基于树模型的严寒地区城市开放空间形态要素与气温关联性研究  被引量:1

Study on the Relationship between Urban Open Space Morphology and Air Temperature in Severe Cold Regions Based on Tree Model

在线阅读下载全文

作  者:金虹[1] 秦淦 JIN Hong;QIN Gan(School of Architecture,Harbin Institute of Technology,Key Laboratory of Cold Region Urban and Rural Human Settlement Environment Science and Technology,Ministry of Industry and Information Technology,Harbin 150006,Heilongjiang,China)

机构地区:[1]哈尔滨工业大学建筑学院,寒地城乡人居环境科学与技术工业和信息化部重点实验室,哈尔滨150006

出  处:《建筑科学》2023年第8期1-9,18,共10页Building Science

基  金:国家自然科学基金重点项目“严寒地区城市微气候调节原理与设计方法研究”(51438005)。

摘  要:城市微气候受城市形态影响显著,且二者关系具有高度复杂性。本文以严寒地区典型城市哈尔滨为例,借助机器学习非参数算法中树算法(决策树、随机森林及XGBoost),建立严寒地区城市开放空间形态要素与空气温度的关系模型,探究各形态要素与气温的关联性。研究发现,基于随机森林的模型性能最优,十折交叉验证R^(2)达到0.941 3;通过变量重要性分析发现,各形态要素中天空开阔度贡献值最大,铺地面积比最小;通过依赖关系分析发现,建筑基底面积比和距水体距离与气温的关系趋近于线性,其余形态要素与气温的关系呈明显的非线性,同时分析了各形态要素与气温的关联性。Urban microclimate is significantly affected by urban morphology,and the relationship between them is highly complex.Taking Harbin,a typical city in a severe cold region,as an example,and based on the tree algorithm in machine learning nonparametric algorithm(Decision Tree,Random Forest,and XGBoost),this paper established the relationship model between urban open space morphology and air temperature in severe cold areas,and explored the relationship between the morphology and air temperature.The study found that the model based on Random Forest has the best performance,and the 10-fold cross-validation R^(2) reaches 0.941 3.Variable Importance Analysis showed that the contribution value of the Sky View Factor is the greatest among the morphology,with Percentage of Pavement Area being the least.Partial Dependence Analysis revealed that the relationship between the Percentage of Building Area and Distance to Water and the air temperature tends to be linear,and the relationship between the other morphology and the air temperature is noticeably nonlinear.And the correlation between each morphology and the air temperature was also analyzed.

关 键 词:城市开放空间形态要素 严寒地区 机器学习 气温影响要素 关联性 

分 类 号:TU831[建筑科学—供热、供燃气、通风及空调工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象