Research On Data Matching Of City Underground Pipeline Data Based On Spatial Similarity

City Underground Pipelines(CUP) are the pipelines and ancillary facilities for underground water, sewerage, gas, fuel, heat, electricity, communications, lighting, radio and television, traffic signals, industrial materials, public video surveillance and so on. They are the important channels of urban material flow, energy flow and information flow, are the important urban infrastructure, and the important life lines in cities. In recent years, with the progress of the construction of the "Digital City", the CUP geoinformation increasingly becomes the core content of the urban space information resources and the CUP geoinformation system (CUP-GIS) also gradually becomes support applications for the urban planning, construction and management. Because of the different purposes of management mode and application of CUP, there exist two types of CUP geoinformation system (CUP-GIS), integrated pipelines GIS for the integrated pipelines management and urban planning, and professional pipelines GIS for the professional pipelines management and the pipeline Operation and maintenance in the pipelines authority sector. Influenced by the GIS platform, technology condition and the purpose of the application, two types of pipeline form two different data resources with different semantics, spatial data models and spatial data accuracy, although their application aims at the same pipeline object in the same area.Generally, the integrated pipeline data has higher location accuracy, but less rich semantic information than the professional pipeline data. Generally, the integrated pipeline data has higher location accuracy, but less rich semantic information than the professional pipeline data. Referencing to the spatial similarity and spatial data matching theories and technologies, this paper studied data matching between professional pipelines and integrated pipelines data for sharing and comprehensive utilization of the CUP spatial data.The research contents and results in this paper are as follows:Firstly, this paper analyzed the causes of differences between professional pipelines and integrated pipelines spatial data. The cause includes subjective ones including detection error, pipeline changes, and objective ones including different data requirements and application purpose. For the above reasons, there generates differences between two types of pipelines, mainly in data source, data accuracy, data model and semantic.Secondly, on the base of common data matching type and on the account of the geometry type, data feature and data matching target in the pipeline data, this paper proposed the matching type of pipeline data referencing to the type of geometry element and its relationship between numbers; this paper discussed the layout feature of urban pipelines network, and hierarchical divided pipeline data following the main road pipeline, the block tree pipeline and independent pipeline according to the above analysis result.Thirdly, this paper proposed spatial similarity measurement methods for pipeline data. This method measured data’s semantic similarity, geometry similarity and topological relationship similarity. This paper used semantic similarity measurement methods of ontology to measure concept similarity between integrated pipeline and professional pipeline under different Categories expression and this paper also used features which included the spatial position, orientation, length, correlation degree to determine the quantitative description way of geometry similarity and topological relationship similarity; Based on geometry similarity and hierarchical expression of pipeline data, this paper designed the pipeline data matching strategy which matched data from whole to the exact one and adjust the similarity index weights on account of the feature of the pipeline data.This paper took gas professional and integrated pipeline spatial data in Nanjing experimental zone for example to build professional and integrated pipeline ontology for semantic similarity measure and designed a process for pipeline data matching. Through the development of a pipeline data matching prototype system, this paper validated and analyzed the proposed matching method.