Enhancing hybrid renewable system performance through load shifting: A multi-objective optimization and forecasting approach

• A comprehensive multi-objective optimization framework was developed to simultaneously optimize cost, reliability, renewable energy utilization, and emissions. Four objectives which were re-cost of energy, loss of power supply probability, dumped renewable energy, and Carbon Dioxide emissions were optimized in pairings to evaluate techno-economic and environmental performance. • A sensitivity analysis and Monte Carlo simulation was conducted to account for uncertainties in fuel prices, O&M costs, component efficiencies, load demand, along with renewable energy resources availability and mention the effect of uncertainty of each one on the system performance. • A comprehensive multi-objective optimization with demand-side management load shifting was developed to simultaneously optimize cost, reliability, renewable energy utilization, and emissions and assessing load shifting impact on cost, reliability, renewable utilization and emissions. • A Neural network model was developed for predicting system cost of energy, reliability, renewable utilization, and emissions for both scenarios (with and without DSM) and verify the accuracy of its predicted data. • Using the Developed Neural network model for both scenarios (with and without DSM) for forecasting the system cost of energy, reliability, renewable utilization, and emissions with an annual load growth rate over several years This paper investigates the impact of demand-side management (DSM), specifically load shifting, on the techno-economic-environmental performance of a standalone hybrid renewable energy system (HRES). A multi-objective optimization model is developed using the Mayfly Algorithm to explore trade-offs between cost, reliability, renewable utilization, and emissions across four key objective pairings. The optimal configurations are evaluated with and without a load-shifting strategy. System robustness is rigorously tested through extensive Monte Carlo simulations (500 scenarios) accounting for uncertainties in economic and technical parameters. A neural network is then employed to predict system performance and conduct long-term forecasting (5 years) under load growth. Comparative analysis reveals that integrating load shifting reduces the cost of energy by 3–5%, significantly decreases dumped renewable energy, and further slashes emissions—for instance, cutting CO₂ by over 50% in the COE–CO₂ case—while maintaining high reliability. This work underscores DSM as a critical tool for maximizing the benefits of hybrid renewable systems.