基于动态多策略语义映射与规则推理的BIM与GIS三维建筑数据融合技术研究

近年来，随着新型智慧城市、智慧建筑的快速发展，BIM与GIS三维建筑数据的融合应用已成为学术界与产业界关注的热点之一。BIM与GIS三维建筑数据融合不仅要考虑几何信息融合，还应关注数据语义融合。然而，当前BIM与GIS融合大多基于数据格式转换与标准扩展的方法，缺乏对各类数据模型语义信息的提取、统一表达与匹配，不可避免造成语义信息丢失，难以实现BIM与GIS三维建筑数据的精准融合。为此，本课题拟将语义网技术引入城市建筑三维建模领域，首先，针对BIM与GIS三维建筑数据，基于IFC与CityGML构建统一规范化语义模型；其次，综合考虑BIM与GIS三维建筑语义模型中概念、属性、实例方面的特征，基于动态权值与改进匹配提取算法建立动态多策略语义映射模型，实现建筑元素概念层语义匹配；最后，构建基于规则推理的BIM与GIS三维建筑数据自动转换机制，实现BIM与GIS三维建筑数据语义层面的深度融合。

Recently, with the rapid development of new-type smart city and smart building, three dimensional building data fusion of Building Information Modeling (BIM) and Geographic Information System (GIS) has become one of research hotspots in academic and industrial domains. Not only the geometric data but also the semantic information should be considered in three dimensional building data fusion of BIM and GIS. However, the existing fusion methods mainly focus on data conversion and standard extension, lack of the extraction, formalization and matching of semantic information, making it difficult to achieve the accurate building data fusion of BIM and GIS. To this end, the Semantic Web technology will be used in the three dimensional modeling in this project. Firstly, a coherent structurized semantic model will be built based on Industry Foundation Classes (IFC) and CityGML. Then, considering the characteristics of all the factors including concept, attribute and instance, a dynamic multi-strategy semantic model will be constructed based on dynamic weight and improved extraction algorithm. Finally, an automatic conversion mechanism of BIM and GIS three dimensional building data will be established based on rule-based reasoning to realize the deep semantic integration of BIM and GIS three dimensional building data.