The Perturb and Observe (P&O) algorithm is one of the most widely adopted Maximum Power Point Tracking (MPPT) methods in photovoltaic (PV) systems due to its simplicity and low computational requirements. However, the P&O method, which heavily relies on the step size selection, suffers from power oscillations around the MPP under steady-state conditions and may incorrectly reflect power variations under rapidly changing conditions. This paper presents an adaptive moment estimation (Adam) by estimation of the first and second moments of the gradients. The insensitivity to the initial learning rate eliminates the influence of the step size selection. The simulation and experimental results show that it has a faster transient response and smaller steady-state power oscillations than the P&O method regardless of the step size.
