Detecção de defeitos em motores de indução trifásicos através da avaliação do comportamento térmico

Abstract

The three-phase induction motor is widely used in modern society as it can be used in almost any application. Although they have high durability, the motors are subject to stresses that can cause defects and failures, and consequently, undesirable stops. This work presents a contribution to the study of defects detection methods in three-phase induction motors, in which two mathematical models were developed that simulate the thermal behavior of the motor for certain operating conditions, in order to verify which of the models presents the best result in comparison with the data obtained in the performed tests. From input data (three-phase voltage, three-phase current and slip) and output data (temperature), the objective was to find a model that describes the operating conditions of the motor. The models selected for the study were Multiple Linear Regression and State Space Modeling. For the study, tests were performed with the normal rotor for determination of data and comparison with the rotor with open bar, in addition, the data obtained were normalized for application of the two methods. From the obtained results, the performance of each method was analyzed, identifying the most adequate to model the system and detect defects, as well as to indicate abnormal operating conditions. Validation was carried out with new data, thus confirming that the State Space model was the most effective for estimating temperature and, consequently, for detecting defects.