Date of Award

Fall 1995

Document Type

Thesis - Restricted

Degree Name

Master of Science (MS)

Department

Civil, Construction, and Environmental Engineering

First Advisor

Brown, Ronald H.

Second Advisor

Belfore, Lee

Third Advisor

Ruchti, Timothy L.

Abstract

The purpose of this thesis is to develop artificial neural network based models to predict daily gas consumption for a portion of metropolitan Milwaukee, Wisconsin. In this thesis, a particular type of artificial neural networks referred to as diagonally recurrent artificial neural networks are used to develop gas load models. These new models are then compared to feed-forward artificial neural network based models and other traditional forecasting techniques. All the models are constructed using historical load and weather data provided by the Wisconsin Gas company. The weather data was obtained from a weather recording site located between 60th street and Capitol Drive in Milwaukee, Wisconsin. Some of the inputs into the models are weather forecasts. Actual weather forecasts are used for these inputs when testing the models. These weather forecasts were also provided by the Wisconsin Gas Company.

Share

COinS

Restricted Access Item

Having trouble?