The integration of solar panel (PV) systems with the grid poses challenges to electrical power quality due to harmonic distortion caused by non-linear loads. To address this issue, a shunt active power filter (SAPF) is incorporated. The work presents a pioneering method designed to identify reference harmonic currents within a SAPF linked to a two-level voltage converter. Fuzzy logic and neural networks, pivotal branches of artificial intelligence, are concurrently employed through an adaptive neuro-fuzzy inference system (ANFIS). ANFIS is used for detecting reference harmonic currents, and Pulse Width Modulation (PWM) is adopted for pulse generation. The study also explores the integration of a photovoltaic system harnessing solar energy, coupled with SAPFs. A comparative analysis of SAPF performance using the ANFIS algorithm and the instantaneous active&reactive (P-Q) technique is presented. This study employs simulation analyses across three different scenarios to assess a range of system parameters. These parameters include source current variations both with and without the use of SAPF utilizing ANFIS and P-Q analysis. The primary objective of this study is to provide a deeper understanding of the efficacy of the ANFIS algorithm in enhancing power quality and addressing harmonic pollution.