Theses and Dissertations

Jian Wu

Issuing Body

Mississippi State University

Advisor

Schulz, Noel N.

Committee Member

Hansen, Eric

Committee Member

Ginn III, Herbert L.

Date of Degree

5-8-2004

Original embargo terms

MSU Only Indefinitely

Document Type

Graduate Thesis - Campus Access Only

Major

Electrical Engineering

Degree Name

Master of Science

College

James Worth Bagley College of Engineering

Department

Department of Electrical and Computer Engineering

Abstract

With the improvements in computer technology, users in utilities expect to receive more advanced functions of their power system management applications by using the data distributed among computer applications. The conventional method of point-to-point interface is not efficient for building large-scale computer systems and is especially difficult for system integration. Integration efforts are carried on to facilitate software applications communicating with others. Defining the same description for a central database and exchangeable data format among applications is the first step for integration. Since current data models in the Common Information Model (CIM) are designed mostly for analyzing the terrestrial transmission power system, they are not sufficient for Shipboard Power Systems (SPS). In order to facilitate software integration in SPS, a fundamentally common semantic for SPS applications needs to be extended from the current CIM. Therefore, these analysis applications for SPS could communicate with each other based on a standard model. In this thesis, a pulsed load model is extended from CIM. This model is a general time dependent pulsed load and the simulation of pulsed railgun load validates the proposed data model. Also, stress upon the power grid caused by the pulsed load is analyzed and the continuous railgun operation is simulated. In addition, new CIM data models are built for passive and active filters to facilitate system simulation and further application. As an active filter is a device that incorporates complicated control strategies, much work focused on finding a general data model to accommodate most common active filters. Finally the simulation for active filter validates the proposed data model.

URI

https://hdl.handle.net/11668/17802

Download

Share

COinS