Expert system for induction motor fault detection

Mathematics and Computers in Simulation, 2003

The paper deals with diagnosis problems of the induction motors in the case of rotor, stator and rolling bearing faults. Two kinds of neural networks (NN) were proposed for diagnostic purposes: multilayer perceptron networks and self organizing Kohonen networks. Neural networks were trained and tested using measurement data of stator current and mechanical vibration spectra. The efficiency of developed neural detectors was evaluated. Feedforward NN with very simple internal structure, used for the detection of all fault kinds, gave satisfactory results, which is very important in practical realization. Experiments with Kohonen networks indicated that they could be used for the initial classification of motor faults, as an introductory step before the proper neural detector based on multiplayer perceptron is used. The obtained results lead to a conclusion that neural detectors for rotor and stator faults as well as for rolling bearings and supply asymmetry faults can be developed based on measurement data acquired on-line in the drive system.

2007

This paper presents an original software tool based on a model of expert system that has been developed to satisfy the requirements of fault diagnosis, repair and maintenance of electrical machines in industrial, educational or other applications. This software comprises an SQL server database and provides different user levels (simple user, power user and administrator). By the introduction of suitable

International Journal of Scientific and Engineering Research

This work involves the development of an artificial intelligent (AI) scheme in the detection of rotor and stator faults in induction machines. Using discrete wavelet transform technique to process the stator current signals measured from faulty motors with isolated cases of rotor and stator faults, the signals are then matched against known fault signatures for the types of fault. The statistical features of fault signals such as mean, Skewness etc. were then extracted and fed to an Artificial Neural Network (ANN) for training. Different ANN architectures were then compared in terms of their accuracy in classification. The network with the lowest mean squared error was then tested with separate data set and the results found to be satisfactory

2013 International Conference on Electrical Engineering and Software Applications, 2013

This paper proposes a neural network based method to achieve an automatic detection of different stator faults of the induction motor. The concerned stator faults are the inter turns short circuit, phase to phase and phase to ground faults. The inputs of the feedforward multi-layer neural network are the indicators of the stator faults while its outputs are the corresponding faults. Therefore, the used indicators of faults are extracted from the symmetrical components of the stator currents which are the magnitude and the angle phase of the negative and zero sequence current. The neural network is trained by the back-propagation algorithm. A faulty simplified multiple coupled circuit model of a 1.1 kW induction motor is used to simulate the different operating conditions of the machine useful to built the data base for the training and the test procedures. The good training and test results show the efficiency of the proposed method.

E3S web of conferences, 2022

Currently, the environmental challenges have been considered as a strategic issue for most industrial companies around the world, threatening their sustainability and profit; This leads to taking the environmental dimensions seriously and preserving natural resources well, since they are a key criterion for sustainable development. In this context, this work calls for innovative solution and new technologies to support the development and integration of environmental considerations through the implementation of an automated fault detection and diagnosis system in induction machines in order to minimize downtime, increase machine utilization rate, get an idea of remaining machine life based on artificial intelligence (AI) and the analysis of collected data. Using the Pattern Recognition methods, this system aims to support decision making in terms of defect classification, through the following process: the collection of relevant data about the stator currents of two induction machines, powered by a converter, one healthy and the other defective, through the CompactRIO device, then the analysis of the data, using programs developed under LabVIEW software, and the extraction of the indicators to form a database. Based on analysis results, several intelligent methods by classification algorithms can organize the acquired data in order to automate the diagnostic process. Ultimately, the setup of an alert system to prevent rather than cure. The outcomes showed that the integration of predictive maintenance could help achieve an energy cost recovery equal to 10% of the total costs of an electric motor system. Hence, the premature detection of faults helps to minimize energy expenditure and achieve overall cost savings, which implies energy optimization.

IEEE Transactions on Industrial Informatics, 2017

Three phase induction motors (TIMs) are the primary means of transformation of electrical energy into mechanical energy in industry, since they are robust and present low cost. However, despite being robust, these machines are subject to electrical or mechanical faults. Thus, identifying a defect in a running motor may decrease the risk of possible damage. This article proposes an alternative approach to identify defects in the stator of these motors, by analysing current signals in the time domain. In addition, it presents the determination of the consequent fault severity by means of two proposals: i) a multi-agent system (MAS) with a classifier behaviour and; ii) a neural estimator. The faults observed are related to short circuits between turns in the stator coil of 1% to 10%. Experimental results are observed with motors of different powers, under various adverse situations of electrical feed and a wide range of load conditions on the machine shaft.

2018

This paper presents the fault diagnosis of a three-phase induction motor using fuzzy logic, neural network and hybrid system. Detailed analysis during voltage unbalance, open phase, low voltage and overload motor fault using these strategies are presented. Stator currents were measured and their root mean square were derived. These values were used to train data. Each output of these diagnosis tools is used to determine the motor conditions whether it is in a healthy state or in a faulty one. A novel hybrid system is design and used in fault diagnosis. Simulation results show that hybrid systems give the best estimation of faults and can be therefore used in monitoring of induction motors with greater efficiency.

… Conference, 2008

Eskişehir Osmangazi Üniversitesi Mühendislik ve Mimarlık Fakültesi Dergisi, 2021

2009 Second International Conference on Emerging Trends in Engineering & Technology, 2009

A degradation of metallurgical equipment is normal process depended on time. Some factors such as: operation process, friction, high temperature can accelerate the degradation process of metallurgical equipment. In this paper the authors analyzed three phase induction motors. These motors are common used in the metallurgy industry, for example in conveyor belt. The diagnostics of such motors is essential. An early detection of faults prevents financial loss and downtimes. The authors proposed a technique of fault diagnosis based on recognition of currents. The authors analyzed 4 states of three phase induction motor: healthy three phase induction motor, three phase induction motor with 1 faulty rotor bar, three phase induction motor with 2 faulty rotor bars, three phase induction motor with faulty ring of squirrel-cage. An analysis was carried out for original method of feature extraction called MSAF-RATIO15 (Method of Selection of Amplitudes of Frequencies -Ratio 15% of maximum of amplitude). A classification of feature vectors was performed by Bayes classifier, Linear Discriminant Analysis (LDA) and Nearest Neighbour classifier. The proposed technique of fault diagnosis can be used for protection of three phase induction motors and other rotating electrical machines. In the near future the authors will analyze other motors and faults. There is also idea to use thermal, acoustic, electrical, vibration signal together.