Three-Phase Induction Motors Faults Recognition and Classification Using Neural Networks and Response Surface Models
- PDF / 1,579,427 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 53 Downloads / 166 Views
Three-Phase Induction Motors Faults Recognition and Classification Using Neural Networks and Response Surface Models Arismar Morais Gonçalves Júnior · Valceres Vieira Rocha e Silva · Lane Maria Rabelo Baccarini · Maria Luíza Figueiredo Reis
Received: 10 June 2013 / Revised: 2 December 2013 / Accepted: 1 February 2014 / Published online: 14 March 2014 © Brazilian Society for Automatics–SBA 2014
Abstract Three-phase induction motors are very robust machines and can be exposed to a wide variety of environmental and operating conditions, what can result in a number of failures during their use. The early detection of faults can prevent these electric motors degradation or even complete breakdown. In this work, five neural networks models with a decision structure, and five response surface models to classify the engine data operating in normal condition or in four failures conditions that can occur in an induction motor: voltage supply unbalance, initial stator coil windings short circuit, mechanical faults, and broken bars are proposed. The proposed technique utilizes current information and it is robust to load variations. The results show the good performance of the implemented model and its ability to identify the faults established for the proposed work. Keywords Three-phase induction motors · Fault detection · Neural networks · Response surface
1 Introduction The three-phase induction motors are the most widely used engines because their operational characteristics allow the A. M. Gonçalves Júnior (B) · V. V. Rocha e Silva · L. M. R. Baccarini · M. L. F. Reis Departamento de Engenharia Elétrica, Universidade Federal de São João del Rei, Praça Frei Orlando, 170, São João del Rei, MG, Brazil e-mail: [email protected] V. V. Rocha e Silva e-mail: [email protected] L. M. R. Baccarini e-mail: [email protected] M. L. F. Reis e-mail: [email protected]
123
drive systems of almost all types of load. Although they are usually well built and robust, several faults can occur during their use, once these motors can be exposed to a wide variety of environments and operating conditions. The early detection of faults in three-phase induction motors aims to prevent deterioration or even breakdown of these engines. Preventive maintenance can be performed during a shutdown of the machines allowing early detection of failures. The major faults of induction machines can generally be summarized as the following: stator faults resulting in the opening or shorting of one or more turns of a stator phase winding; abnormal connection of the stator windings; broken rotor bar or cracked rotor end rings; static and/or dynamic air-gap irregularities; bent shaft (similar to dynamic eccentricity) which can result in rubbing between the rotor and stator, causing serious damage to stator core and windings; and bearing and gearbox failures (Nandi et al. 2005; Siddique et al. 2005; Jasim et al. 2011). The application of artificial intelligence techniques for eletric motors faults diagnosis allows practical analysis and, in many cases,
Data Loading...
