An Improved Variable Step Size LMS Harmonic Current Detection Method for Active Power Filter

Abstract

With the development of adaptive filter theory, many adaptive filtering techniques are used for harmonic current detection, and achieve some good results. But it is difficult to detect harmonic current in the low signal-to-noise ratio (SNR) condition when using conventional algorithms. Based on adaptive noise cancellation theory, this paper proposed a modify variable step size least mean square (MVSS-LMS) algorithm for harmonic current detection. The step size of the proposed method is controlled by the improved generic Sigmoid function relationship with the error signal. The MVSS-LMS has fast convergence and good tracking performance of fundamental and harmonic components from distorted signals, and simulation results prove the algorithm having a good feature.