Optimisation of Sustainable Energy Harvesting in Hybrid Wind-Solar Systems Using AI-Driven Predictive Maintenance Models

Abstract

Hybrid wind-solar systems have become a key solution for sustainable energy generation; however, their inherent complexity and fluctuating environmental conditions pose significant challenges in terms of operational reliability and maintenance optimization. This study proposes an AI-driven predictive maintenance model integrated with a multi-objective optimization framework to enhance the efficiency and economic viability of hybrid wind-solar systems. The predictive maintenance model employs Deep Neural Networks (DNN) to predict equipment failures with high accuracy, enabling proactive maintenance scheduling and minimizing unplanned downtimes. In parallel, a multi-objective optimization algorithm using Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) is implemented to achieve optimal trade-offs between maximizing energy production and minimizing maintenance costs. The integrated approach achieved a 25% reduction in system downtime, a 30% decrease in maintenance costs, and a 15% increase in overall energy production compared to traditional strategies. The results indicate that the proposed framework significantly improves system reliability, optimizes energy harvesting, and supports the economic sustainability of hybrid renewable energy projects. This study provides a comprehensive solution for addressing the complexities of hybrid systems, offering valuable insights for practitioners and policymakers seeking to implement advanced AI-driven maintenance and optimization strategies in renewable energy systems.