System identification
Methodologies of Using Neural Network and Fuzzy Logic Technologies for Motor Incipient Fault Detecti
Methodologies of Using Neural Network and Fuzzy Logic Technologies for Motor Incipient Fault Detecti
Hi-index | 0.00 |
In this paper we examine the problem of knowing the value of steady-state electromagnetic torque in induction motors installed in industrial plants. The models derived from two parametric black-box identification techniques (polynomial and neural) are implemented and tested for two motors and compared with the analytical model provided by the equivalent circuit theory. Both provide better performances when compared to the latter; the best performance is given by the neural model.