Harapajan Singh

Fault Detection and Diagnosis of Induction Motors Using Motor Current Signature Analysis and a Hybrid FMM–CART Model

In this paper, a novel approach to detect and classify comprehensive fault conditions of induction motors using a hybrid fuzzy min-max (FMM) neural network and classification and regression tree (CART) is proposed. The hybrid model, known as FMM-CART, exploits the advantages of both FMM and CART for undertaking data classification and rule extraction problems. A series of real experiments is conducted, whereby the motor current signature analysis method is applied to form a database comprising stator current signatures under different motor conditions. The signal harmonics from the power spectral density are extracted as discriminative input features for fault detection and classification with FMM-CART. A comprehensive list of induction motor fault conditions, viz., broken rotor bars, unbalanced voltages, stator winding faults, and eccentricity problems, has been successfully classified using FMM-CART with good accuracy rates. The results are comparable, if not better, than those reported in the literature. Useful explanatory rules in the form of a decision tree are also elicited from FMM-CART to analyze and understand different fault conditions of induction motors.

Offline and online fault detection and diagnosis of induction motors using a hybrid soft computing model

In this paper, a hybrid soft computing model comprising the Fuzzy Min-Max (FMM) neural network and the Classification and Regression Tree (CART) for motor fault detection and diagnosis is described. Specifically, the hybrid model, known as FMM-CART, is used to detect and classify fault conditions of induction motors in both offline and online environments. A series of experiments is conducted, whereby the Motor Current Signature Analysis (MCSA) method is applied to form a database containing stator current signatures under different motor conditions. The signal harmonics from the power spectral density (PSD) are extracted, and used as the discriminative input features for fault classification with FMM-CART. Three main induction motor conditions, viz. broken rotor bars, stator winding faults, and unbalanced supply, are used to evaluate the effectiveness of FMM-CART. The results indicate that FMM-CART is able to detect motor faults in the early stage, in order to avoid further damage to the induction motor as well as the overall machine or system that uses the motor in its operations.

A modified fuzzy min-max neural network for data clustering and its application to power quality monitoring

When no prior knowledge is available, clustering is a useful technique for categorizing data into meaningful groups or clusters. In this paper, a modified fuzzy min-max (MFMM) clustering neural network is proposed. Its efficacy for tackling power quality monitoring tasks is demonstrated. A literature review on various clustering techniques is first presented. To evaluate the proposed MFMM model, a performance comparison study using benchmark data sets pertaining to clustering problems is conducted. The results obtained are comparable with those reported in the literature. Then, a real-world case study on power quality monitoring tasks is performed. The results are compared with those from the fuzzy c-means and k-means clustering methods. The experimental outcome positively indicates the potential of MFMM in undertaking data clustering tasks and its applicability to the power systems domain.

Application of the fuzzy min–max neural network to fault detection and diagnosis of induction motors

In this paper, an application of the motor current signature analysis (MCSA) method and the fuzzy min–max (FMM) neural network to detection and classification of induction motor faults is described. The finite element method is employed to generate simulated data pertaining to changes in the stator current signatures under different motor conditions. The MCSA method is then used to process the stator current signatures. Specifically, the power spectral density is employed to extract harmonics features for fault detection and classification with the FMM network. Various types of induction motor faults, which include stator winding faults and eccentricity problems, under different load conditions are experimented. The results are analyzed and compared with those from other methods. The outcomes indicate that the proposed technique is effective for fault detection and diagnosis of induction motors under different conditions.

Detection and classification of electrical supply voltage quality to electrical motors using the Fuzzy-Min-Max neural network

Generally electrical motors have a designed lifespan of more than 20 years when supplied with the proper rated voltage and acceptable voltage waveforms. However when motors are subjected to lowered levels of supply voltage quality conditions appearing simultaneously due to disturbances of overvoltage or undervoltage, voltage unbalance and voltage waveform distortions, it causes motor windings to be seriously overheated and reduce significantly the lifespan of the motor. Detection and classification systems are expensive compared to the low cost of motors as such detection systems are seldom integrated together with the motor. In this paper, a control methodology to enable a simple low cost for the proper detection and classification of electrical supply voltage condition to electrical machines using Fuzzy-Min-Max neural network to significantly improve the satisfactory operation and life span of the electrical machines is presented. The proper application of supply voltage quality levels can reduce the downtime and operating expenses of the electrical machines, thus improving return of investment on assets managed by the organization. In this paper, a simple control methodology for the early stage detection and classification of the electrical voltage supply condition in electrical machines based on Fuzzy-Min-Max neural network is presented. The condition of the supply voltage quality to electrical machines is diagnosed and classified using Fuzzy Min-Max neural network. It will be shown that the developed method is simple in dealing with any supply voltage condition to detect and allows for the ease in classification of the supply voltage pattern. Test results for the classified patterns have shown that the method used for this classification scheme is able to correctly identify supply voltage conditions, and the adopted Fuzzy-Min-Max neural network condition monitoring based method is efficient.

Detection and diagnosis of broken rotor bars and eccentricity faults in induction motors using the Fuzzy Min-Max neural network

Fault detection and diagnosis of electrical machines is gaining importance in regards to machine downtimes, where an unpredicted shutdown of operations owing to unavailability of machines can be very costly. As such, an early warning system for incipient machine faults using condition monitoring is of significance in practical applications. In this paper, we propose a fault detection and diagnosis system to detect and classify broken rotor bars and eccentricity faults of induction motors using the Fuzzy MinMax (FMM) neural network. A series of real experiments is conducted, where the acquired current signals under various motor conditions is used to build a database. The Power Spectral Density is then used to extract the discriminative input features for fault detection and classification with FMM. The results are comparable, if not better, than those from the MultiLayer Perceptron neural network and other methods reported in the literature.

Power Quality Analysis Using a Hybrid Model of the Fuzzy Min–Max Neural Network and Clustering Tree

A hybrid intelligent model comprising a modified fuzzy min-max (FMM) clustering neural network and a modified clustering tree (CT) is developed. A review of clustering models with rule extraction capabilities is presented. The hybrid FMM-CT model is explained. We first use several benchmark problems to illustrate the cluster evolution patterns from the proposed modifications in FMM. Then, we employ a case study with real data related to power quality monitoring to assess the usefulness of FMM-CT. The results are compared with those from other clustering models. More importantly, we extract explanatory rules from FMM-CT to justify its predictions. The empirical findings indicate the usefulness of the proposed model in tackling data clustering and power quality monitoring problems under different environments.

Condition monitoring of electrical supply voltage quality to electrical machines using RBF neural network

Three phase electrical machines are normally exposed to lowered levels of supply voltage quality conditions which can appear simultaneously due to voltage disturbances of overvoltage or undervoltage, voltage unbalance and voltage waveform distortions. These voltage disturbances can cause effects of seriously overheating winding insulation resulting in degradation and reduced lifespan of the machines. The supply of electrical power with proper rated voltages and acceptable voltage waveforms can significantly improve the satisfactory operation and life span of the machines. The proper application of supply voltage quality levels can reduce the downtime and operating expenses of the electrical machines, thus improving return of investment on assets managed by the organization. In this paper, a control methodology for the early detection and classification of the electrical voltage supply condition in electrical machines based on radial based function (RBF) neural networks is presented. The condition of the supply voltage quality to electrical machines is diagnosed and classified using RBF neural networks. It will be shown that the developed method is simple in dealing with any supply voltage condition to detect and allows for the ease in classification of the supply voltage pattern. Test results for the classified patterns have shown that the method used for this classification scheme able to correctly identify supply voltage conditions, and the adopted RBF neural network condition monitoring based method is efficient.

Classification of Power Quality Disturbance Based on Continuous S-Transform-Windowing Technique (CST-WT) and ANOVA as a Feature Selection

This paper was conducted in order to identify and classify the different types of Power Quality Disturbances (PQD) based on a new approach the Analysis Of Variance (ANOVA). ANOVA is used as feature selection for the PQD parameters. The datum of PQD from the PSCAD/EMTDC® simulation and Power Quality Monitoring has been validated before feature extraction analysis can be commenced. The obtained datum is then analyzed by using Windowing Technique (WT) based on Continuous S-Transform (CST) to extract the features and its characteristics. Moreover, the study focuses an important issue concerning the identification of PQD selection, detection and classification. The feature and characteristics of three types of signal such as sag, swell, and transient signal are obtained. The outcome of the analysis shows that a new approach framework ANOVA-Based Before and After Neural Network (NN) classification has a slightly increases to 15-25% in term of classification of PQD.

ANOVA Based Feature Analysis and Selection in Power Quality Disturbances Identification

This study was conducted in order to identify the different types of PQD based on a new approach the Analysis Of Variance (ANOVA). ANOVA is used as feature selection for the Power Quality Disturbances (PQD) parameters. The datum of PQD from the PSCAD/EMTDC® simulation has been validated before feature extraction analysis can be commenced. The obtained datum is then analyzed by using cycle windowing technique based on Continuous S-Transform (CST) to extract the features and its characteristics. Moreover, the study focuses an important issue concerning the identification of PQD selection and detection. The feature and characteristics of four types of signal such as Sag, Swell, Transient and sinusoidal normal signal are obtained. The outcome of the analysis shows that a new approach ANOVA have a different result in term of identification of PQD.