Non-parametric induction motor rotor flux estimator based on feed-forward neural network

The conventional induction motor rotor flux observer based on current model and voltage are sensitive to parameter uncertainties. In this paper, a non-parametric estimator feed-forward neural network is proposed. This operating without parameters therefore it independent from trained using Levenberg-Marquardt algorithm offline. All the data collection, training testing process fully performed in MATLAB/Simulink environment. A forced iteration of 1,000-epochs imposed process. There overall 603,968 datasets used modeling four-input two-output capable providing estimation for field-oriented control systems with 3.41e-9 mse elapsed 28 minutes 49 seconds time consumption. proposed tested reference speed step response result indicates that improves observers