paper presents an intelligent methodology for diagnosing incipient faults in rotating machinery. In this fault diagnosis system, wavelet neural network techniques are used in combination with a new evolutionary learning algorithm. This new evolutionary learning algorithm is based on a hybrid of the constriction factor approach for particle swarm optimization (PSO) technique and the gradient descent (GD) technique, and is thus called HGDPSO. The HGDPSO is developed in such a way that a constriction factor approach for particle swarm optimization (CFA for PSO) is applied as a based level search, which can give a good direction to the optimal global region, and a local search gradient descent (GD) algorithm is used as a fine tuning to determine the optimal solution at the final. The effectiveness of the HGDPSO based WNN is demonstrated through the classification of the fault signals in rotating machinery. The simulated results show its feasibility and validity.

Broken Rotor Bars Fault Detection in Induction Motors Using Park's Vector Modulus and FWNN Approach

In this paper a new integrated diagnostic method based on the current Parks Vector modulus analysis and fuzzy wavelet neural network classifier is proposed for the diagnosis of rotor cage faults in operating three-phase induction motors. Detection of broken rotor bars has long been an important but difficult job in the detection area of induction motor faults. The characteristic frequency components of a faulted rotor in the stator current spectrum are very close to the power frequency component but by far less in amplitude, which brings about great difficulty for accurate detection. In order to overcome the shortage of broken rotor bars characteristic components being submerged by the fundamental one in the spectrum of the stator line current, Parks Vector modulus(PVM) analysis is used to detect the occurrence of broken rotor bar faults in our work. Simulation and experimental results are presented to show the merits of this novel approach for the detection of cage induction motor broken rotor bars.

Hybrid PSO based wavelet neural networks for intelligent fault diagnosis

A model of wavelet neural network (WNN) using a new evolutionary learning algorithm is proposed in this paper. This new evolutionary learning algorithm is based on a hybrid of Particle Swarm Optimization (PSO) and gradient descent algorithm (GD), and is thus called HGDPSO. The Particle Swarm Optimizer has previously been used to train neural networks and generally met with success. The advantage of the PSO over many of the other optimization algorithms is its relative simplicity and quick convergence. But those particles collapse so quickly that it exits a potentially dangerous property: stagnation, which state would make it impossible to arrive at the global optimum, even a local optimum. HGDPSO was proposed for neural network training to avoid premature and eliminate stagnation in PSO. The effectiveness of the HGDPSO based WNN is demonstrated through the classification of the fault signals in rotating machinery. The simulated results show its feasibility and validity.

Development and application of internet-based remote monitoring and diagnostic platform for key equipment

Requirements for high efficiency in manufacturing, while pushing the development of key equipment, need high-performance maintenance as a safeguard. Internet-related techniques are enabling remote monitoring and maintenance to extend beyond local areas. Considering the need for software architecture with openness and interchangeability for flexible selection and replacement of data analysis techniques, this research develops a standardized multi-layer platform using Web Services, realizing data transmission among networks. The independence between the software platform and data analysis guarantees their research and development in parallel. Studies were done on nonlinear-data-analysis techniques, such as wavelet neural networks for fault diagnostics, particle filters for estimates, and support vector regression for condition prediction. These studies were included in a comprehensive application. The experimental results support condition-based maintenance for key equipment.

Fault Feature Extraction by Using Adaptive Chirplet Transform

The vibration generated by industrial machines always contains nonlinear and nonstationary signals. It is expected that a desired time-frequency analysis method should have good computation efficiency, and have good resolution in both time domain and frequency domain. In this paper, the adaptive Gaussian chirplet distribution for an integrated time-frequency signature extraction of the machine vibration is presented. The adaptive Gaussian chirplet spectrogram is nonnegative, has a high time-frequency resolution, and is free of cross term interference, so it offers the advantage of good localization of the vibration signal energy in the time-frequency domain. Experimental results show that the proposed method is very effective

Rolling bearings fault diagnosis based on adaptive gaussian chirplet spectrogram and independent component analysis

Condition monitoring of rolling element bearings through the use of vibration analysis is an established technique for detecting early stages of component degradation. The location dependent characteristic defect frequencies make it possible to detect the presence of a defect and to diagnose on what part of the bearing the defect is. The difficulty of localized defect detection lies in the fact that the energy of the signature of a defective bearing is spread across a wide frequency band and hence can be easily buried by noise. To solve this problem, the adaptive Gaussian chirplet distribution for an integrated time-frequency signature extraction of the machine vibration is developed; the method offers the advantage of good localization of the vibration signal energy in the time-frequency domain. Independent component analysis (ICA) is used for the redundancy reduction and feature extraction in the time-frequency domain, and the self-organizing map (SOM) was employed to identify the faults of the rolling element bearings. Experimental results show that the proposed method is very effective.

Transient Fault Signal Detection Using Wigner Higher-Order Moment Spectra Methods

Most mechanical faults in machinery reveal themselves through transient events in vibration signals. That is, the vibration generated by industrial machines always contains nonlinear and non-stationary signals. Many analysis methods in current use are optimal for Gaussian and linear models but are suboptimal when the problem is one of analysing non-stationary, non-linear and non-Gaussian signals. Focusing on the defects of different joint time-frequency representations, the analysis of transient vibration signals is considered using Wigner higher-order based time-frequency distributions in this paper. The Wigner higher-order moment spectra (WHOS), which are the extensions of the Wigner-Ville distribution (WVD) to higher-order moment spectra domains, can describe the higher-order moment spectral characteristics from the time domain and the frequency domain simultaneously. The simulation case show that this technique has high time-frequency resolution and reduced interference terms

A self-constructing compensatory fuzzy wavelet network and its applications

By utilizing some of the important properties of wavelets like denoising, compression, multiresolution along with the concepts of fuzzy logic and neural network, a new self-constructing fuzzy wavelet neural networks (SCFWNN) using compensatory fuzzy operators are proposed for intelligent fault diagnosis. An on-line learning algorithm is applied to automatically construct the SCFWNN. There are no rules initially in the SCFWNN. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter learning. The advantages of this learning algorithm are that it converges quickly and the obtained fuzzy rules are more precise. The proposed SCFWNN is much more powerful than either the neural network or the fuzzy system since it can incorporate the advantages of both. The results of simulation show that this SCFWNN method has the advantage of faster learning rate and higher diagnosing precision.

A new intelligent prediction method for grade estimation

In this paper, a novel PSO–SVR model that hybridized the constrict particle swarm optimization (PSO) and support vector regression (SVR) is proposed for grade estimation This hybrid PSO–SVR model searches for SVRs optimal parameters using constrict particle swarm optimization algorithms, and then adopts the optimal parameters to construct the SVR models The hybrid PSO–SVR grade estimation method has been tested on a number of real ore deposits The result shows that method has advantages of rapid training, generality and accuracy grade estimation approach It can provide with a very fast and robust alternative to the existing time-consuming methodologies for ore grade estimation.

Joint time-frequency and kernel principal component based SOM for machine maintenance

Conventional vibration signals processing techniques are most suitable for stationary processes. However, most mechanical faults in machinery reveal themselves through transient events in vibration signals. That is, the vibration generated by industrial machines always contains nonlinear and non-stationary signals. It is expected that a desired time-frequency analysis method should have good computation efficiency, and have good resolution in both time domain and frequency domain. In this paper, the auto-regressive model based pseudo-Wigner-Ville distribution for an integrated time-frequency signature extraction of the machine vibration is designed, the method offers the advantage of good localization of the vibration signal energy in the time-frequency domain. Kernel principal component analysis (KPCA) is used for the redundancy reduction and feature extraction in the time-frequency domain, and the self-organizing map (SOM) was employed to identify the faults of the rotating machinery. Experimental results show that the proposed method is very effective.