Abstract:

To improve the efficiency and lifespan of wind-solar hydrogen production systems and resolve the mismatch between renewable energy intermittency and electrolyzer operational characteristics, this paper proposes a power allocation strategy based on hybrid electrolyzer operational features. An improved fuzzy C-means clustering algorithm incorporating dynamic time warping (DTW) distance is introduced to accurately capture the temporal fluctuation characteristics of renewable generation output, enabling effective classification of typical daily profiles from annual wind and solar power data. Based on the operational differences between alkaline electrolyzers (AEL) and proton exchange membrane electrolyzers (PEL), a power allocation strategy is formulated with AEL as the main hydrogen production unit and PEL as the auxiliary regulation unit, while an economic operation evaluation model is established for the integrated system. Simulation results demonstrate that the proposed strategy enhances hydrogen production and annual profit while significantly reducing renewable energy curtailment, lifespan penalty costs, and AEL power fluctuation standard deviation, thereby validating its effectiveness.

Key words: wind-solar hydrogen production system, hybrid electrolyzers, power allocation strategy, fuzzy c-means clustering, typical day