Techno-economic evaluation of reverse osmosis desalination system considering emission cost and demand response

Abstract

Desalination is a viable solution to meet water scarcity but is considered cost-intensive. The reverse osmosis technique of desalination is commonly used in coastal areas for its cost-effectiveness, but remains energy-intensive. Conventional fossil energy sources emit carbon gas, which has environmental and cost implications, while renewable energy sources are limited by intermittency. This study formulates a mathematical optimisation model using the conventional grid and a diesel generator in comparison with a system that integrates photovoltaic energy to power a reverse osmosis desalination unit. The model is formulated to include a balance between power supply and demand, water demand and water produced, cost functions and a time-of-use demand response (TOU DR) programme. The cost functions include carbon emission cost, DR cost and components cost. The objective of the optimisation model is to minimise the annualised cost of the system and carbon emission, while maximising the quantity of freshwater production subject to different economic and system reliability constraints. The model is implemented in three case scenarios. Case 1 is a system with only the grid as the energy source, Case 2 has the grid and a diesel generator, while Case 3 combines the grid, a diesel generator and photovoltaic energy. The economic impact of the integration of a renewable energy source and a DR programme was examined. The techno-economic analysis of the three cases was performed using levelised cost of energy (LCOE), cost of water (COW) and annualised cost of system (ACS) matrices and the results were compared. The results show that Case 3 brings about a very significant improvement in the reduction of carbon emission, with a value of 751766 kgCO2-e and 648315 kgCO2-e with the introduction of the DR programme and also the lowest ACS ($1158801 with TOU DR). Moreover, the result of a sensitivity analysis shows the impact of increasing the carbon tax rate on the LCOE and COW; again, Case 3 proves to be less affected, as its proportional cost-variation is significantly less than that of Case 1 and Case 2.