A novel method for mining the most hazardous failure path of a cascade dam group under risk transfer effect

Cascade dam groups are critical to water resource management but face complex, interconnected risks due to hydraulic interactions and structural dependencies. Traditional risk assessment methods often overlook the transfer effects of risks within such systems, leading to incomplete safety evaluations. This study develops a comprehensive risk transfer evaluation indicator system that incorporates inherent resistance characteristics and hydraulic risk transfer effects. The relative importance of each indicator is determined by combining ranking relation analysis (G1) and entropy weighting methods. A risk transfer model is then devised, and the risk transfer coefficient is quantified using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). To identify the most hazardous failure path, the minimum system reliability criterion is introduced, employing structural non-probabilistic reliability analysis theory optimised via the Harris Hawk optimisation (HHO) algorithm. A case study of a three-dam system demonstrates the method’s effectiveness, identifying the path from Dam B to Dam A as the most critical. This approach provides technical support for the failure risk assessment of cascade dam groups and offers key insights for prioritising safety interventions.