Research On Engineering Network Planning Optimization Based On Particle Swarm Optimization Algorithm

As the increasing of computer technolgoies,the development of intelligent algorithms provides many new ideas to promote intelligent construction.In order to improve the resource utilization,reduce the material and energy consumption,and accelerate the construction of the change of the pattern,this paper combined the resource optimization problem of network planning with practical engineering,and developed a resource optimization model under static and dynamic conditions to develop the evolutionary approaches based on adaptive particle swarm optimization to obtain the optimized solutions regarding resource leveling and shortest construction period.Combined with the cost of manual scheduling,the comprehensive optimization problem of static and dynamic network planning was proposed,and an improved multi-objective particle swarm optimization algorithm was designed to solve the problem,which also achieved good optimization effect,and provided an intelligent and optimized multi-objective construction scheme for the project.The main research contents are as follows:(1)In view of "resource-leveling with a fixed duration",an adaptive dissipative particle swarm optimization(ADPSO)algorithm was first designed.Compared with the generic particle swarm optimization(PSO)and the adaptive mutation particle swarm optimization(AMPSO),the advantages of ADPSO algorithm in terms of precision and stability were verified,and the intelligent optimization of resource equilibrium construction scheme was realized.Furthermore,in the face of the dynamic network planning where the time parameters of network planning change during the construction process due to "shutdown delays" and other reasons,the "dynamic resource-leveling with a fixed duration" problem is proposed.The network planning time parameters are updated in real time efficiently by performing the developed ADPSO algorithm iteratively.The obtained optimization results respectively are81.94%,76.37%,85.12% and 69.56%,and the intelligent optimization of the construction scheme of dynamic resource balance is achieved.(2)In view of the "resource-constrained and shortest construction time" problem,an adaptive heuristic particle swarm optimization(AHPSO)algorithm was designed.Compared with genetic and PSO algorithms,the advantages of the developed AHPSO algorithm in solving the problem of shortest construction period in terms of accuracy and stability were verified.In order to solve dynamic network planning problem,and solve the dynamic problem of "resource-constrained and shortest construction time",the developed AHPSO algorithm is iteratively performed and the network planning is updated in real time based on the the time parameters.The intelligent construction scheme with the shortest construction period and resource limitation after delay is obtained,and the intelligent optimization of the construction scheme with the shortest construction period under the dynamic condition is realized.(3)Based on the research of resource optimization problem,a multi-objective optimization problem of "resource-time-cost" under resource constraint is proposed in combination with the objective of labor scheduling cost optimization.The improved multi-objective particle swarm optimization(i MOPSO)algorithm is proposed based on the improved method of AHPSO algorithm.Compared with the experimental results of multi-objective particle swarm optimization,the advantages of i MOPSO algorithm in terms of distribution and convergence are verified and demonstrated.The cost of the optimal manual scheduling is about 1/5-1/4 of the original scheme,and the intelligent optimization of the multi-objective construction scheme is realized.Furthermore,a dynamic "resource-time-cost" problem model for dynamic network planning is established,and the i MOPSO algorithm is iteratively performed to solve the model in real time after updating the network planning time parameters in real time,so as to realize the intelligent optimization of the multi-objective construction scheme in dynamic situation.