RLMMN adaptive filtering based control scheme for multi-objective GPV system

Abstract

This paper presents a robust least mean mixed norm (RLMMN) adaptive control scheme for multi-objective grid integrated solar photovoltaic (GPV) system under abnormal grid conditions. The control scheme serves manifold objectives such as load balancing, harmonics elimination, improving active power penetration into the distribution network while having active shunt filtering capabilities. The estimated perturb and observe (EPO) scheme is used to harvest crest power from solar photovoltaic (PV) array under variable atmospheric conditions. The proposed control scheme is robust under impulsive power system environments and has the advantages of low steady-state oscillations, low complexity, less mean square error and good dynamic response. The comparative performance with the conventional algorithms depict the satisfactory performance under dynamic condition. The DC link voltage is adapted in proportion with PCC voltage to reduce VSC (Voltage Source Converter) converter losses and its tripping under weak distributed grid conditions. Test results demonstrate the satisfactory behavior under steady-state and dynamic conditions of load unbalancing, variable solar insolation and grid voltage fluctuations. The total harmonic distortions (THDs) of grid currents are observed within limits of grid codes compliance according to an IEEE 519 standard.