Three-dimensional subway alignment recreation considering tunnel construction gauges

Abstract

Due to errors generated by measurement, construction and structural settlement, railway tracks in subway tunnels will inevitably deviate from their originally-designed alignments. Thus, for subsequent construction processes and for guaranteeing operational safety, it is crucial to recreate the subway sections with severe deviations based on actually-measured tunnel point data. Unfortunately, to date, little attention has been devoted to the recreation of subway alignments while considering the relevant complex constraints. Hence, this paper focuses on recreating 3-dimensional (3-D) constrained alignments of subways. Primarily, an optimization model for recreating subway alignment geometry is developed to minimize the total deviations between recreated alignments and actually-measured points. In this model, complicated tunnel construction and alignment geometric constraints are also specified. Afterward, a two-stage solution method is developed. In the first stage, a “cross-vertical-integration” approach is designed to generate 3-D feasible regions for subway alignments by handling constraints. Based on these feasible regions, recreated alignments are then searched through a swing iteration heuristic incorporating a “horizontal-vertical-integration” search strategy. Finally, by being applied to a real-world subway project, the computer-generated alignment solution is compared with two other solutions, i.e., a manually-designed solution produced by experienced engineers and a recreated alignment generated by a genetic algorithm. The effectiveness of the proposed method is elaborated through detailed resulting data analyses.