The Research On Road Alignment Optimization Method Based On Genetic Algorithms

This article start from the theory of road designing optimization,study the road alignment problems, creat a road alignment optimization model based on the cost. From the perspective of the theory of highway optimal design, this article researches and discusses the problems of the selection of highway route, and proposes a cost-based model for its selection. The purpose of this research is to improve the efficiency of the selection of highway route,optimize its result, and put forth new ideas and methods for designers and researchers, so that they can find the most optimal space alignment plan of newly-constructed highway more efficiently, scientifically, and accurately.With the given starting point and ending point of a highway, the model transforms various factors interrelated road alignment to cost factor, identifying a problem of optimizing highway space alignment plan with a minimum cost. Based upon GIS as a platform, the model imports the original CAD digital terrain mapstratified into GIS after grooming The original CAD digital terrain map, sets up road regional a spatial database of the region of highway selection, adds different fields property to different layers in accordance with the requirement of the model data, and solves the problem by genetic algorithm.In the geographic information system, first, according to the control point, a series of new highway alignment space programs are randomly generated, and each alternative's linear vertical section and horizontal alignment is designed automatically in detail. Then all the cost resulted from the candidate road alignment itself shall be calculated, including essential construction cost, earthworks cost, coverage costs, the cost of bridge or tunnel construction, maintenance cost; at the same time, the expenditure of road users'shall be calculated, including the cost of traveling time and vehicledriving cost of driving ; penalty fees shall be paide when if the road passes thrill through the wetlands and arable land, impose some punishment costs, so that the alignment as far as possible to try to avoid these high-cost regions. The total cost here above , will be part of the cost of the candidate alternative plans,as a cumulative sum of the total cost of the program, in GIS when the automatic highway alignment design can not fully meet all the restrictions of the design specifications, punish function shall be applied to make up. In genetic algorithm, we regard the alternative plans as population, the control points in each plan as genes in population's chromosome, and the cost of the plan the users'expenditure and the punish function fees as the fitness value of the population. An optimal highway alignment will be worked out by initializing population, calculating the population's fitness value, selecting, overlapping and transforming genetic algorithm to develop and meet the conditions.The model of the highway alignment selection is effective, feasible through digital tests, and it proves to have good applicable value.