In order to improve voltage regulation and power efficiency, this study proposes a renewable energy management system that combines solar photovoltaic (PV) with a bicycle dynamo generator using a Multiple-Input Single-Output (MISO) SEPIC converter. The SEPIC converter is perfect for renewable energy applications because it can manage varying input voltages while producing a steady output. The variation in sun irradiance, which impacts power generation, is a significant difficulty in solar PV systems. Inefficient energy extraction results from the slow response times and steady-state oscillations of traditional Perturb and Observe (P&O) MPPT methods. An Artificial Neural Network (ANN)-based MPPT algorithm is used to get around this, enabling accurate tracking, quicker convergence, and real-time adaptation to shifting environmental conditions. MATLAB/Simulink simulations are used to analyse the system and assess its power efficiency, voltage stability, and adaptability. The outcomes show that by increasing energy harvesting efficiency, the ANN-based MPPT performs better than traditional techniques. By continuously modifying system parameters, this AI-driven method guarantees efficient power utilization, which makes it ideal for off-grid energy solutions applications. The suggested system offers a scalable and effective solution for contemporary renewable energy applications, supporting intelligent and sustainable energy management. © 2025 Elsevier B.V., All rights reserved.