见物见人——时空大数据支持下的存量规划方法论

City Sensing: An Inventory Planning Tool Based on Spatial-temporal Big Data

段冰若
北京清华同衡规划设计研究院有限公司技术创新 中心 规划师,硕士

摘要: 相比于传统的增量规划,存量规划中主要在产权本质、时间逻辑和空间处理尺度上有着本质的不同。因此,存量规划对现有用地 现状和性质的精准刻画提出了更高的要求。作为当前传统的用地现状分析图在存量规划中存在着地块特征刻画精度有限与用 地分类维度过低等不足。对用地类型的混合、同种用地类型的规模、同一地块的时间属性等用地特征,传统的现状分析图也难以 进行描述。随着互联网LBS(Location-Based Service)服务的发展,越来越多LBS时空数据因其巨大的用户基数和完善的时空地理 信息,受到规划师的关注。这些新的时空地理数据使得对用地功能和人口活动特征的详细刻画成为可能。使用互联网某LBS平台 人口分时活动密度数据,叠加百度POI(Place of Interest),通过非监督分类和非负矩阵分解的方法,分别对北京市六环内的地块尺度、500 m网格尺度和25 m点阵尺度进行用地功能的识别与分类。通过多维度分类结果的叠加,对研究区域的用地功能、人口时空活动特征进行深入刻画,探讨通过大数据进一步辅助存量规划的用地功能研究方法。

Abstract: Compared to the traditional incremental planning in China, inventory planning is different in the perspective of property, time and space. Thus, a higher demand for the depiction of existing space is needed in the inventory planning process. However, the depiction method used in incremental planning such as land use analysis map cannot fulfill this demand. With the prevalence of internet LBS (Location-Based Service)data, planners see a new opportunity to make a more detailed depiction of the existing space. This paper intends to use an LBS data of population density by hour, together with POI (Place of Interest) from Baidu. With the help of unsupervised learning algorithm, a detailed depiction of land use and population activity pattern will be presented, showing more opportunities for big data analysis in the current urban planning research.

关键词:存量规划、机器学习、用地分类、 LBS数据

Keyword: Inventory planning,Machine learning,Land use clustering,LBS data

中图分类号:中图分类号TU981

文献标识码: 文献标识码A

[1]邹兵. 增量规划,存量规划与政策规划[J]. 城市规划,2013(2):35-37.
ZOU Bing. Increment planning, inventory planningand policy planning[J]. Urban Planning, 2013(2):
35-37.
[2]Halder A, Ghosh A, Ghosh S. Supervised andunsupervised land-use map generation from remotely sensed images using ant based systems[J].
Applied Soft Computing Journal, 2011, 11(8):5770-5781.
[3]梁松. 城市规划动态监管卫星遥感关键技术研究[D]. 北京:中国矿业大学博士学位论文,2010.
LIANG Song. Study on the key technology of satellite remote sensing for urban planning dynamic supervising[D]. Beijing: The Dessertation
for Doctoral Degree of China University of Mining and Technology, 2010.
[4]王芳,高晓路,许泽宁. 基于街区尺度的城市商业区识别与分类及其空间分布格局——以北京为例[J]. 地理研究,2015,34(6):1125-1134.
WANG Fang, GAO Xiaolu, XU Zening.Identification and classification of urban commer c i a l districts at Block Scale[J] .Geographical Research, 2015,34(6): 1125-1134.
[5]王德,王灿,谢栋灿,等. 基于手机信令数据的上海市不同等级商业中心商圈的比较——以南京东路、五角场、鞍山路为例[J]. 城市规划学刊,2015(3):50-60.
WANG De, WANG Can, XIE Dongcan, et al.Comparison of retail trade areas of retail centerswith different hierarchical levels: a case study ofEast Nanjing Road, Wujiaochang, Anshan Road in Shanghai[J]. Urban Planning Forum, 2015(3): 50-60.
[6]张晓亮,陈智宏,刘冬梅,等. 一种基于多源数据的出租车分布预测方法研究[J]. 道路交通与安全,2015(1):47-51.
ZHANG Xiaoliang, CHEN Zhihong, LIUDongmei, et al. A taxi travel forecasting methodbased on multi-source data[J]. Road Traffic andSafety, 2015(1): 47-51.
[7]龙瀛,张宇,崔承印. 利用公交刷卡数据分析北京职住关系和通勤出行[J]. 地理学报,2012,67(10):1339-1352.
LONG Ying, ZHANG Yu, CUI Chengyin.Identifying commuting pattern of Beijing using bus smart card data[J]. Acta Geographica Sinica, 2012, 67(10): 1339-1352.
[8]裘炜毅,刘杰. 手机大数据视角下的工业区职住平衡分析方法[C]//中国城市规划年会论文集:城市规划新技术应用. 北京:中国建筑工业出版社,2015.
QIU Weiyi, LIU Jie. Industrial job-housing balance in the perspective of mobile phone big data[C]// China Annual Urban Planning Forum Proceedings.Beijing: China Architecture & Building Press, 2015.
[9]龙瀛,沈尧. 数据增强设计——新数据环境下的规划设计回应与改变[J]. 上海城市规划,2015(2):81-87.
LONG Ying, SHEN Yao. Data augmented design:urban planning and design in the new data environment[J]. Shanghai Urban Planning Review, 2015(2): 81-87.

微信扫一扫
关注“上海城市规划”
公众号