New Railway Profile Optimization Algorithm And Program Design

In the design process of the new railway, controlling engineering investment andreducing the amount of civil works are always the hotspot in the research of design workers.The optimization design of the vertical alignment, which is under the condition that thehorizontal alignment is set, according to the major technical standards that are selected,determine the design of the line reasonably. In accordance with the objective function andoptimization method, though there are many optimization models now, a lot of work are needto be done in the synthesizing of algorithm and the improving of optimization model.This article is mainly based on the comprehensive analysis of the optimization designalgorithm of the existing new railway, using satisfaction theory and genetic algorithm toresearch the algorithm of the optimization design of the vertical alignment, combining thecalculated results with the AutoCAD drawing, and develop AutoCAD drawing function,achieving the overall process of data input, the optimization of vertical alignmentautomatically and mapping.This article mainly conducted the following research work:(1) Establish the new railway optimization design models of vertical alignmentThe article selects the distance and height of knick point as design variables, introducingthe satisfaction theory to construct a multi-objective optimization function to take engineeringcosts and operating costs into account, with the minimum slope length, maximum gradientand difference in gradients as constraint conditions to establish the correspondingmulti-objective optimization model.(2) The generation of the initial program groupAccording to the characteristics of complex terrain conditions of the new railway, thisarticle use the feasible region to modify the initial ground lines that have many extreme casesand smoothing the ground line, according to the result of processing and main technicalstandard to determine the number and the location of knick point. Using the fitting methodand normal distribution method and combined with a variety of other methods to generate theinitial program group, not only ensures the diversity of initial population, but also provide agood initial solution for later algorithm optimization.(3) Improvement design of the genetic algorithmIn order to make the genetic algorithm can better adapt to optimization design of the newrailway vertical alignment, appropriate improvement is required. This article usemulti-objective fitness function based on satisfaction, and design genetic algorithm from theselection, crossover, and mutation and so on. Use the roulette wheel selection operator, usethe distributed in the whole solution space of the whole arithmetic crossover operator, and mutation operator is adopted evenness variation, variation and inconsistency combinedadaptive variation. That make the genetic algorithm can better solve the problem of complexvertical alignment optimization design, determine the termination criteria based on thesatisfaction at the same time, and make the algorithm more operability.(4) AutoCAD graphics developmentCombined with specific engineering, the optimization algorithm according to thecalculated results with the AutoCAD, and combined with the problem, using VBAprogramming language to develop auxiliary CAD automatic optimization procedure based onExcel data source, enable the system to implement the vertical alignment optimization designof the new railway, and use AutoCAD map the vertical alignment of line directly at the sametime.By specific engineering test, using the optimization algorithms combined satisfactionwith genetic algorithm is better able to get the optimization design scheme of verticalalignment of the line. That will play a very important role to improve efficiency of design andscientificity of decision-making.