Novel data-driven energy management of a hybrid photovoltaic-reverse osmosis desalination system using deep reinforcement learning

•Novel deep reinforcement learning-accelerated energy management system is developed.•Energy management of a hybrid grid-connected desalination process is investigated.•The partial observability dilemma involved in PV-powered energy systems is resolved.•Developed model was superior to 4 advanced deep reinforcement learning algorithms.•Developed algorithm was benchmarked against a powerful energy management system.