Hardware Implementation of Back Propagation Neural Networks for DC-DC Converter

This paper presents cost effective hardware realization of back propagation neural network for DC- DC Converter. Artificial neural network (ANN) has become very popular in many control applications due to their high speed operation in real time applications which can be achieved if the networks are implemented using parallel hardware architecture. To achieve these parallelism, modularity and dynamic adaptation of ANN, reconfigurable Field Programmable Gate Array (FPGA) is used which offer flexible designs, savings in cost and design cycle. The training patterns for the neuron controller were obtained from the synergetic controller which is implemented in simulink and the designed neural network controller was implemented in FPGA using Very High Speed Integrated Circuits (VHSIC) Hardware Description Language (VHDL). The simulation was done by using MATLAB and Xilinx ISE software.