This paper investigates the performances of adaptive linear neural network (ADALINE) based least mean square (LMS) control algorithm for distributed static compensator (DSTATCOM) to mitigate the power quality disturbances in a weak AC grid with multifarious load. The algorithm is based on the tracking of active unit voltage templates to maintain minimum error. This adaptive algorithm helps to extracts fundamental weighs of load currents to estimate controlled reference current signals. These reference current signals further compared with the actual weak AC grid current signals to generate controlled switching signals for the DSTATCOM to enhance the power quality. This adaptive control algorithm has been investigated successfully in current harmonics alleviation, voltage regulation, load balancing and reactive power compensation in the presence of non-linear load and induction furnace load at the point of common coupling. © 2019 IEEE.