As the rolling bearings being the key part of rotary machine, its healthy condition is quite important for safety production. Fault diagnosis of rolling bearing has been research focus for the sake of improving the economic efficiency and guaranteeing the operation security. However, the collected signals are mixed with ambient noise during the operation of rotary machine, which brings great challenge to the exact diagnosis results. Using signals collected from multiple sensors can avoid the loss of local information and extract more helpful characteristics. Recurrent Neural Networks (RNN) is a type of artificial neural network which can deal with multiple time sequence data. The capacity of RNN has been proved outstanding for catching time relevance about time sequence data. This paper proposed a novel method for bearing fault diagnosis with RNN in the form of an autoencoder. In this approach, multiple vibration value of the rolling bearings of the next period are predicted from the previous period by means of Gated Recurrent Unit (GRU)-based denoising autoencoder. These GRU-based non-linear predictive denoising autoencoders (GRU-NP-DAEs) are trained with strong generalization ability for each different fault pattern. Then for the given input data, the reconstruction errors between the next period data and the output data generated by different GRU-NP-DAEs are used to detect anomalous conditions and classify fault type. Classic rotating machinery datasets have been employed to testify the effectiveness of the proposed diagnosis method and its preponderance over some state-of-the-art methods. The experiment results indicate that the proposed method achieves satisfactory performance with strong robustness and high classification accuracy.

Fault diagnosis of rolling bearings with recurrent neural network-based autoencoders.

Signal based condition monitoring techniques for fault detection and diagnosis of induction motors: A state-of-the-art review

Stator/rotor winding faults are the common faults in squirrel-cage induction motor systems, which motivates the study of incipient fault detection and isolation (IFDI) to improve the safety and reliability of CRH (China Railway High-speed) trains. In this paper, a dynamic model for squirrel caged induction motor in d-q coordinate system is established firstly, further, the models and characteristics of incipient broken-rotor-bar fault and turn-to-turn short fault are analyzed. After that, a novel robust diagnosis design is proposed for the possible incipient stator/rotor winding faults. Experimental analyses show that the detection and isolation scheme designed in this paper provides high sensitivity and accurate isolation to incipient winding faults.

Incipient winding fault detection and diagnosis for squirrel-cage induction motors equipped on CRH trains.

Motor fault diagnosis using attention mechanism and improved adaboost driven by multi-sensor information

A methodology for automatic incipient broken rotor bar detection in induction motors (IMs) is presented. Sparse representations of signals are applied as a diagnosis technique. The novelty of this technique is that it can analyze the frequency spectra from vibration signals even when the differences among signals are small. This representation allows decomposing or reconstructing signals through a trained dictionary that has learned the features of one specific group/class. The main feature of this paper is the use of overcomplete dictionaries trained from sets of signals with faults to be detected. In this way, trained dictionaries perform the decomposition of signals using the orthogonal matching pursuit (OMP) algorithm. The decomposition is evaluated and classified by error-based criteria and a majority decision classifier, allowing the detection of early damage, ranging from 1 mm to one broken bar. The detection is performed by the decomposition of vibration signals from three axes ( ${x}$ , ${y}$ , and ${z}$ ) of IMs under three load conditions (unloaded, half loaded, and three-fourths loaded) and different levels of damage (healthy or 0 mm, 1–9 mm, and one broken bar). These signals are processed by the Fourier transform and the spectrum obtained is evaluated by the OMP algorithm. Finally, the retrieved information is evaluated and the diagnosis is given. All algorithms are developed in MATLAB software and the detection accuracy is higher than 90% for damages as small as 1 mm.

Incipient Broken Rotor Bar Detection in Induction Motors Using Vibration Signals and the Orthogonal Matching Pursuit Algorithm

The knowledge of the broken rotor bars characteristic frequencies and amplitudes has a great importance for all related diagnostic methods. The monitoring of motor faults requires a high resolution spectrum to separate different frequency components. The Discrete Fourier Transform (DFT) has been widely used to achieve these requirements. However, at low slip this technique cannot give good results. As a solution for these problems, this paper proposes an efficient technique based on a neural network approach and Hilbert transform (HT) for broken rotor bar diagnosis in induction machines at low load. The Hilbert transform is used to extract the stator current envelope (SCE). Two features are selected from the (SCE) spectrum (the amplitude and frequency of the harmonic). These features will be used as input for neural network. The results obtained are astonishing and it is capable to detect the correct number of broken rotor bars under different load conditions.

Detection of broken rotor bar faults in induction motor at low load using neural network.

It is well known that stator winding faults such the inter-turn short circuit are the most frequent source of breakdowns in induction motors. Early detection of any small inter-turn short circuit and location of the faulty phase at different load would eliminate some subsequent damage to adjacent coils and stator core, reducing then the repair cost. To achieve this purpose, the present paper presents a new method of diagnosis and detection of inter turn short circuit fault using discrete wavelet transform (DWT) and neural networks (NN). This method consists in analyzing the stator current by DWT in order to compute the energy associated with the stator fault in the frequency bandwidth. Then, this energy is used as input for a NN classifier. The results obtained are astonishing and the approach is able to detect any small number of shorted turns and the faulty phase even under different load of the machine.

Wavelet transform and neural network techniques for inter-turn short circuit diagnosis and location in induction motor

DWT and Hilbert Transform for Broken Rotor Bar Fault Diagnosis in Induction Machine at Low Load

In this paper, a new robust mean shift tracker is proposed by utilizing the joint color and texture histogram. The contribution of our work is to take local phase quantization (LPQ) operator advantage of texture features representation, and to combine it with a color histogram mean shift tracking algorithm. The LPQ technique can be applied to obtain the texture features which represent the object. In texture classification, The LPQ operator is much robust to blur than the well-known local binary pattern operator (LBP). Compared with traditional color histogram mean shift algorithm which considers only color statistical information of the object, the joint color-LPQ texture histogram is more robust and overcome some difficulties of the traditional color histogram mean shift algorithm. Comparative experimental results on numerous challenging image sequences show that the proposed algorithm obtains considerably better performance than several state-of-the-art methods, especially traditional mean shift tracker. The algorithm is evaluated by numerical parameters: the center location and the average overlap, it proved the tracking robustness in presence of similar target appearance and background, motion blurring.

Improved object tracking via joint color-LPQ texture histogram based mean shift algorithm

This paper presents the design of mobile robot visual navigation system in indoor environment based on fuzzy logic controllers (FLC) and optical flow (OF) approach. The proposed control system contains two Takagi–Sugeno fuzzy logic controllers for obstacle avoidance and goal seeking based on video acquisition and image processing algorithm. The first steering controller uses OF values calculated by Horn–Schunck algorithm to detect and estimate the positions of the obstacles. To extract information about the environment, the image is divided into two parts. The second FLC is used to guide the robot to the direction of the final destination. The efficiency of the proposed approach is verified in simulation using Visual Reality Toolbox. Simulation results demonstrate that the visual based control system allows autonomous navigation without any collision with obstacles.

/pdf/mobile-robot-visual-navigation-based-on-fuzzy-logic-and-2n2s8qnise.pdf

Mohamed Boumehraz

Papers

Detection of broken rotor bar faults in induction motor at low load using neural network.

Wavelet transform and neural network techniques for inter-turn short circuit diagnosis and location in induction motor

DWT and Hilbert Transform for Broken Rotor Bar Fault Diagnosis in Induction Machine at Low Load

Improved object tracking via joint color-LPQ texture histogram based mean shift algorithm

Mobile robot visual navigation based on fuzzy logic and optical flow approaches