Multi-objective Optimization of Passive Flexible Protection Network Layouts for Rockfall Hazards Based on Evolutionary Algorithms

Rockfall disasters pose a severe threat to infrastructure safety, and the rational design of passive flexible protection network layouts is critical for disaster mitigation. Taking the Guanmenla rockfall area in Ji’an City, Jilin Province as a case study, this paper develops an optimization framework for protection net placement by coupling rockfall numerical simulation with multi-objective evolutionary algorithms. Firstly, a three-dimensional rockfall motion model was established using the RAMMS software. Through 129,000 Monte Carlo simulations, the statistical distribution characteristics of key dynamic parameters—including rockfall trajectories, velocities, kinetic energy, and bounce heights—were obtained. Subsequently, a multi-objective optimization model aimed at maximizing the interception rate and minimizing engineering costs was constructed, with the Pareto optimal solution set solved via the NSGA-II algorithm. The results indicate that the 95th percentile bounce height along the road is 2.04 m, with kinetic energy ranging from 500 to 3200 KJ The optimized scheme adopts a zonal cascading protection strategy, incorporating a total of four flexible protection nets. This configuration achieves a total interception rate of 98.43% with a total engineering cost of 368,440 RMB. Verification through back-simulation in RAMMS demonstrated an error of less than 0.5%. This research establishes a systematic workflow from dynamic parameter extraction to scheme optimization and validation, providing a scientific basis for the design of rockfall protection engineering.