Multiobjective swarm intelligent algorithm based on sensor layout optimization method

Proper layout of tunnel sensors is essential to ensure the long-term safe operation of the infrastructure. In addition to incomplete consideration of influencing factors, the optimised layout of multi-objective sensors in existing studies is coarse. Therefore, this study focuses on achieving comprehensive monitoring of the tunnel section under the condition of limited monitoring points in order to balance monitoring performance and layout cost. On this basis, this paper proposes an optimization method based on a multi-objective swarm intelligence algorithm：multi-objective snow ablation optimization algorithm (MOSAO), and experimentally compares it with multi-objective particle swarm optimization (MOPSO), nondominated sorting genetic algorithm (NSGA-II), and strength Pareto evolutionary algorithm (SPEA). Evaluation metrics such as Generational Distance (GD), Spacing (SP), Maximum Spread (MS) and Hypervolume (HV) were used to quantitatively analyse the performance of each optimised algorithm. The data set of the simulation experiment is formed based on the information of a city shield tunnel. The results show that the MOSAO algorithm significantly outperforms other algorithms in GD, MS and HV metrics, indicating faster convergence and robustness in solving the layout optimisation problem, with better Pareto front and providing uniform distribution. This study provides new insights for the application of multi-objective optimization algorithms in the field of sensor layout optimization.