Battery versus hydrogen storage for renewable-powered RO desalination: A technoeconomic and environmental analysis in coastal Saudi Arabia

Research output: Contribution to journalArticlepeer-review

Abstract

This study presents a comprehensive and systematic assessment comparing battery and hydrogen energy storage systems for powering reverse osmosis (RO) desalination. The research introduces a multi-scenario optimization framework applied to ten coastal locations in Saudi Arabia, encompassing photovoltaic (PV), wind turbine (WT), hybrid PV/WT, and PV/WT/diesel configurations, each integrated with either battery or hydrogen storage. System optimization using the Artificial Rabbit Optimization (ARO) algorithm for a four-stage RO unit reduced energy demand by approximately 41 % relative to a single-stage design. Two optimization modes were employed: a single-objective formulation minimizing the Levelized Cost of Water (LCOW) and a multi-objective formulation minimizing both LCOW and Annual Carbon Emissions (ACE). Battery storage proved most cost-effective, achieving an LCOW of 1.66 $/m3 in PV-powered systems, whereas hydrogen storage minimized ACE to 10 tons/year in wind-powered configurations. No configuration was universally optimal, underscoring trade-offs between economic and environmental objectives.

Original languageEnglish
Article number152466
JournalInternational Journal of Hydrogen Energy
Volume197
DOIs
StatePublished - 5 Jan 2026

Bibliographical note

Publisher Copyright:
© 2025 Hydrogen Energy Publications LLC

Keywords

  • Desalination
  • Hydrogen storage
  • Renewable energy
  • Reverse osmosis
  • Saudi Arabia

ASJC Scopus subject areas

  • Renewable Energy, Sustainability and the Environment
  • Fuel Technology
  • Condensed Matter Physics
  • Energy Engineering and Power Technology

Fingerprint

Dive into the research topics of 'Battery versus hydrogen storage for renewable-powered RO desalination: A technoeconomic and environmental analysis in coastal Saudi Arabia'. Together they form a unique fingerprint.

Cite this