Design of Frost Protection System for Cold Region Tunnels Based on Improved Genetic Algorithm

The aim of this study is to improve the performance of the genetic algorithm (GA) in optimization problems. By employing strategies such as rank-based fitness assignment, adaptive crossover probability, and non-uniform mutation, the improved genetic algorithm (IGA) effectively reduces the probability of premature convergence, increases population diversity, and optimizes the search process. Experimental results demonstrate that IGA significantly enhances solution quality and convergence speed compared to traditional genetic algorithms. Furthermore, the introduction of hybrid genetic algorithms (HGAs) further improves solution quality and convergence speed. The effectiveness of adaptive parameter adjustment in IGA is also demonstrated, providing the algorithm with enhanced robustness in handling different problem types. These findings provide theoretical and empirical support for the application of IGA in complex optimization problems.