A Study on Heuristic Operation Strategy Based on Predictive Data for PV Integrated Hybrid UPS–ESS Systems

This paper presents an operational strategy for a Photovoltaic(PV)-integrated hybrid system combining a Battery Energy Storage System(BESS) and an Uninterruptible Power Supply(UPS). The proposed Hybrid UPS–ESS(HUE) enables UPS to supply backup power during grid outages, while ESS improves efficiency and flexibility under normal conditions. PV generation reduces grid dependency and supplies charging power to the ESS. To manage PV and load forecast uncertainty, a Long Short-Term Memory(LSTM) network is used for short-term prediction and integrated into a heuristic control that adjusts the ESS’s lower State-of-Charge(SOC) bound, traditionally reserved for UPS use. This integration enhances ESS utilization while ensuring UPS reliability. Simulation results demonstrate that the proposed strategy enhances UPS usage rate, ESS availability, SOC stability, and grid power smoothing compared to rule-based methods.