A fast on-line neural-network training algorithm for a rectifier regulator

F. Kamran, R.G. Harley, B. Burton, T.G. Habetler, M.A. Brooke
1998 IEEE transactions on power electronics  
This paper addresses the problem of deadbeat control in fully controlled high-power-factor rectifiers. Improved deadbeat control can be achieved through the use of neuralnetwork-based predictors for the input-current reference to the rectifier. In this application, on-line training is absolutely required. In order to achieve sufficiently fast on-line training, a new random-search algorithm is presented and evaluated. Simulation results show that this type of network training yields equivalent
more » ... yields equivalent performance to standard backpropagation training. Unlike backpropagation, however, the random weight change (RWC) method can be implemented in mixed digital/analog hardware for this application. The paper proposes a very large-scale integration (VLSI) implementation which achieves a training epoch as low as 8 s. Index Terms-Current regulator, neural network, on-line training, power electronic rectifier.
doi:10.1109/63.662857 fatcat:yizghzt6sjhhlcjti3j6ecxspm