Research On The Generation Design Of Primary And Secondary Education Building Based On Multi-agent System

In the field of educational architecture design,campus planning and layout,building shape and space generation are one of the most important research focus in the design process.In the early stage of design,a large number of reasonable possible schemes are enumerated according to the requirements of the task book and a variety of design specifications.With the development of artificial intelligence technology,the research methods of Genetic Algorithm,Multi-Agent System,Cellular Automata,Machine Learning and other disciplines have been introduced into the research of architecture discipline.With these methods,architects can greatly improve their work efficiency and explore various new complex building forms under the limited rules.As a new architectural design method,architectural generative design represented by the above research methods has gradually become an important branch of research on the exploration of architectural form and space generation in recent years.The Multi-Agent System is a computing system composed of multiple agents interacting in one environment,and there are a lot of cross studies between this theory and architecture.The work of this paper is mainly divided into two parts.The first part of this paper is a review of the relevant research in the field of building generation design,especially the application of multi-agent system in the field of building planning.According to the different agents of agents,it is divided into three categories and makes a detailed review of the research in this field.By summarizing and combing the previous research on multi-agent system in the field of generation design,this paper will guide the direction of the future research.The latter part of this paper is a multi-agent system generation and design experiment,which is divided into two parts: the experiment of campus general layout and the experiment of education building functional space layout,and the concept of natural lighting of education building is introduced in the experiment of space layout.It includes the design of algorithm rules,the generation of experimental steps,the generation of results analysis and other links,as well as the whole workflow from the campus general layout to the layout of building functional space into a real project for verification,and completed a conceptual design of a multi-agent system oriented by natural lighting as a generation tool.The experimental part of this study takes the layout of primary and secondary education buildings as the research object,based on the self-organized behavior mode of the multi-agent system,using the multi-agent agent building function module,based on pycharm and rhino+ The grasshopper platform,through Python Programming and parametric modeling,initially constructs the layout tools of primary and secondary education buildings that meet the requirements of educational building spacing,functional space layout rules,site boundary conditions constraints,group gathering,etc.,as well as the functional space layout of primary and secondary education buildings that meet the requirements of functional quantity constraints,evacuation distance constraints,natural lighting constraints,etc Tools.The core of this experiment is to translate the program language of traditional architecture vocabulary,and to establish the operation rules of multi-agent system according to the design characteristics of primary and secondary schools.