jailcd2023002: A review of architectural literature on "number intelligence"

Abstract

Buildings are the primary places of human activity, and future high-performance buildings should provide spatial quality to their users while being constructed and operated as efficiently as possible using resources. In recent years, on the premise of the rapid development of deep learning, "intelligent number" buildings also appear. In this review, we will discuss the two aspects of building design generation and building energy consumption prediction and investigate and study the automatic layout generation method and building energy consumption prediction method based on envelope structure design. The application of hybrid method goes beyond the limitation of single science and integrates and expands the research field of "digital intelligence" architecture. To promote the development of "digital intelligence" architecture. Using algorithmic and data-driven solutions, they can optimize design solutions during the design process, or explore creative solutions for new buildings. In addition, they can better understand the existing building stock or changes in building policy: empowering architects, urban designers, legislators and the public to make more informed decisions to create sustainable cities of the future.