Application of universal learning networks to PV-supplied DC motor drives

This paper describes the application of Universal Learning Networks (ULNs) to the speed control of a separately excited dc motor drives fed from Photovoltaic (PV) generators via dc-dc buck-boost converters. In this application, two ULNs are used: the first is the Universal Learning Network Identifier (ULNI) that used to emulate the dynamic performance of the motor system. The second is the Universal Learning Network Controller (ULNC) that used to control the converter duty ratio so that the motor speed can follow an arbitrary reference signal. In addition to that, the overall system can operate at the Maximum Power Point (MPP) of the PV source. The free parameters of both networks are updated online by the forward propagation scheme, which is considered as an extended version of the Rel Time Recurrent Learning (RTRL). The simulation results showed good performance for the controller and the identifier networks. Promising results are also observed when the identifier network is trained in an environment contaminated with noise.