Object-oriented electric distribution system data modeling in a GIS context

In response to the new competition resulting from the deregulation in the electric power industry, utilities are looking for better technology which can improve productivity, streamline operations, provide accurate and up-to-date system information, and enable timely and flexible data access. One of the strategic movement by the utilities is to define an integrated information system which can provide a common database support for multi-disciplinary distribution functions. Such a system is called Distribution Management System (DMS) in this dissertation. The approach taken in this research work is to develop a DMS using Geographic Information Systems (GIS), an emerging technology in 1990s. Currently there are many limitations to the traditional GIS data models which hinder the development of engineering applications in a DMS environment. The objectives of this dissertation are to propose an object-oriented GIS data model and to design a DMS database based on the new GIS data model.; The GIS Domain Space Model proposed in this dissertation allows developers to describe the real-world in terms of objects instead of geometric primitives. Multiple data abstraction hierarchies defined in different domains for electric power distribution systems provide the flexibility of data representation at different graphical/map scale and at different analytical scale. Data abstraction also enhances the extendibility of the data model. Through data encapsulation, a virtual graphic data layer is developed to provide the independence of specific graphic systems. The dissertation also proposed a concept of Nerve Center, a rule-based system to model many business rules inside the DMS database. The purpose of this nerve center is to enhance data integrity and to eliminate the need for the external applications to enforce the rules.; To validate the design concept, a prototype system consisting of an Object-Relational DBMS (POSTGRES) and a commercial GIS system (ARC/INFO) is implemented. The test results of the prototype system prove that the proposed data model for electric power distribution systems is comprehensive, flexible, and extensible. It can serve as an integrated data environment for multi-disciplinary purposes including real-time applications.