Detecção de anomalias em cargas rotativas utizando o motor de indução como estimador de conjugado

Abstract

The continuous development experienced by industrial processes has led to growing interaction among their component parts. As each part are individually subject to failures and malfunction, making the process operation safe and reliable is a challenge to engineers and technical personal involved. As induction motor is an essential part of the processes subsystems, keep the operation of mechanical load driven by it free of failures is a very important requirement. To do this it is necessary to use condition monitoring and fault detection systems. For economy and safety reasons, fault detection and condition monitoring methods that do not use sensors directly installed on mechanical load are of great interest. In this context, it is proposed in this research work a fault detection aid system based on the mechanical load torque. Two questions are dealt with before: (i) the characterization of typical faults on mechanical loads and (ii) a study on the induction motor acting as an input transducer. The proposed fault detection aid system is validated on laboratory experiments using a load emulator and a water pumping system.