基于差分粒子群算法的装配式住宅项目进度优化研究  

Hybrid particle swarm algorithm based on differential evolution for the research of prefabricated housing project schedule optimization

作　　者：赵平[1] 吴昊[1] 李萍萍[1] 乔媛媛[1] 

机构地区：[1]西安建筑科技大学土木工程学院,陕西西安710055

出　　处：《西安建筑科技大学学报（自然科学版）》2016年第2期178-182,共5页Journal of Xi'an University of Architecture & Technology(Natural Science Edition)

基　　金：陕西省自然科学基础研究基金(2014JM2-5046);国家自然科学基金青年项目(51308441);陕西省科技统筹创新工程计划项目(2015KTCQ03-18;2015KTZDSF03-05-03)

摘　　要：为了有效的解决装配式住宅项目的进度优化问题,通过探讨装配式住宅项目调度问题的约束条件,提出了基于差分进化的改进粒子群算法(DEPSO),建立了以项目工期最优为目标的进度优化模型.通过在差分算法和粒子群算法之间建立信息互融机制,克服了差分算法和粒子群算法单独使用易产生局部最优和精度低的缺陷,以达到最优工期的预期目标.通过实例分析,进行了三种算法的比较,证明了DEPSO算法在求解装配式住宅项目进度优化中高效、合理、鲁棒性强,对装配式住宅的推广和发展起到一定的积极作用.In order to solve the prefabricated housing project schedule optimization problem effectively,through discussing the constraints of prefabricated housing project scheduling problem,based on the particle swarm optimization(DEPSO) of differential evolution,the schedule optimization model is built whose objective is the optimal fabricated project period.The new algorithm established an information exchange mechanism between DE and PSO which greatly overcomes the defects of local optimum and low accuracy when DE or PSO is used alone to achieve the schedule optimization target.Ending with analyzing the case,comparing the three algorithms,the result proves that DEPSO is reasonable and efficient in solving assembled project schedule optimization with a strong robustness which plays a positive role in promotion and development of the prefabricated housing project.

关 键 词：装配式住宅 差分算法 粒子群算法 进度优化 

分 类 号：TU722[建筑科学—建筑技术科学]