An intelligent computing technique to analyze the vibrational dynamics of rotating electrical machine

In this study, bio-inspired computational intelligence is exploited to analyze the nonlinear vibrational dynamics of rotating electrical machine (VD-REM) model by applying artificial neural networks (ANNs), genetic algorithms (GAs) and active-set methods (ASMs). The superintended mathematical relation of VD-REM is modeled with ANNs by employing an unsupervised error function. Design parameters of the networks are trained with meta-heuristic approach based on GAs, used as a tool for effective global search method, hybrid with ASM for efficient local search. The design scheme is evaluated for VD-REM models by taking different values of shaft stiffness along with an amplitude of force and parametric excitations. The performance of the proposed scheme is validated through the comparison of results from Adam numerical method with the help of different measures based on mean absolute derivation, root mean square error, and Nash-Sutcliffe efficiency.