| Groundwater is an important part of water resources in moutainous area, and it is the water supply of many areas. Traditional locating of groundwater relies on manual anaylysis, which is low efficent, besides, the lack of expert knowledge and exprince of management and designing department of groundwater of result in failure of water supply locating project.The distribution and buried features of groundwater are closely linked with rock types, geological structures, topography, hydrology and meteorology. Of all the above factors, the lithology and structure is the prerequisite for the existence of groundwater, which plays a controlling role in enrichment of groudwater. As a reflection of the geological structures, Geological map comprehensively demonstrates the relations of strata and geological structures.This paper researches into the basic elements and constraints of watre storage sturctures in bedrock areas, based on hydro geological theory, the paper abstracts the conceptual model of water storage, and describes the elements of its abrastract data types. Combined with the data organization of ArcGIS GeoDatabase, the paper finally constructed the physical model of water storage structure.Based on data model of water storage structure, the paper collects the expert knowledge and rules of water storage structure identification. Production rules are used to express these rules and knowledge, so that they can be managed by relational database. In this way, rules and knowledge are well linked to spatial data.Of various types of water storage structures, the paper classified them into water proof type, fault type, fold type and contact type with the causes and characteristics of the water storage. By analysing the spatial distribution and structual characteristics of each type of water storage, the paper developed the algorithm of each single type of water storage structure. the prototype system was developed by using of ojbect oriented method and ESRI spatial analysis cmoponents.Based on the above research, solutions of intelligent identification of water storage structures based on geological map are propoesde. The verification of prototype system used the 1:50000 geological map of Tangshan. |