| The spatial grid structure is widely used in large-scale public buildings at home and abroad because of its structural efficiency and spatial flexibility.Meanwhile,the rapid development of digitization has greatly enriched the expression forms of buildings,and various long-span and complex surface shapes emerge one after another,which also brings great challenges to structural design,especially in the collaborative design of beautiful architectural schemes and structural schemes.Consequently,the traditional grid structure design and selection methods have been difficult to meet its requirements,that is,the current spatial grid structure has many problems,such as difficult grid generation,low efficiency of structural design,strong experience of selection and so on.How to realize the optimization and rapid analysis of grid scheme according to architectural surface is of great significance.Based on this,taking the spatial grid structure as the research object,a preliminary study on the intelligent generation and selection optimization of spatial grid structure is made by using parametric technology and intelligent optimization algorithm.The main results are as follows:Firstly,the parametric grid generation of spatial structure grid is studied.Based on the theory of non-uniform rational B-splines(NURBS)in rhino,two grid generation methods,parameter domain bisection method and isoparametric line segmentation method,are introduced.The former directly discretizes the parameter domain of the surface to obtain the grid nodes,which has high calculation efficiency and flexible and controllable grid form.The latter performs isoparametric line segmentation based on the isoparametric line of the surface to ensure the equal length and high uniformity of the grid.In order to make the application of the two methods more specific,the free-form surface is divided into triangle,quadrilateral and circular boundary surfaces according to the boundary shape.Moreover,the concept of generalized topological parameters is proposed.Compared with the traditional topological optimization method,the grid generated by this method is more constructable and easier to apply.The control parameters of different surfaces are determined,and a rich single-layer and double-layer grid library is established,as a result,the rapid generation of grid under given surface and grid topology control parameters is realized.In addition,the automatic design method of spatial grid structure is studied.The limitations of the current mainstream structural parametric design methods and tools are analyzed.In view of the long design cycle of spatial grid structures,the C# programming language is used to develop the data interface between Grasshopper and SAP2000,including section definition,element definition,and load definition,model assembly and analysis,automatic design,post-processing of calculation results 6 major functional modules,with a total of 40 components,covering the whole process of spatial grid structure design.What’s more,the algorithm is developed to solve the key problems such as the generation of loading virtual surface,the local axis adjustment of members,the designation of calculation length and slenderness ratio,and the iterative optimization of section.And the efficiency of the program is verified by engineering examples,which provides technical conditions for the automation,digital design and real-time evaluation of scheme performance of large and complex spatial grid structures.Last but not least,the optimization method of spatial grid structure selection is studied.In view of the limitations of the traditional phased design method,an integrated selection optimization method considering both grid topology and member section design is proposed,and its mathematical optimization model is constructed.This method considers the combination of multiple working conditions according to the specification,uses the minimum steel quantity to determine the optimal grid topology on the premise of ensuring structural safety,and adopts the intelligent optimization algorithm based on elite strategy genetic algorithm to realize the selection method.Numerical examples show that this method has high accuracy and efficiency,and the steel consumption of the optimized grid topology scheme is significantly reduced,which fully explains the rationality of this method,and there is no need for manual intervention in the whole process,so as to realize the intelligent generation of grid.The above research results are encapsulated as integrated components in parametric design platform Rhino + Grasshopper using C# language,which has high engineering application value and provides innovative ideas and practical tools for related research. |
