大连理工大学学报
大連理工大學學報
대련리공대학학보
JOURNAL OF DALIAN UNIVERSITY OF TECHNOLOGY
2015年
1期
39-46
,共8页
投资-效益准则%全寿命造价%多目标优化模型%精英保留非劣排序遗传算法
投資-效益準則%全壽命造價%多目標優化模型%精英保留非劣排序遺傳算法
투자-효익준칙%전수명조개%다목표우화모형%정영보류비렬배서유전산법
cost-benefit criterion%total life-cycle cost%multi-obj ective optimization model%elitist non-dominated sorting genetic algorithm
投资-效益准则是基于结构性能抗震设计的重要原则,它所追求的设计目标是在结构的初始造价与地震损失期望之间达到一种和谐的优化平衡,使结构在全寿命周期内总费用最小。在以上的单目标优化模型基础上,提出了考虑结构初始造价、损伤期望、检查和维护费用、拆除费用及残余价值和环境污染费用的多目标全寿命优化设计模型。基于所提出的多目标优化设计模型,应用精英保留非劣排序遗传算法,建立了以截面尺寸、纵筋和箍筋的配筋率为决策变量的近海桥梁结构全寿命抗震性能多目标优化设计模型,并给出了优化设计流程及具体实现。结果表明:得到的非劣解在目标空间分布均匀,算法收敛性和鲁棒性较好。
投資-效益準則是基于結構性能抗震設計的重要原則,它所追求的設計目標是在結構的初始造價與地震損失期望之間達到一種和諧的優化平衡,使結構在全壽命週期內總費用最小。在以上的單目標優化模型基礎上,提齣瞭攷慮結構初始造價、損傷期望、檢查和維護費用、拆除費用及殘餘價值和環境汙染費用的多目標全壽命優化設計模型。基于所提齣的多目標優化設計模型,應用精英保留非劣排序遺傳算法,建立瞭以截麵呎吋、縱觔和箍觔的配觔率為決策變量的近海橋樑結構全壽命抗震性能多目標優化設計模型,併給齣瞭優化設計流程及具體實現。結果錶明:得到的非劣解在目標空間分佈均勻,算法收斂性和魯棒性較好。
투자-효익준칙시기우결구성능항진설계적중요원칙,타소추구적설계목표시재결구적초시조개여지진손실기망지간체도일충화해적우화평형,사결구재전수명주기내총비용최소。재이상적단목표우화모형기출상,제출료고필결구초시조개、손상기망、검사화유호비용、탁제비용급잔여개치화배경오염비용적다목표전수명우화설계모형。기우소제출적다목표우화설계모형,응용정영보류비렬배서유전산법,건립료이절면척촌、종근화고근적배근솔위결책변량적근해교량결구전수명항진성능다목표우화설계모형,병급출료우화설계류정급구체실현。결과표명:득도적비렬해재목표공간분포균균,산법수렴성화로봉성교호。
Cost-benefit criterion is significant in performance-based seismic design.The obj ective of this criterion can make a reasonable and optimum balance between the initial cost of structures and potential loss expectation of structures,so as to minimize the total cost of structures in the whole life cycle.The multi-obj ective optimum design method based on single-obj ective optimum design method is firstly proposed considering initial cost, cost expectation, maintenance and inspection costs, dismantling fee,residual value and cost for potential environmental pollution.And then,elitist non-dominated sorting genetic algorithm is used to establish life-cycle performance-based seismic design method for bridge structures by taking section size,reinforcement ratio of longitudinal reinforcement and stirrup as decision variables.The optimization design process and the concrete implementation are given in detail.The experimental results show the obtained non-inferior solution has uniform spatial distribution and the target algorithm has good convergence and robustness.