Machine Learning for Layout Optimization of Braces of Building Frames

来源 :Asian Congress of Structural and Multidisciplinary Optimizat | 被引量 : 0次 | 上传用户:aa6512048
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  Machine learning has been extensively applied in various fields of engineering including image processing,automatic control,etc.Binary decision tree (BDT),support vector machine (SVM),artificial neural network (ANN),and association rule are the basic tools of machine learning.However,only a few studies can be found for application of machine learning to structural optimization.In 1990s,ANN was extensively used mainly for estimating the structural responses in the process of structural optimization.For example,nonlinear seismic responses of a 3D-frame were estimated;the maximum stress ratio of a plate composed of functionally graded material was estimated;and a basic framework of structural optimization based on estimation of constraint functions was proposed.Biedermann and Grierson[l]trained member grouping in the process of optimization of plane frames.However,most of the applications were for estimation of function values rather than classification of solutions that can be a candidate of approximate optimal solution.Furthermore,most of applications of ANN are in the second generation,when application to only small toy problems were possible.
其他文献
In some cases,samples are inevitably contaminated by noise.In order to provide satisfactory approximations for stochastic (noisy) systems,some regression-type metamodels are adopted in current researc
会议
In aerodynamic optimization,the number of design variables needs to be increased to improve the effect and accuracy of optimization.However,surrogate optimization is extremely sensitive to the number
We present the cone-shaped heterogeneous double split hollow sphere(CsHDSHS) which are based on a helmholtz resonator whose resonance frequency is observed[1,2].CsHDSHS are made by paper cone attached
会议
会议
Robot arms have been widely utilized in various manufacturing and automation applications,such as aerospace,automobile,energy,machinery,semiconductors,etc.Especially in the development of Industry 4.0
会议
In the reliability analysis of practical engineering structures,performance functions are generally complex and implicit with high nonlinearity,which makes the traditional methods of reliability analy
The hypervolume indicator has been proved as an outstanding metric for the distribution of Pareto points,and the derived hypervolume based expected improvement (HVEI) has received a particular attenti
会议
Passive free-layer damping is an effective means in vibration control,especially in the vibration suppression of automotive and commercial airplane.Damping performance of the passive layer damping str
会议
Steel truss structures for point-supported glass curtain walls with a web member configuration are presented using topology optimization based on a truss-like material model.The initial structure comp
会议
Kriging model is widely used in structural and multidisciplinary optimization.A few algorithms can deal with the expensive objective function and constraint functions,but the computational efficiency