摘要: |
深入认知建成环境对街道活力的影响效应,是街道
环境优化与城市更新的基础。但多数研究忽略了二者之间的非
线性关系,难以有效指导设计实践。基于多源大数据测度街道
活力与建成环境,使用机器学习算法来分析其非线性效应,并
针对不同类型街道进行探索。结果表明:1)提升开发强度是促
进街道活力的最有效措施;2)建成环境要素对街道活力的影响
表现出非线性特征,将其控制在合理范围内,街道活力才会有
效提升;3)街道环境要素的综合设置应考虑其交互效应,一个
要素的影响会随着另一个要素的变化被放大或缩小;4)老城
区、商务片区、工业与区域交通设施周边区域及景观性街道的
活力形成机制存在显著差异。相关规律可为街道的精细化设计
提供人本尺度的理论参考。 |
关键词: 风景园林 城市更新 精细化设计 街道活力 建
成环境 非线性效应 机器学习 |
DOI: |
|
基金项目: |
|
Nonlinear Effect of Built Environment on StreetVitality: A Multi-source Big Data Analysis Based onXGBoost Model |
WU Wanshu,MA Ziying,GUO Jinhan,ZHAO Kai |
Abstract: |
Understanding the influence of built environment
on street vitality is the basis of street environment optimization
and urban renewal. However, most studies ignore the nonlinear
relationship between them, which makes it difficult to effectively
guide the design practice. Based on multi-source big data to
measure street vitality and built environment, machine learning
algorithm is used to analyze the nonlinear effect, and different
types of streets are explored. The results show that: 1) Enhancing
development intensity is the most effective measures to promote
street vitality; 2) The influence of environmental factors on the
street vitality shows nonlinear characteristics. Only by controlling
them within reasonable ranges can the street vitality be effectively
improved; 3) The comprehensive setting of street environment
elements should consider their interaction effects, and the influence
of one element will be enlarged or reduced with the change of
another element; 4) There are significant differences in the vitality
formation mechanisms among the streets in the old urban area,
the business district, the surrounding areas of industrial lands and
regional transportation facilities and the landscape streets. Relevant
laws can provide a theoretical reference of people-oriented scale for
the refined design of streets. |
Key words: landscape architecture urban renewal refined design street vitality built environment nonlinear effect machine learning |