Font Size: a A A

Windowed Evolutionary Structural Optimization And Research On Its Application In Double-deck Concrete Box Girder

Posted on:2019-05-13Degree:MasterType:Thesis
Country:ChinaCandidate:L J WangFull Text:PDF
GTID:2392330596994766Subject:Engineering
Abstract/Summary:PDF Full Text Request
In structural design,engineers expect to achieve the best performing structure with the least amount of material,the lowest cost,and the simplest process.As a mathematical method to optimize the material distribution in the initial region under a given load,constraints and performance indicators,structural topology optimization has received extensive attention from designers.The progressive structure optimization algorithm has been widely developed with its advantages of clear concept,simple rules and convenient calculation.However,due to its low computational efficiency,unstable algorithm and serious checkerboard format in the structural optimization calculations with a large number of finite element units,this paper has carried out several researches on the problems in the progressive structure optimization algorithm.(1)Evolutionary structural optimization(ESO)is usually used to find the optimal topology shape of the structure and guide the design of the structure.nevertheless,there are some problems exist such as its defects on elimination criteria which cause low efficiency.Windowed evolutionary structural optimization(WESO)is proposed in this paper based on a modified elimination criteria of ESO.It considers structural average strain energy as the elimination criteria for the optimization and its elimination rate is set as a self-adaption state by an adjustable window.(2)Analysis by example shows,WESO,the difficulties in traditional ESO including low optimization efficiency and optimization results of distortion have been solved to some extent,and its application scope to obvious expansion is enabled since the high-order finite element analysis and optimization become feasible.The reliability of the topology of WESO has also been proved by comparing with the Michell theory.(3)The tie-and-strut model is one of the effective tools for studying the stress state of complex reinforced concrete structural materials.In order to study the stress mechanism of webs under the bending conditions of double-layer traffic concrete box girder,this paper uses a topology optimization method with strong optimization ability and high computational efficiency,which is called " Windowed Evolutionary Structural Optimization(WESO)" to construct the tie-and-strut model.The experimental phenomenon is compared with the tie rod-compression rod model,and the accuracy of the method is verified.At the same time,it is recommended to design a similar component with a round hole with a diameter not greater than h/3 at the distance between the two ends of the support,and the center should be inclined to the tension zone.The trapezoidal hole is the most reasonable.The tie-and-strut model of the I-beams under different boundary conditions is used to provide reference for the design of similar components.(4)Based on the WESO method,the optimal design of the webs under the bending conditions of the double-layer traffic concrete box girder is presented,which provides the design ideas for the following research.According to the topology optimization solution,the model of single box three-room double-deck concrete box girder under bending condition is established,Its web opening ratio is 1.5 times that of the non-optimized design and only one beam is required in the span,at the same time,the stress value range of the optimized model is smaller than the test model and the mid-span displacement is 0.5 times of the test model,the comprehensive comparison optimization model has superior mechanical properties,low engineering cost,and better ventilation and lighting effects.
Keywords/Search Tags:structural optimization, evolutionary structural optimization method, windowed, Double-deck traffic, Reinforced concrete box girders, The tie-and-strut model
PDF Full Text Request
Related items