Robust estimation of power system harmonics using a hybrid firefly based recursive least square algorithm

• FA–RLS algorithm is proposed in this paper for harmonic estimation in power system.
• Proposed algorithm is validated with experimental data in real time environment.
• Performance of the proposed FA–RLS algorithm is evaluated in presence of noise.
• FA–RLS algorithm outperforms others in terms of accuracy, convergence and computational time.

This paper presents a new and hybrid algorithm based on Firefly Algorithm (FA) and Recursive Least Square (RLS) for power system harmonic estimation. The hybrid FA–RLS algorithm is developed for estimating harmonics, inter harmonics and sub harmonics from a distorted and noise corrupted power signal. The basic strategy of the proposed algorithm is to integrate FA for getting the optimum initial weights for RLS algorithm that sequentially updates the unknown parameters (weights) of the harmonic signal. Simulation and practical validation is made with experimentation of the algorithms with real time data obtained from a solar connected inverter system. Comparison of results amongst recently proposed Artificial Bee Colony Least Square (ABC–LS), Bacteria Foraging Optimized Recursive Least Square (BFO–RLS) and FA–RLS algorithms reveals that proposed FA–RLS algorithm is the best in terms of accuracy, convergence and computational time.