三维城市模型是智慧城市与数字孪生城市的核心内容.IFC(industry foundation classes)标准构建的建筑信息模型(building information model,BIM)为城市空间规划和建筑设计提供了丰富而精细的城市三维模型数据.由于标准的差异,BIM无法与三维地理信息系统无缝集成和交互应用,极大地限制了 IFC标准BIM的应用优势.本文挖掘IFC、CityGML(city geography markup language)标准三维建筑模型的语义表达差异,提出BIM源模型向不同细节层次(level of detail,LOD)层级模型的几何转换方法;构建一套完整的转换流程,实现IFC标准BIM向CityGML标准三维建筑模型的几何转换;设计并开发转换原型系统.研究表明:本方法及实现流程能够有效实现IFC-BIM模型向CityGML-BIM模型的不同LOD层级的转换,且对多数单体模型相当实用;转换后模型未见错误与偏差,视觉效果良好;具备灵活性,可根据不同模型的转换进行调整.研究成果为BIM与地理信息系统的集成应用提供了新思路,为智慧城市与数字孪生开辟了一种新的数据来源通道.
A method for conversion of standard 3D building models from IFC to CityGML
The three dimensional city model serves as the cornerstone of both smart cities and digital twin cities.The building information model(BIM),constructed in accordance with the industry foundation classes(IFC)standard,offers a wealth of detailed urban three-dimensional model data crucial for urban spatial planning and architectural design.However,discrepancies between standards hinder the seamless integration and interactive application of BIM data with the three dimensional geographic information system(3DGIS),thereby constraining the full potential of BIM data utilization and the spatial analysis capabilities of GIS within the IFC framework.This paper delves into the disparities in the semantic representation of 3D building models under the IFC and CityGML standards.It introduces a geometric transformation methodology from BIM source models to various level of detail(LOD)models,establishing a comprehensive transformation workflow that facilitates geometric conversion from BIM models conforming to IFC standards to 3D models adhering to CityGML specifications.Moreover,a prototype system for this transformation has been designed and developed.Using the conversion from an IFC model to a CityGML LOD-2 model as a case study,it illustrates how at this level,the building surface can be perceived as a simplified rendition of the IFC model.During the conversion,primary resources are extracted from the IfcWall and IfcSlab entities,while auxiliary resources stem from the IfcColumn,IfcBeam,and IfcCurtainWall classes to construct the building surface model.Consequently,the WallSurface object within the CityGML schema is generated by extracting external walls from the IfcWall class,followed by model simplification to eliminate unconvertible walls and edges.Additionally,geometric data from IfcCurtainWall elements can also be mapped onto the WallSurface class in CityGML.Surfaces within IfcWall entities are represented via the WallSurface class,whereas RoofSurface(representing roofs)necessitates the simplification of the geometric structure derived from IfcSlab entities linked to IfcRoof.This simplification process filters out non-convertible geometric information,ensuring the resulting RoofSurface adheres to specified criteria.The conversion approach and its implementation detailed herein effectively enable the transition across different LOD levels from IFC-BIM to CityGML-BIM models,proving highly practical for a majority of individual cases;the resultant models exhibited no errors or deviations,presenting satisfactory visual outcomes.Furthermore,the proposed conversion method boasts considerable adaptability,enabling tailored adjustments for diverse model transformations.It paves the way for a novel data fusion technique facilitating the integrated deployment of BIM and GIS,thereby inaugurating a fresh avenue for data sourcing in the realms of digital twins and smart cities.
NETL
NSTL
万方数据
