FPGA Implementation of Neural Network-Based Controllers for Power Electronics Applications

Abstract

This paper presents an innovative approach for the hardware implementation of neural networks (NN's) in power electronics applications. Effective NN-based applications in power electronics require hardware implementations that exploit the inherent parallelism of NNs. Nonetheless, most NN hardware implementations have been realized using digital signal processors (DSP) or computers that offer serial processing. Nowadays, parallel programmable logic devices, such as the field programmable gate array (FPGA) with embedded microprocessors, have become powerful hardware options, offering low cost, high execution speed, reconfigurability and parallelism. This work intends to exploit the current available resources in commercial FPGAs to implement NN-based controllers for power electronics applications. Simulation and experimental results included in this paper show the viability of exploiting the parallelism and modularity of a low cost FPGA to implement a NN-based controller for a buck converter.