基于遗传算法的直埋供热弯管的优化设计  

Optimum Design of Buried Heating Elbow based on Genetic Algorithm

作　　者：高晔明 王国伟[2] 王飞[2] 

机构地区：[1]太原钢铁(集团)设计院,太原030009 [2]太原理工大学环境科学与工程学院,太原030024

出　　处：《太原理工大学学报》2011年第4期415-419,共5页Journal of Taiyuan University of Technology

基　　金：山西省科技攻关基金资助项目(2007032069)

摘　　要：直埋供热管道弯头的设计过程是一种试算的过程,即给定曲率半径和壁厚等参数来进行应力验算,直到应力验算合格为止。提出以曲率半径、壁厚和强度条件等作为约束条件,以弯头成本最低为目标函数,进行弯头的优化设计。通过对DN=1 000 mm以下各种规格管道的优化计算,得出了两臂无限长水平弯头的最优曲率半径和最优壁厚,得到了直埋弯头采用遗传算法进行优化设计的结论;且应用惩罚函数法来处理非线性约束问题,收敛稳定一致,收敛速度快。"遗传算法"应用到直埋供热管道弯头的工程设计计算中,不管弯头尺寸是否受到严格的场地限制,均可优化设计,既增强了弯头布置的灵活性,也提高了弯头的可靠性,降低了工程投资。The design process of elbow of the direct-burying heating pipeline is one kind of pilot calculation process, namely assigns parameters as radius of curvature and pipe wall thickness to carry on the stress checking calculation, until stress checking calculation is qualified. In this article the radius of curvature, pipe wall thickness and intensity condition were used as the constraints, and the lowest cost of elbow were used as the objective function, to conduct the optimization design of elbow. Through optimized computation of pipelines with different specifications below DN1000mm, the optimal radius of curvature and pipe wall thickness were obtained for the horizontal elbow with two infinite long arms. The genetic algorithm can be used to carry out the optimum design for the elbow. The problems of misalignment restraint, can be dealt with by using the SUMT to reach stably consistemt convergence and fast convergence rate. The application of the genetic algorithm to the engineering design computation of elbow of the direct-burying heating pipeline can realize optimized design no matter whether the elbow size is restricted by location. The arrangement flexibility and reliability of the elbow can be improved and the project investment can be reduced.

关 键 词：遗传算法 惩罚函数法 直埋供热 弯管 优化设计 

分 类 号：TU995.3[建筑科学—供热、供燃气、通风及空调工程]