Enhanced PV Energy Harvesting: An Integrated Approach using ANN and SMC

Harnessing renewable energy through photovoltaic (PV) modules is becoming increasingly prevalent. PV power output depends on numerous factors including solar irradiance, temperature, partial shading conditions, and operating voltage. Of these, only the operating voltage can be readily controlled, which makes effective voltage regulation crucial for maximum power extraction. Although numerous maximum power point tracking (MPPT) techniques exist, an integrative approach that ensures optimal operation under variable conditions is needed. This paper introduces a solution that combines an artificial neural network (ANN) with a sliding mode controller (SMC) in an MPPT converter. The ANN effectively generates a reference signal for the maximum power point under various rapid changes, while the SMC ensures smooth voltage output from the boost converter, even under fluctuating input voltage. This combination results in a quick response with minimal DC voltage ripples, maximizing energy extraction from the PV system. The paper presents the performance of this ANN-SMC combination for boost converter-based MPPT. To mirror real-world conditions, a Hardware-in-the-Loop (HIL) setup was used, built using two OPAL-RT modules. The paper contains comprehensive results obtained from the HIL setup, providing a robust evaluation of the proposed method under different case scenarios.