Application of integrated PCA and FIS approach to the selection of current and vibration signal features in mechanical fault classification of induction motor

2022 ◽  
pp. 1-19
Author(s):  
D. Gunapriya ◽  
C. Muniraj ◽  
K. Lakshmi
Keyword(s):  
Induction Motor ◽  
Research Work ◽  
Industrial Process ◽  
Principal Component ◽  
Vibration Signal ◽  
Integrated Approach ◽  
Classification Performance ◽  
Fault Classification ◽  
Synthesis Process ◽  
Mass Unbalance

The detection as well as analysis of faults in Induction Motor (IM) is prominent in the industrial process in recent decades, since it has been a demanding issue in industries to confirm the safe and reliable operations of IM. Though the electrical faults, mechanical faults and environmental faults cause damages in IM, as per Electric Power Research Institute (EPRI) statistical studies, the faults due to (i) rotor mass unbalance and (ii) rotor shaft bending substantially contribute 8-9% of the total motor fault. This present research work focuses on the issue of detecting and analysing the faults by studying the current and vibration data obtained from the three-phase squirrel cage IM under healthy and faulty conditions using the experimental workbench. It also depicts the development of a fault detection model for IM which comprises the integrated approach of Principal Component Analysis (PCA) and Fuzzy Interference System (FIS) and two level decision fuzzy measures. Besides, fuzzy integral data fusion technique has been used in this work for the improvement of diagnosing accuracy. The data acquired from the workbench system are first investigated through the PCA to extricate the appropriate features that provide the major information of collected data without reducing its dimensions. The projected data space using the principal components is non-deterministic for further synthesis process of fault classification. Hence, to classify the faults in IM, the obtained feature vectors from PCA are fed into FIS as an input and the classification performance is compared finally. The work experiment has been carried out under the healthy and different faulty conditions of motor and the proposed integrated approach is executed by using MATLAB.

scholarly journals Fault Diagnosis for Rolling Element Bearings Based on Feature Space Reconstruction and Multiscale Permutation Entropy

Entropy ◽  
2019 ◽  
Vol 21 (5) ◽  
pp. 519 ◽  
Author(s):  
Weibo Zhang ◽  
Jianzhong Zhou
Keyword(s):  
Diagnostic Algorithm ◽  
Feature Space ◽  
Vibration Signal ◽  
Rolling Bearing ◽  
Ensemble Empirical Mode Decomposition ◽  
Classification Performance ◽  
Fault Classification ◽  
Permutation Entropy ◽  
Support Vector ◽  
Intrinsic Mode Functions

Aimed at distinguishing different fault categories of severity of rolling bearings, a novel method based on feature space reconstruction and multiscale permutation entropy is proposed in the study. Firstly, the ensemble empirical mode decomposition algorithm (EEMD) was employed to adaptively decompose the vibration signal into multiple intrinsic mode functions (IMFs), and the representative IMFs which contained rich fault information were selected to reconstruct a feature vector space. Secondly, the multiscale permutation entropy (MPE) was used to calculate the complexity of reconstructed feature space. Finally, the value of multiscale permutation entropy was presented to a support vector machine for fault classification. The proposed diagnostic algorithm was applied to three groups of rolling bearing experiments. The experimental results indicate that the proposed method has better classification performance and robustness than other traditional methods.

scholarly journals Fault Detection and Diagnosis in Process Data Using Support Vector Machines

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Fang Wu ◽  
Shen Yin ◽  
Hamid Reza Karimi
Keyword(s):  
Dimensional Space ◽  
Industrial Process ◽  
Principal Component ◽  
Original Data ◽  
Fault Detection And Diagnosis ◽  
Fault Classification ◽  
Recursive Feature Elimination ◽  
Support Vector ◽  
Process Data ◽  
Tennessee Eastman Process

For the complex industrial process, it has become increasingly challenging to effectively diagnose complicated faults. In this paper, a combined measure of the original Support Vector Machine (SVM) and Principal Component Analysis (PCA) is provided to carry out the fault classification, and compare its result with what is based on SVM-RFE (Recursive Feature Elimination) method. RFE is used for feature extraction, and PCA is utilized to project the original data onto a lower dimensional space. PCAT2, SPE statistics, and original SVM are proposed to detect the faults. Some common faults of the Tennessee Eastman Process (TEP) are analyzed in terms of the practical system and reflections of the dataset. PCA-SVM and SVM-RFE can effectively detect and diagnose these common faults. In RFE algorithm, all variables are decreasingly ordered according to their contributions. The classification accuracy rate is improved by choosing a reasonable number of features.

scholarly journals Induction Motor Stator Fault Classification Using PCA-ANFIS Technique

2020 ◽  
Vol 19 (1) ◽  
pp. 26-32
Author(s):  
Ayodele Isqeel Abullateef ◽  
Mohammed Faiz Sanusi ◽  
Olabanji Sunday Fagbolagun
Keyword(s):  
Induction Motor ◽  
Membership Function ◽  
Induction Motors ◽  
Fuzzy Inference ◽  
Principal Component ◽  
Fault Classification ◽  
Function Type ◽  
Inference System ◽  
Stator Fault

Induction motors are used commonly for industrial operations due to their ease of operation coupled with ruggedness and reliability. However, they are subjected to stator faults which result in damage and consequently revenue losses. The classification of stator fault in a three-phase induction motor based on Adaptive neuro-fuzzy inference system (ANFIS) in combination with Principal Component Analysis (PCA) is proposed in this study. A burnt motor was redesigned and rewound while data acquisition was developed to acquire the current and vibration data needed for the fault classification. The data feature extraction for the fault classification was carried out by PCA while backpropagation and the least-squares algorithms were used for the training of the data. Three principal components, which severs as input for the ANFIS, were used to represent the entire data. The ANFIS was tested under four different paradigms, while the membership function type and epoch number were changed at each instant. The ANFIS model based on the triangular membership function and 10 epoch number was found appropriate and used, bringing the accuracy of the model to over 99% with the lowest ANFIS training RMSE error of      1.1795e-6. The ANFIS validation results of the fault classification show that the results are accurate, indicating that the PCA-ANFIS technique is applicable in fault diagnosis and classification of stator faults in induction motors.

Simple Threshold Method in Fault Diagnosis for Voltage Source Inverter in a Direct Torque Control Induction Motor Drive

2019 ◽  
Vol 13 ◽  
Author(s):  
Farid Kadri ◽  
Mohamed Assaad Hamida
Keyword(s):  
Fault Diagnosis ◽  
Induction Motor ◽  
Direct Torque Control ◽  
Classification Performance ◽  
Torque Control ◽  
Fault Classification ◽  
Voltage Source ◽  
Motor Drive ◽  
Induction Motor Drive ◽  
Voltage Source Inverter

Background: The need for a diagnosis today, becomes a necessity for variable speed AC drives in several industrial applications. An important research axis is oriented towards monitoring the state of the converter supplying the electric motor. Indeed, the voltage source inverter is likely to have switching faults. Therefore, an emergency stop of the motor drive must be done. Objective: After reviewing related patents and works, the objective of this paper is to identify the feasibility of fault detection and diagnosis in a three-phase inverter feeding an induction motor. Methods: The proposed approach is a simple threshold fault classification method applied to fault diagnosis of a direct torque control (DTC) induction motor drive using the stator Concordia mean current vector. With a fault occurrence, a localization domain consisting of seven patterns is constructed. Results: Simulated results on 1.5-kW induction motor drive show the effectiveness of the proposed approach with a good classification performance. Conclusion: The classification performance of our simple diagnosis system is acceptable for one fault occurrence compared to others methods. Faulty switch detection and identification is completed within a few periods of current. Using intelligent technique should improve classification performances for multiple faults occurrence.

Induction Motor Fault Diagnosis and Classification Through Sparse Representation

2017 ◽  
Author(s):  
Jianjing Zhang ◽  
Peng Wang ◽  
Chuang Sun ◽  
Ruqiang Yan ◽  
Robert X. Gao
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Sparse Representation ◽  
Induction Motor ◽  
Vibration Signal ◽  
Performance Measure ◽  
Fault Classification ◽  
Human Intervention ◽  
Robustness To Noise ◽  
Fault Characteristic

Condition monitoring and fault diagnosis of induction motor play a critical role in operation safety and production efficiency. In recent study, sparse representation has demonstrated its simplicity in training, robustness to noise and high accuracy in classification. This paper evaluates the effectiveness of sparse representation as an alternative approach to induction motor fault diagnosis with fault classification rate and robustness to noise as performance measure. Aiming at eliminating the human intervention in fault characteristic frequency detection and extensive feature extraction steps in traditional method, the spatial pattern of the vibration signal is studied as the classifier input. The residual sparsity index (RSI) is proposed to quantify the degree of multi-class data separation and evaluate the reliability of classification results. Experimental results show that the sparse representation method using vibration signal achieves high motor multi-fault classification accuracy and good robustness to noise, with no human intervention required for fault characteristic pattern detection and the need for long feature extraction eliminated. Finally, RSI confirms the high overall reliability of classification results.

Fault Diagnosis for Motor Rotor Based on KPCA-SVM

2011 ◽  
Vol 143-144 ◽  
pp. 680-684 ◽  
Author(s):  
Ping Li ◽  
Xue Jun Li ◽  
Ling Li Jiang ◽  
Da Lian Yang
Keyword(s):  
Fault Diagnosis ◽  
Dimensional Space ◽  
Principal Component ◽  
Vibration Signal ◽  
Training Sample ◽  
Fault Classification ◽  
Kernel Principal Component Analysis ◽  
Support Vector ◽  
Nonlinear Properties ◽  
Nonlinear Features

Aimed at the nonlinear properties of motor rotor vibration signal,a fault diagnosis method based on kernel principal component analysis (KPCA) and support vector machines (SVM) was proposed. Initial feature vectors of motor vibration signal were mapped into higher-dimensional space with kernel function. Then the PCA method was used to analyze the data in the high dimensional space to extract the nonlinear features which is used as training sample of SVM fault classifier. Then the rotor fault is identified using the trained classifier. The classification effect of KPCA-SVM is compared with PCA-SVM and SVM. The result shows that the method based on KPCA-SVM can identify motor rotor fault efficiently and fulfill fault classification accurately.

scholarly journals Transformer Oil Quality Assessment Using Random Forest with Feature Engineering

Energies ◽  
2021 ◽  
Vol 14 (7) ◽  
pp. 1809
Author(s):  
Mohammed El Amine Senoussaoui ◽  
Mostefa Brahami ◽  
Issouf Fofana
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Oil Quality ◽  
Principal Component ◽  
Condition Assessment ◽  
Classification Performance ◽  
Transformer Oil ◽  
Classification Model ◽  
Insulation Degradation ◽  
Transformer Oils

Machine learning is widely used as a panacea in many engineering applications including the condition assessment of power transformers. Most statistics attribute the main cause of transformer failure to insulation degradation. Thus, a new, simple, and effective machine-learning approach was proposed to monitor the condition of transformer oils based on some aging indicators. The proposed approach was used to compare the performance of two machine-learning classifiers: J48 decision tree and random forest. The service-aged transformer oils were classified into four groups: the oils that can be maintained in service, the oils that should be reconditioned or filtered, the oils that should be reclaimed, and the oils that must be discarded. From the two algorithms, random forest exhibited a better performance and high accuracy with only a small amount of data. Good performance was achieved through not only the application of the proposed algorithm but also the approach of data preprocessing. Before feeding the classification model, the available data were transformed using the simple k-means method. Subsequently, the obtained data were filtered through correlation-based feature selection (CFsSubset). The resulting features were again retransformed by conducting the principal component analysis and were passed through the CFsSubset filter. The transformation and filtration of the data improved the classification performance of the adopted algorithms, especially random forest. Another advantage of the proposed method is the decrease in the number of the datasets required for the condition assessment of transformer oils, which is valuable for transformer condition monitoring.

scholarly journals Hyperspectral Image Classification with Localized Graph Convolutional Filtering

2021 ◽  
Vol 13 (3) ◽  
pp. 526
Author(s):  
Shengliang Pu ◽  
Yuanfeng Wu ◽  
Xu Sun ◽  
Xiaotong Sun
Keyword(s):  
Hyperspectral Image ◽  
Principal Component ◽  
Representation Learning ◽  
Classification Performance ◽  
Hyperspectral Data ◽  
Spectral Graph Theory ◽  
Feature Reduction ◽  
Graph Representation ◽  
Novel Method ◽  
Local Graph

The nascent graph representation learning has shown superiority for resolving graph data. Compared to conventional convolutional neural networks, graph-based deep learning has the advantages of illustrating class boundaries and modeling feature relationships. Faced with hyperspectral image (HSI) classification, the priority problem might be how to convert hyperspectral data into irregular domains from regular grids. In this regard, we present a novel method that performs the localized graph convolutional filtering on HSIs based on spectral graph theory. First, we conducted principal component analysis (PCA) preprocessing to create localized hyperspectral data cubes with unsupervised feature reduction. These feature cubes combined with localized adjacent matrices were fed into the popular graph convolution network in a standard supervised learning paradigm. Finally, we succeeded in analyzing diversified land covers by considering local graph structure with graph convolutional filtering. Experiments on real hyperspectral datasets demonstrated that the presented method offers promising classification performance compared with other popular competitors.

Sign in / Sign up

Export Citation Format

Share Document