Research On Multi-objective Optimization Of Subway Project Based On Multi-population Ant Colony-Particle Swarm Algorithm

Project duration,cost and quality are three major management goals that a project must pay attention to and determine.There are vague and non-linear interface characteristics between them.Therefore,when control the construction period,quality,and cost targets of subway projects,a multi-objective optimization management concept should be formed.Based on the current research of the scholars,this paper proposes a multi-group ant colony-particle swarm fusion algorithm to solve the multi-objective optimization problem in the construction management of metro engineering projects.Specifically,this article focuses on the following aspects:First,analyze the research status of intelligent algorithms in multi-objective optimization research projects in domestic and foreign.Propose a way which be used to improve the algorithm.The first stpe is to design a multi-population ant colony algorithm.First,three ant sub-populations of duration,cost,and quality were established in order,let various groups of ants update the path to be searched according to the probability transfer formula,and update the pheromone according to the specified pheromone mechanism within the population,and finally complete the search output optimal solution set.In the second step,according to the number of optimization targets,the particle population is divided into three groups of particle swarms and one main particle group to perform parallel search from the particle swarm and the main particle swarm.;The solution sets of multiple colony ant colony algorithms are used as the initial solution of each particle in the particle swarm,to improve the particle search capability.Speed up the iteration and shorten the search time to achieve the efficiency of the fusion algorithm.Secondly,analyze the definition of the project duration,cost and quality objectives respectively,and establish the corresponding function model based on the construction process.The critical path method is used to construct the duration function model.When construct the cost function model,the cost is divided into direct cost and indirect cost.When establish the quality function model,the expert scoring method is used to score the quality of different construction methods in each process and proportion of this type of construction organization in the entire project is also calculated use the expert scoring method too.Based on the score data,the entropy weight method is used to determine the weight of each construction process.Effectively compatible with the advantages of the subjective and objective empowerment method.On the basis of the objective function model,the multi-objective comprehensive optimization function model is established based on the target threshold defined by the owner.Finally,a real metro project is used in combination as an example to validate the algorithm and the comprehensive optimization model,write algorithm code in MATLAB R2014 a software environment is used to solve the multi-objective optimization problem of metro project.At the same time contrast with the results obtained from separate multiple ant colony algorithm and multi-objective particle swarm algorithm.The results show that the multi-population ant colony-particle swarm fusion algorithm has more Pareto solution sets than the the latter two algorithms,that is,the alternative construction organization scheme is more and the convergence speed is faster.The research in this paper provides new ideas for multi-objective optimization for metro construction projects.The relevant theory of quality quantification has been enriched,and under the premise of ensuring quality pursuing the aim of realizing short construction period and maximizing profits,the construction unit has been provided with a variety of reasonable and scientific construction schemes,and the construction unit received more effective encouragement.