权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Optimization of the architectural form using the life evolution process

利用生命进化过程优化建筑形式

基本信息

批准号：
10650603
负责人：
AOKI Yoshitsugu
金额：
$ 1.47万
依托单位：
Tokyo Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
1998
资助国家：
日本
起止时间：
1998 至 2001
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-10650603/
关键词：
life evolution architectural form optimization genetic algorithm

项目摘要

It has be made to be a first purpose in developing the technique which efficiently obtains architecture form which maximizes the value of given evaluation function, on the basis of the genetic algorithm devised by Holland, and discovering the new architecture form using such technique. In addition, and the searching the new design principle from common features in architectural forms evaluated excellent has be also made to be a purpose.This study has largely been composed of three following parts.1 : Improvement genetic algorithm of the architectural form optimization methodIt may become a new genetic code in which the phenotype does not make a sense, and the calculation efficiency becomes bad in order to remove it. Then, gene expression method for following features of the architecture form was proposed. In addition, in order to reduce too big calculation load on evaluation of architectural form, the method that it learns the concise evaluation function by neural network in the process of evolution was developed, because the calculation load was big.2 : Optimization of architecture and city form and finding new design principlesThe above method is applied on optimization problems on 1) the room allocation, 2) the middle corridor type plan, 3) the form of apartment house under the sunshine condition form of multiple dwelling house, 4) the facilities allocation in the city. In addition, features of being with this optimum solution in common were extracted, and design know-how and design principle of architecture and city were obtained.3 : The new design principle of architecture and citySome of design principles obtained above suggest a design theory for the new architecture and city which are not in the traditional design.

在Holland设计的遗传算法的基础上，有效地获得给定评价函数值最大的建筑形式，并利用该技术发现新的建筑形式，是发展该技术的首要目的。此外，从被评价为优秀的建筑形式的共同特征中寻找新的设计原则也是一种目的。本研究主要由以下三个部分组成。1：改进建筑形式优化方法的遗传算法它可能成为一个新的遗传密码，其中表型没有意义，为了去除它而计算效率变差。然后，提出了建筑形态以下特征的基因表达方法。此外，为了减少建筑形式评价的计算量过大，由于计算量大，开发了在进化过程中利用神经网络学习简洁评价函数的方法。2 .建筑与城市形态的优化，寻找新的设计原则。将上述方法应用于1)房间配置优化问题，2)中间走廊式平面优化问题，3)阳光条件下的公寓住宅形式优化问题，4)城市设施配置优化问题。此外，还提取了与该最优解共同存在的特征，得到了建筑与城市的设计诀窍和设计原则。3：建筑与城市的新设计原则上述的一些设计原则为传统设计中没有的新建筑与新城市提供了一种设计理论。