摘要
针对翻板钢水闸门的结构优化问题,选择国内某翻板钢水闸门为例,运用ANSYS软件构建有限元模型,然后融合径向基神经网络和遗传算法组成混合优化算法,以在控制响应计算精度和优化部件方案的前提下,大幅提高优化方案的计算效率。采用MATLAB软件实现混合优化算法,运行算法后,对比原方案和优化后方案,发现工字型钢、槽钢的型号以及梁的数量与配置完全不变,但角钢从不等边75 mm×50 mm×10 mm,换为等边45 mm×45 mm×5 mm型号,面板厚度从10 mm减少到5 mm,优化后方案最大应力从142 MPa增大到162 MPa,最大位移从4.66增大为4.92 mm,但都远小于规范规定的容许值,同时优化后水闸门总耗钢量为0.1051 m 3,相比于原设计方案的0.1567 m 3节约了32.93%的钢材,优化后的方案经济效益显著。
Aiming at the structural optimization of flap steel gate,a domestic flap steel gate is selected as an example.The finite element model is constructed by using ANSYS software,and then the hybrid optimization algorithm is composed of radial basis function neural network and genetic algorithm,so as to greatly improve the calculation efficiency of the optimization scheme on the premise of controlling the calculation accuracy of response and optimizing the component scheme.MATLAB software is used to realize the hybrid optimization algorithm.After running the algorithm and comparing the original scheme with the optimized scheme,it is found that the models of I-section steel and channel steel and the number and configuration of beams are completely unchanged,but the angle steel model is changed from 75 mm×50 mm×10 mm to 45 mm×45 mm×5 mm;the panel thickness is reduced from 10mm to 5mm;the maximum stress of the optimized scheme is increased from 142 MPa to 162 MPa;the maximum displacement is increased from 4.66 mm to 4.92mm,but they are far less than the allowable value specified in the specification.At the same time,the total steel consumption of the optimized sluice gate is 0.1051 m 3,which saves 32.93%steel compared with 0.1567 m 3 of the original design scheme,and the economic benefit of the optimized scheme is remarkable.
作者
贺弘扬
HE Hongyang(Yangling Vocational and Technical College,Xi'an 712100,Shaanxi China)
出处
《粘接》
CAS
2022年第12期133-136,共4页
Adhesion
关键词
径向基神经网络
遗传算法
翻板钢水闸门
有限元
radial basis function neural network
genetic algorithm
flap steel gate
finite element
