Sign Regressor LMS Algorithm Based Adaptive Control Algorithm for Multi-objective Solar PV Generator System Connected to Grid

The work proposes sign regressor least mean square algorithm (SRLMS), an adaptive control algorithm for grid-connected solar PV generator system. Proposed adaptive filtering based algorithm is designed for DC to AC conversion with desired voltage and frequency values along with correct phase sequence for grid synchronization. Also, the algorithm aims to incorporate functionalities such as harmonic compensation and VAR generation, load balancing, adaptive DC link voltage regulation, etc. A two-stage three-phase grid-connected PV system is used to test the adaptive algorithm with linear as well as non-linear loads. The system is simulated in MATLAB with Simulink toolbox. Simulation results obtained show the effectiveness of the applied algorithm under steady and dynamic state operations. The total harmonic distortions (THD) of grid current and point of common coupling (PCC) voltage are achieved well within recommended limits according to an IEEE-519 standard.