Real-time implementation of online trained neurocontroller for a BLDC motor

Implementation and experimental verification of a simple neurocontroller (SNC) as a speed controller for a brushless DC motor is presented. The SNC with one weight and a linear hard limit activation function is trained online using the backpropagation algorithm. A modified error function is used to ensure good performance during the online training which has been used without previous off-line training. The SNC was implemented using a computer-interface card mounted on a PC. The driving system performance was investigated by a number of experimental tests for a variety of input reference speed trajectories