Font Size: a A A

Application Of Algorithmic Generating Form Based Force In Architectural Design

Posted on:2021-01-03Degree:MasterType:Thesis
Country:ChinaCandidate:J ZhaoFull Text:PDF
GTID:2392330611983909Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
With the help of computer technology,building form design based on mechanical principle is a new design method developed in recent years.In the process of architectural design,the stress characteristics of the structure are analyzed,and physical simulation is carried out through computer programs,which are continuously calculated,iterated,and optimized to find the relationship between spatial form and structural rationality.Although there are many current mechanical form finding methods,designers know little about them and have less application in practical engineering.In view of this,this article takes the interactive mechanical form finding method as the main line,and combs the four methods of form-finding of structures.Taking actual engineering as an example,these methods are used for structural optimization analysis and form design.The finite element analysis of the form results was performed to verify its reliability and superiority.The details are as follows:(1)Introduce the principle of graphic statics,force-to-form analysis process and genetic algorithm.With the help of Structure FIT,use the actual truss as an example to carry out diverse design and optimize the form.(2)The thrust network analysis method is introduced,Rhino Vault is used as an analysis tool,and the lactticed shell structure of the funnel-shaped light well is used as an example to optimize the lactticed shell form.(3)Combining the finite element analysis method and simulated annealing algorithm,based on the basic principle of generating the compression-only shell shape based on the reverse physical suspension model,the finite element analysis tool Karamba and the single-objective optimization tool Galapagos were used to optimize the shell form.(4)Using the bidirectional evolutionary structural optimization and taking Ameba as the software platform,the shell obtained in the previous chapter is furthertopology optimized.After the above research,the following conclusions were reached:(1)By comparison with the nodal method,the maximum error of the internal force solution of the graphic statics is 0.38 %,and the method is accurate and reliable.Using genetic algorithm for optimization design,six kinds of truss optimization results were obtained,all of which meet the requirements of the design code.Compared with the original structural design,the amount of steel used was reduced by up to 44 %.The superiority of this method were verified.(2)Using thrust network analysis method,five new lactticed shell structural forms were obtained.Among them,the maximum displacement was only 1.49 mm and the maximum stress was only 1.37 Mpa.The forces were reasonable and met the requirements of the specification.The reliability of the thrust network analysis method was verified.(3)Using the finite element method,the shell structure simulated by the reverse suspension model has a small tensile region and a small tensile stress,and the shell is basically under full pressure,which proves the correctness of the simulation results.After optimization,the average strain energy and displacement of the shell are reduced.Compared with the initial shell,the average strain energy is reduced by60.4%,the deformation is reduced by 85.8 %,and the area under pressure is significantly increased.The characteristics of the only pressure shell are more obvious.Verifying the rationality and practicability of this form optimization method.(4)Based on the bidirectional evolutionary structural optimization,the shell obtained in the previous section is topologically optimized.The shell shape obtained is novel,the material consumption is significantly reduced,and the maximum displacement is 164 mm,which meets the requirements of the specification.
Keywords/Search Tags:mechanical form finding, interactivity, graphic statics, thrust network analysis, structural optimization
PDF Full Text Request
Related items