DSP implementation of state observers for electrical drive with elastic coupling

Abstract

Article presents the results of implementation of three discrete methods of estimation of the state vector of the dynamic system-Luenberger observer, Kalman filter and neural network estimator. Implementation was done in C ++, in an VisualDSP 5.0 environment, assuming the use of algorithms to observe the state of the mechanical part of the electric drive, characterized by a elastic connection with the working machine. Autors had used object-oriented programming technique.