In this proposed system Artificial Neural Network (ANN) based on the Levenberg-Marquardt (LM) algorithm and the Perturb and Observe (P&O) algorithm are deployed for maximum power point tracking (MPPT) in a solar photovoltaic (PV) system. The goal is to conduct a comparative performance analysis of these two algorithms. Using the MATLAB/Simulink environment, a maximum power point tracking energy harvesting system is designed, and the Artificial Neural Network Toolbox is utilized to analyse the developed model. The ANN model is trained with a dataset comprising 1000 samples of solar irradiance, temperature, and voltage. The dataset is divided into three parts: 60% is used for training, 25% for validation, and 15% for testing. The comparative analysis focuses on evaluating the performance of the algorithms in handling the trained dataset, providing insights into their efficiency and accuracy in optimizing the energy harvesting process.