A Modified Particle Swarm Optimization Algorithm In Optimization Design Of Highway Subgrade Widening

Road-widening is an important problem in highway construction. It demands economical and practical scheme that can make full use of the landform condition. So optimization and alteration are the main tasks in road rebuilding design to reduce the cost. In this paper, old road widening optimization based on Particle Swarm Optimization was studied with the hope to obtain the economical scheme which can provide reference for project.First, arithmetic principle of Particle Swarm Optimization was expatiated in the paper. Based on LDW and AMPSO, a Modified Particle Swarm Optimization was brought out, which can adjust particle's global searching ability.Then, designing method of the farmiliar subgrade retaining structure: gravity retaining wall and reinforced earth retaining wall were introduced. Optimization models were built respectively for them according to the cost. With the Modified Particle Swarm Optimization, retaining wall designs were optimized in consider of different landform conditions. According to the acquired cost curve, wall height-cost equations of the two retaining walls were established under different filling height. Also, the economical height for each was obtained.Further, optimization model of soil nailing retaining wall was constructed. Based on the Modified Particle Swarm Optimization, two programs were worked out: the searching program of the most dangerous sliding surface and optimization program of soil nailing retaining wall. With the first one, the highest steady height and biggest steady angle of slop without reinforcing were analysed. For the second one, soil nailing retaining wall cost of different height were calculated with the acquisition of correlative expense curve.Finally, the whole optimization model of subgrade widening were built for the fill-cut situation in consider of subgrade retaining wall, slope stabilization, occupying area and so on. In combination of the retaining wall optimization program, the most economical scheme were achieved, providing a reference for the futher design.