Papers by Keyword: Fault Diagnosis

Abstract: The background noise makes it difficult to detect incipient faults through vibration analysis. The stochastic resonance (SR) method can be applied to enhance the signal-to-noise ratio (SNR) of a system output using the unavoidable environmental noise. The parameters selection is the most important to generate SR. The proposed fault diagnosis method utilizes the artificial bee colony algorithm to find the best parameters of SR so as to match input signals and detect faults. The performance of the proposed method is confirmed as compared to the fixed parameters method.

Abstract: The vibration signals of rolling element bearings are non-linear and non-stationary and the corresponding fault features are difficult to be extracted. EEMD (Ensemble empirical mode decomposition) is effective to detect bearing faults. In the present investigation, MEEMD (Modified EEMD) is presented to diagnose the outer and inner race faults of bearings. The original vibration signals are analyzed using IMFs (intrinsic mode functions) extracted by MEEMD decomposition and Hilbert spectrum in the proposed method. The numerical and experimental results of the comparison between MEEMD and EEMD indicate that the proposed method is more effective to extract the fault features of outer and inner race of bearings than EEMD.

Abstract: A novel fast algorithm for lndependent Component Analysis is introduced, which can be used for blind source separation and machine fault diagnosis feature extraction. It is shown how a neural network learning rule can be transformed into a fixed-point iteration, which provides an algorithm that is very simple, does not depend on any user-defined parameters, and is fast to converge to the most accurate solution allowed by the data. The purpose of this paper is to review the application of blind source separation in the machine fault diagnosis,including the following aspects: noise elimination and extraction of the weak signals,the separation of multi-fault sources,redundancy reduction,feature extraction and pattern classification based on independent component analysis. And its application in machine fault diagnosis is illustrated by the examples. In addition, some prospects about using blind source separation for machine fault diagnosis are discussed.

Abstract: Aero-Engine health management generally involves a series of activities over the period from the aerospace breaking down until it returning to normal, including signal processing, monitoring, health assessment, decision supporting, human-computer interaction, and so on. As one of the key technology of Aero-Engine health management, fault diagnosis plays a very important role on the safe operation of Aero-Engine. Currently, for effective challenging Aero-Engine health management, a fault diagnosis of Aero-Engine based on Principal Component Analysis (PCA) s proposed. Firstly, based on a variety of significant parameters of the collected information, principal component analysis model is established. Secondly, the fault diagnosis of engine operating conditions is realized by comparing the T² statistic and Squared Prediction Error (SPE) statistic as an engine running in good condition threshold limits. Finally, through the variable's cumulative contributions diagram with the behavior of SPE overrun, the fault variables are effectively worked out. Experimental results show that the proposed PCA method can efficiently come true Aero-Engine health management o and has some engineering applications values.

Abstract: The various formulations of the technical diagnostics problem are possible when creation and use of analog circuits. One approach allows deviation a sufficiently large set of the elements parameters. The results of this diagnosis can be used to improve the technology of analog circuits or predict their behavior depending on the time of exposure or destabilizing factors. The power sources (current sources and emf) commonly used as a testing influence on circuit in realization test diagnosis. A short circuit approach of the testing experiments organization with active analog circuits with partly inaccessible nodes is considered.

Abstract: Aiming at the problems of less study sample, large network scale and long training time existing in current fault diagnosis field, we develop a new method based on KPCA and selective neural network ensemble. First, reducing the data size by using KPCA to extract the sample features. Then achieving a selective neural network ensemble method based on improved binary particle swarm optimization algorithm (IBPSOSEN), and combining the two methods for fault diagnosis. In selective neural network algorithm, bagging method is used to take a number of different training sets of fault samples to solve the problem of less fault samples. Finally, simulation experiments and comparisons over Tennessee Eastman Process (TE) demonstrate the effectiveness and feasibility of the proposed method.

Abstract: This article are discussed Sliding Bearing for Turbine instability failure research status in detail; Gives the theoretical analysis On sliding bearing oil film instability failure mechanism, to further explore the oil whirl and oil whip manifestations and signal spectral characteristics. And analysis of the oil whirl speed changes the vibration characteristics of typical regions, given the time-domain waveform oil whirl and oil whip axis trajectory. Details of the turbine generator film and the main reason for the instability factors and made a film instability fault governance approach. It helped for Fault Diagnosis of Turbine film and achieving security and stability of Turbine provides technical reference.

Abstract: A new method for bearing fault diagnosis is proposed based on Probabilistic Principal Component Analysis (PPCA) and Cyclic Bispectrum (CB). The first procedure is signal de-noised using PPCA and the second procedure is the CB analysis. The effectiveness of the proposed method is demonstrated by numerical simulation and experimental investigation of a rolling bearing with outer race fault.

Abstract: To deal with the lack of effective experimental data under the current condition for gearbox fault pattern recognition, the Wind Turbine Drivetrain Diagnostics Simulator (WTDS) was used for experimental investigation and gained large number of gear fault samples. The wavelet transform is employed to decompose the vibration signal to obtain the energy ratio in each frequency band. Taking energy ratios as feature vectors, the pattern recognition results are obtained by the support vector classification (SVC). The experimental results show that the hybrid approach is robust to noise and has high classification accuracy.

Abstract: Fault diagnosis is essentially a kind of pattern recognition. In this paper propose a novel machinery fault diagnosis method based on supervised locally linear embedding is proposed first. The approach first performs the recently proposed manifold learning algorithm locally linear embedding on the high-dimensional fault signal samples to learn the intrinsic embedded multiple manifold features corresponding to different fault modes. Supervised locally linear embedding not only can map them into a low-dimensional embedded space to achieve fault feature extraction, but also can deal with new fault samples. Finally fault classification is carried out in the embedded manifold space. The ball bearing fault signals are used to validate the proposed fault diagnosis method. The results indicate that the proposed approach obviously improves the fault classification performance and outperforms the other traditional approaches.