FPGA design of a new frequency relay based on evolutionary algorithms

This paper presents an efficient technique for frequency estimation in Electrical Power Systems (EPS) based on Genetic Algorithms (GAs). Frequency estimation of a distorted electrical signal is modeled as an optimization problem. The motivation concerning the use of the GA approach is related to its intrinsic robustness to noise. The GA is implemented in a FPGA (Field-Programmable Gate Array) device and the estimation process is computed in real-time. This is made possible due to (a) the implicit parallelism of FPGAs (b) the suitable choice of the GA operators to explore this parallelism and (c) the determination of the appropriate architecture for a FPGA that parallelizes the GA routine and ensures its good performance. In order to evaluate the performance of the proposed method, an EPS was simulated using the ATP software for some typical operation conditions. The resulting signals were analyzed by the proposed GA/FPGA approach and they were compared to the performance of a commercial relay. Promising results were found with the proposed methodology.