Optimization and simulation of a grid-connected PV system using load forecasting methods: A case study of a university building

Abstract Distributed generation represents a paradigm shift from the traditional electric grid to localized of power along with capability incorporating renewable energy (RE) sources into mix. Responding need for sufficient analysis, simulation, and study feasibility distributed generation, this aims design hybrid system university building analyze its economic benefits. The viability existing load forecasting methodologies systems is also presented in paper. was determined through predictive modeling profile using historical data optimization machine learning methods, namely auto-regressive integrated moving average (ARIMA) long short-term memory (LSTM). Simulations were run HOMER Pro. Results show that grid-connected solar photovoltaic (PV) installed on roof coupled an storage (ESS) will provide most benefits because it yields reduced cost (COE) per kilowatt-hour (kWh) building. This shows line efforts transition towards clean energy, RE not only have benefits, but can ensure security environmental sustainability.