Dynamic scheduling for solid-state green hydrogen storage using improved NSGA-II with binary search algorithm

Hydrogen energy, as a secondary clean energy source, is receiving increasing attention. However, the intermittency and fluctuations of renewable energy sources such as wind and solar present new challenges for green hydrogen (GH₂) storage. Addressing challenges from intermittent renewable energy in GH₂ storage, we develop a dynamic scheduling strategy − dynamic constraint reactor scheduling NSGA-II (DCRS-NSGA-II) − for solid-state hydrogen storage systems composed of multiple Mg-based hydrogen storage reactors (Mg-HSRs). A binary search-based dynamic allocation method, leveraging Mg hydrogen absorption kinetics, is coupled with an improved NSGA-II algorithm to optimize reactor scheduling. To manage declining absorption rates and fluctuating GH₂ supply, the strategy employs dynamic constraints and time-segmented chromosome encoding. Using a case study where the coefficient of variation reaches 51.6 % for a 9.398 kg/day GH₂ supply, the Pareto-optimal solution demonstrates that 15 10 kg-Mg-HSRs meet daily demand with only 15 start-stop cycles and 22 h total operation. This significantly outperforms traditional serial hydrogen absorption (20 reactors, 20 cycles) and parallel hydrogen absorption (169 h operation) strategies. Compared with traditional scheduling strategies, the DCRS-NSGA-II can effectively improve the dynamic response rate under daily fluctuations of GH₂. The hydrogen allocation among different reactors is adjusted reasonably according to the demand, allowing the system to operate more efficiently.