Life cycle cost assessment of PEM water electrolysis systems: a system dynamics–intuitionistic fuzzy bayesian network approach

Abstract

Proton exchange membrane water electrolysis is a core technology for green hydrogen production, but its widespread adoption is hindered by a prohibitively high and uncertain life cycle cost. To address the dynamic complexity and multi-source uncertainties inherent in cost assessment, this paper proposes an integrated modeling framework that combines system dynamics with an intuitionistic fuzzy bayesian network. The system dynamics model captures the macro-level feedback loops driving long-term cost evolution, such as technological innovation, economy-of-scale effects, and other critical factors. To model and infer causal dependencies among uncertain variables that are challenging to specify precisely within the system dynamics model, the intuitionistic fuzzy bayesian network is incorporated, enabling quantification of relationships under conditions of incomplete data and cognitive fuzziness. Through comprehensive simulations, the framework forecasts the cost evolution trajectories. Results indicate a potential 77 % reduction in the unit power cost of a 1 MW system by 2060. Uncertainty analysis revealed that the initial prediction variance for the catalyst layer was approximately 20 %, significantly higher than the 6.5 % for the bipolar plate, highlighting a key investment risk. A comparative analysis demonstrates that the proposed framework achieves a superior forecast accuracy, with a mean absolute percentage error of 4.8 %. The proposed method provides a more accurate and robust decision support tool for long-term investment planning and policy formulation for hydrogen production through proton exchange membrane water electrolysis technology.