Simplified Modelling and GA-Optimisation of a Proportional-Integral Load Frequency Controller for Stand-Alone Small Hydropower Plants

Abstract

This paper presents load frequency control for standalone small hydropower plants using an electronic load controller (ELC) tuned via Genetic Algorithm (GA) optimisation. The ELC for a standalone generating system is developed using an Arduino microcontroller, solid-state relays, a zero-crossing detection circuit, and heating elements to serve as damper loads for channelling the generator's excess power during load changes using phase cut modulation. A high-fidelity (HF) MATLAB/Simulink model of the physical system was developed using the physical characteristics of the experimental model while a simplified transfer function model of the system with fast execution time was derived using output data obtained from the real-time experimental model and GA optimisation. GA optimisation was also used to tune a PI controller to minimise the output frequency error of the simplified model. The obtained optimised PI gains were then deployed to the HF model and the physical system in real time. Comparisons of the results by simulation and real-time experimentation show that the parameters of the phase modulated PI-controlled ELC can be tuned relatively quickly via GA optimisation with the simplified model to achieve satisfactory performance.