Low-Cost Implementation of an Adaptive Neural Network Controller for a Drive with an Elastic Shaft

This paper deals with the implementation of an adaptive speed controller applied for two electrical machines coupled by a long shaft. The main parts study are synthesis neural and hardware using low-cost system based on STM Discovery board. framework between control system, power converters, motors is established ARM device. A radial basis function network (RBFNN) used as controller. net coefficients updated (online mode) to ensure high dynamics correct work under disturbance. results contain transients achieved in simulations experimental tests.