Wind-Resistant Optimization Of Large Span Roofs Based LCQPSO Algorithm

The large-span roof structures are wind-sensitive,which are characterized by light mass,small damping,small stiffness and low natural vibration frequency.However,there are few studies on wind resistance optimization of large-span roof structures,and it will bring some safety risks if the wind resistance optimization of high-rise structures are completely referred to.As new optimization algorithms,intelligent algorithms have been widely used in civil engineering,especially in structural optimization design.Therefore,this paper will adopt one of the most widely used intelligent algorithms-the Particle Swarm Optimization(PSO)algorithm,its improved algorithm-the Quantum-behaved PSO(QPSO) algorithm and Logistic chaotic QPSO(LCQPSO) algorithm to the wind resistance of long-span roof structure Optimization design,and analyze the influence of the parameters of algorithms on the Optimization.The main contents of this paper are as follows:(1)Harmonic Excitation Method(HEM) is used to deal with data from wind tunnel test to obtain the wind-induced vibration of long-span roof structure.Then the equivalent static wind load at the most adverse wind angle is calculated by Load Response Correlation(LRC).A mathematical model for wind resistance optimization of large-span roof is established to minimize the cross-sectional area of the structure under the action of equivalent wind load and the constraint of displacement and stress.Using the fitness function of the sum of particle distance value and penalty function to select the optimal design scheme in the optimization process.(2)Since the QPSO algorithm is easy to fall into local maximum,chaos mapping is introduced into the QPSO algorithm.Chaos map is used to replace the random number which is in particles updated formula in QPSO algorithm with a certain mutation probability.Chaotic local search technique is used to update the global optimal solution of each iteration.(3)This paper compiles wind resistance optimization program based on PSO,QPSO and LCQPSO algorithms.The effects of algorithm parameters on long-span wind resistance optimization are analyzed.The optimal parameters of the three algorithms collectively are set as: the maximum number of iterations is 400,and the population value is 10.Other optimal parameters optimized based on PSO algorithm are set as: the value of maximum speed limit[0.0076,0.0141],the value of inertia weight is 0.5,and the value of learning factors are 2.Other optimal parameters optimized based on LCQPSO algorithm are set as: mutation probability value is 0.3,the initial value of chaos values is a random value between 0 and1,and that chaos method chooses Logistic chaos mapping.(4)Wind-resistant optimization based on PSO,QPSO and LCQPSO algorithms can achieve convergence before 400 generation,and the convergence of solutions are less than the total weight of the initial structure satisfying the constraint conditions,show the algorithms in the applicability of the wind-resistant optimization of large-span structures.By analyzing the optimal solution of the three algorithms,the minimum total weight obtained from LCQPSO algorithm is 320.27 t,which is reduced by 55.67% compared with initial structure.The cross-sectional areas of the bars are less than the initial value by QPSO and LCQPSO algorithm.The cross-sectional area of the 11 th member is larger than that before optimization by PSO algorithm,which indicates that the PSO falls into local maximum in the optimization process.During the optimization process,the displacement constraint redundancy is always less than zero,while the stress constraint redundancy exceeds the constraint limit many times in the iteration,which indicates that the stress constraint has a greater effect on design variables.