Power system harmonic parameter estimation using Bilinear Recursive Least Square (BRLS) algorithm

The impact of nonlinear loads produces harmonic pollution in electrical power system. It is considered as a serious concern now a day. Whereas, many algorithms have been proposed for harmonic estimation to improve the power quality performance but till date the accurate estimation of power quality parameters remains a challenge. In this paper a non-linear adaptive algorithm, called Bilinear Recursive Least Square (BRLS), has been applied for the first time for estimating the amplitudes, phases and frequency in case of time varying power signals containing harmonics, sub harmonics, inter harmonics in presence of White Gaussian Noise. The technique is applied and tested for both stationary as well as dynamic signals containing harmonics. Practical validation of the proposed algorithm is also made along with the real time data obtained from a Variable Frequency Drive (VFD) panel used for controlling the speed and torque of the induction motor used at a large paper industry. Comparison of the results achieved with the proposed BRLS algorithm with two recently reported non-linear adaptive algorithms, Volterra Least Mean Square (VLMS), and Volterra Recursive Least Square (VRLS), reveals that the proposed BRLS algorithm is the best in terms of estimation accuracy and computational time.