Open AccessStudy of the section optimization of gravity dam based on improved PSO
Author(s) -
Shiqiang Zheng,
Kai Yang,
Lingzhi Huang
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/330/2/022007
Subject(s) - particle swarm optimization , inertia , mathematical optimization , convergence (economics) , mathematics , rate of convergence , control theory (sociology) , computer science , algorithm , physics , artificial intelligence , computer network , channel (broadcasting) , control (management) , classical mechanics , economics , economic growth
As the linearly-decreasing inertia weight of particle swarm optimization (PSO) can easily lead to local extremum, this paper proposes an improved PSO which changes the dynamic changing pattern of Inertia Weight w based on the properties of trigonometric function. The inertia weight value of improved PSO is large at the early stage and relatively small at the later stage, thus improving the global searching ability as well as the convergence performance of the algorithm. The Optimal Section Area A(X) of non-overflow gravity dam calculated through improved PSO is 5147.3 m2 which is 6.93% lower than that calculated through SPSO (A(X)=5530.7m2). To obtain the optimal solution, the calculation steps needed by improved PSO and SPSO respectively are 15 and 22. The convergence rate of improved PSO is 31.8% higher than that of SPSO. The calculation results of the two algorithms show that the improved PSO is able to obtain better optimization results in higher convergence rate and thus is suitable for the optimization design of large-scale hydraulic structures.