城市公共空间三维路径系统寻径 方法与应用 —以淮安井神盐化工业区更新改造为例.

宋哲昊; 唐 芃; 宋 伟; 龙 灏

Urban public spaces are an important element of the urban environment. With the increasing volume of business, landscape, and leisure activities, the spatial form and traffic flow of urban public spaces are becoming increasingly complicated. As such, it is essential to design an accessible, convenient, and safe path system for urban public spaces. Exploring the three- dimensional path in urban public spaces is a multi-objective optimization which is difficult for artificial design methods to solve optimally. Many researchers have tried to introduce pathfinding algorithms to explore solutions. However, existing studies on path exploration only discuss the shortest path on a two- dimensional plane and don't provide a complete workflow in series connection with subsequent scheme design. In this study, a three-dimensional path system was explored using the case study of the reconstruction of the Jingshen Industrial Zone in Huan'an into a large urban industrial heritage park. The roaming paths at the site were generated based on pathfinding results provided by the algorithm. Based on the Java language, a pathfinding algorithm model of a three-dimensional path system in the urban public space oriented to multi-element influences was developed using the Dijkstra algorithm as the prototype. This model was connected with digital modeling tools to support the scheme design. A complete "digital chain" analysis-design combined workflow was formed. Firstly, the site conditions for path planning and design in the project were extracted. Influencing factors of path planning and design were analyzed and then translated into program language. A three-dimensional lattices pathfinding network was constructed. The "moving cost" of pathfinding points in the network reflects whether the three-dimensional path can pass through a point or the comprehensive moving cost of influencing factors when passing through a point. On this basis, the algorithm model was further optimized based on the codes of building design to produce three-dimensional paths that can meet spatial evaluation requirements and design codes. Finally, the spatial data of pathfinding results was input into Grasshopper modeling software and the details of the path system were generated automatically, forming a complete "digital chain" workflow from site analysis to path system exploration and to scheme deepening. Based on the above process, this study completed preliminary planning design of a three-dimensional cruising path system composed of ground driving lanes and landscape overbridges in the reconstruction project for Jingshen Industrial Zone in Huan'an. Designers created deep plans on this basis and completed the final scheme. In urban design, the quantification and calculation bases of path design are relatively clear, and the design objectives are supported by appropriate algorithms. Path exploration was completed using an appropriate programming algorithm. In this study, the algorithm tools corresponding to three-dimensional path exploration were developed to make path planning and design in urban public spaces independent from subjective judgment. In view of the design results, research methods in this study reached the expected objective of three- dimensional path exploration in the urban public space. Since the algorithm operation is based on a specific pathfinding network, the path exploration results are not completely optimal and still have many problems before artificial adjustment, such as pixilation and multiple turns. This study proved the feasibility and effectiveness of digital technology throughout the whole process, from analysis to design. This approach can be applied to other complicated problems in urban design projects and form a connection between data-driven design and analysis.

JAVA (Indonesia); PUBLIC spaces; URBAN planning; PATH analysis (Statistics); URBAN policy; TRAFFIC flow; DIGITAL technology

South Architecture / Nanfang Jianzhu, 2022, Issue 9, p19

1000-0232

Academic Journal

10.3969/j.issn.1000-0232.2022.09.003