结构工程
人口
预制混凝土
遗传算法
计算机科学
算法
工程类
数学
数学优化
人口学
社会学
作者
José V. Martí,Víctor Yepes,Fernando González‐Vidosa,A. Luz
标识
DOI:10.1016/j.rimni.2013.04.010
摘要
This paper deals with the minimum cost automatic design of precast bridge decks made of U-beams and an upper slab. It uses a hybrid memetic algorithm that combines the population search of solutions by genetic algorithms and a search by variable neighborhood. This algorithm is applied to a bridge made of two isostatic U-beams of 20-40 m of span and a width of 12 m. This example has 40 discrete variables. The evaluation module takes into account the service and ultimate limit states usually considered for these structures, i.e. flexure, shear, torsion, cracking, deflections, etc. The use of the memetic algorithm requires its previous calibration. Each of the heuristics is run 12 times, obtaining information about the minimum and average values, as well as the scatter. The parametric study showed a good correlation for the cost, the number of strands and the steel and concrete quantities with the span length. Savings have been found between 8 and 50% compared to other structures really executed. The presented procedure allows the practical application to the real design and its adaptation to the precast process.
科研通智能强力驱动
Strongly Powered by AbleSci AI