Font Size: a A A

Scheme Selection And Optimum Design Of Single-Layer Reticulated Domes

Posted on:2006-09-06Degree:MasterType:Thesis
Country:ChinaCandidate:Z F LiuFull Text:PDF
GTID:2132360155472852Subject:Engineering Mechanics
Abstract/Summary:PDF Full Text Request
Latticed dome is a type of space structural system which is applicable and promising. At present, single-layer latticed domes have been applied widely to large-span building. the structural type of single-layer latticed domes is various. if single-layer latticed domes is designed in traditional method, the workload is excessive. At the same time, it is difficult to compare different scheme because of time and cost. Thus, final plan is usually not the best one. Although some of structural design software have the part of optimum design in China, many adopt optimum methods of full-stress. They are fit to the structure whose constraint condition is less and more simple and aren't suitable for the structure as single-layer latticed domes whose every component is spacial beam element and whose bearing capacity depend not only on the member bar's strength but also on resistance to overturning and local stability of single-layer latticed domes. When single-layer latticed dome is designed, the use of constraint conditions, such as strength, stiffness, stability and physical dimension, is necessary. Moreover, every member bar's section of single-layer latticed domes must match applicable standard and model number. Namely optimum design of single-layer latticed domes belong to structural optimum design of discrete variable. The Genetic Algorithm which has been developed recently is the optimum Algorithm based on natural genetic mechanism of biologic evolutionism, which includes regeneration, crossover and mutation etc. In search of the best solution, Genetic Algorithm encodes the variable and forms binary encoding. So the Algorithm is suitable for optimum design not only of continuous variable but also of discrete variable, even non-number. In searching optimum solution, it use only objective function, but not use the other ancillary information and differential coefficient information. So Genetic Algorithm has no specific requirements for objective function, design variable and feasible zone. It can solve kinds of optimization question, Specially fit to complex and nonlinear question which can't be solved in traditional methods. It have good overall optimization performance and can converge upon optimum solution. The dissertation solve the member bar's section optimization problem of single-layer reticulated domes with adaptive genetic algorithm, and write the program of adaptive genetic algorithm and finite element in VC++ language according to the introduced algorithm, which can realize automatic optimum design of single-layer reticulated domes. Theory and practice has proved that optimum design methods can get good result, but the application don't reach the level that the subject have reached. One of the cause is that the method will reduce structural safety excessively. The ratio of buckling load and designing load, which can evaluate bearing and safety performance, will be got though the buckling analysis of the single-layer reticulated domes which are optimum designed according to applicable procedures and the single-layer reticulated domes which are designed according to applicable procedures but not be optimized. Thus reliability of optimum design and performance of kinds of single-layer reticulated domes can be assessed. Through the optimum design in this paper, the following conclusions are obtained . â‘ The optimum result of the single-layer reticulated domes of K6-7 and K8-7, which is designed by the optimum design program of Latticed domes based on adaptive genetic algorithm and finite element theory, indicated that the optimum design program have the character of simple theory, easy convergence, good and reliable result. 90% of optimum solution can be attained easily, but it is very difficult to get real optimum solution especially to single-layer reticulated domes which is statically indeterminate and multiple constraint structure, as adaptive genetic algorithm is good at global searching and unable to get to local solution. In fact, it is also difficult to get to the optimum solution of single-layer reticulated domes in other optimum methods. It is enough to search a better optimum solution rapidly. â‘¡Results of nonlinear analysis indicate that buckling load of the single-layer reticulated domes, which are optimum designed according to applicable procedures, reduces more, compared with buckling load of the single-layer reticulated domes that designed according to applicable procedures but not be optimized. namely the former's safety stock also reduces more than the latter's. On the other hand, the buckling load of the single-layer reticulated domes with optimum design is 5 time more than that of the single-layer reticulated domes without optimum design, as is specified in the technical regulation of Latticed domes. So optimum design according to applicable procedures can meet safety requirement. â‘¢The ratio of buckling load and designing load will change as the ratio of rise to span of single-layer reticulated domes that have the same structural style and span. bigger ratio of rise to span , bigger ratio of buckling load and designing load. So The single-layer reticulated domes with large ratio of rise to span has higher safety stockthan that with small ratio of rise to span. When ratio of rise to span is less than 1/7, ratio of the buckling load and designing load will approach the limited value of applicable standards. So when smaller single-layer reticulated domes will be designed, constraint condition should be improved to increase safety stock.
Keywords/Search Tags:single-layer reticulated domes, buckle analysis, structural optimum design, genetic algorithm
PDF Full Text Request
Related items