Design of a robust adaptive neural network MPPT controller for a photovoltaic energy storage system

Abstract

This paper presents the design of an L1 Adaptive Neural Network (ANN) Maximum Power Point Tracking (MPPT) controller for a Photovoltaic Energy Storage (PV-ES) system consisting of a PV module, a DC–DC boost converter, and a battery storage unit. The controller combines the L1 adaptive control method with a Radial Basis Function (RBF) neural network to approximate unknown nonlinear dynamics and compensate for system uncertainties. This integration enables fast adaptation while preserving robustness, thereby overcoming key limitations of conventional MPPT strategies. Simulation results demonstrate that the proposed L1 ANN-MPPT controller ensures rapid convergence to the maximum power point, reduced steady-state oscillations, and enhanced battery charging efficiency under highly variable irradiance and temperature conditions. These findings quantitatively confirm that the proposed controller eliminates the conventional trade-off between dynamic response and steady-state precision, offering superior speed, stability, and accuracy in maximum power point tracking. Overall, the results validate its effectiveness and highlight its potential for real-time deployment in renewable energy systems.