Carbon Emission Reduction of Tunnel Construction Machinery System Based on Self-Organizing Map-Global Particle Swarm Optimization With Multiple Weight Varying Models

The impact of greenhouse gas emissions from construction activities on the environment is becoming increasingly evident. It is imperative to adopt appropriate techniques and management measures to restrain carbon emissions from the construction industry, especially the construction of infrastructure such as transportation tunnels. Through the life cycle assessment theory and carbon emission factor method, we established a carbon emission model that can be applied to quantify the carbon emissions of the mechanized construction system of tunnels constructed using the drill-and-blast method. This model exhibits considerable non-convexity after incorporating various constraint conditions, and thus, its task of finding the optima becomes an NP-hard problem. Then, we propose a self-organizing map-global particle swarm optimization algorithm that incorporates multiple weight varying strategies and self-organizing mapping networks for adaptive adjustment of particle trajectories to improve the ability to search for optimums. The feasibility and advantages of the proposed algorithm were verified through a series of experiments. Finally, it was combined with a quantitative model and applied to the optimization of the construction machinery unit combination for the F4 section of the Wushaoling Tunnel. Good optimization results were achieved; the optimized configuration reduced the cycle duration from 660 min to 432 min, while the total emission of carbon equivalent was reduced by 88.4% compared to the original machinery system configuration, which offers reference and inspiration to other studies on carbon emission reduction in the construction of tunnels or other infrastructures.