Bancada experimental de controle de velocidade sem sensor para máquinas elétricas rotativas.

Abstract

In a feedback loop control system, the use of sensors is fundamental to determine de error signal between the measured quantity and the desired value. However, in situations that the controlled variable can’t be measured, either for lack of instruments or to reduce the project’s costs, it is necessary to use topologies which allows determine the error signal indirectly from other vriables of the system. Thus, the Kalman Filter can be implemented to observe the measureble quantities and, from the model, determine the value of the variables that can’t be measured. In this work, will be presented and discussed the project of an electrical power conversion system that withdraw energy from the power distribution network and supply it to, with the power factor as close as possible to 1, a permanent magnet continuous current electrical machine, which will have its angular velocity estimated indireclty and optimally controlled.