Two-stage optimization of an integrated renewable CCHP multi-energy system for a university building in Melbourne

This study evaluates an integrated renewable combined cooling, heating, and power (CCHP) system for the Victoria University building in Melbourne, Australia. Annual building loads and associated emissions were simulated in DesignBuilder, and the resulting hourly profiles were used to assess a system integrating photovoltaic–thermal (PVT) collectors, an organic Rankine cycle (ORC), cooling and heating subsystems, and sensible thermal energy storage. A two-stage optimization framework was developed using Response Surface Methodology (RSM) and an artificial neural network (ANN) coupled with NSGA-II to maximize exergy efficiency and minimize system cost rate. The RSM optimum achieved an exergy efficiency of 18.62% and a core-system cost rate of 2.418 $/h. The ANN-based optimization predicted 18.6115% and 2.4592 $/h, respectively; re-evaluation of the ANN solution in the EES thermodynamic model yielded 18.6340% and 2.4485 $/h, corresponding to relative differences of 0.12% and 0.43% from the ANN predictions. Over the annual operating period, the system generated 138,450.0 kWh of electricity against a building electricity demand of 99,737.6 kWh. On-site generation directly covered 89.27% of annual electricity demand, while 10,702.0 kWh was imported from the grid and 49,414.4 kWh of surplus electricity was exported. The system also produced 82,150.0 kWh of heat and 94,820.0 kWh of cooling; chilled-water and hot-water storage capacities were 1200 kWh and 950 kWh, respectively. Compared with the conventional baseline, annual operational CO₂ emissions fell from 112.53 to 8.78 t/year, a reduction of 92.20%; accounting for exported electricity, the net carbon balance was −31.74 t CO₂/year. The proposed configuration supports integrated energy management based on thermal storage and grid-interactive power exchange. In warm seasons, excess cooling is stored during off-peak hours and used during peak demand, whereas in cold seasons, surplus heating is stored for later use, and surplus electricity generation is exported to the local electrical grid to balance generation profiles, maximize economic revenue, and reduce reliance on grid imports without requiring dedicated electrochemical battery storage. Overall, the results indicate that the system can improve multi-energy utilization, reduce grid dependence and operational emissions, and provide a flexible energy-management option for large educational buildings.