Development of artificial neural network models to predict daily gas consumption

The development of feedforward artificial neural network (ANN) models to predict daily gas consumption is the subject of this paper. A methodology based on network sensitivities and intuition is discussed. The methodology is applied to two regions in Wisconsin served by the Wisconsin Gas Company (WGC). Training results show that ANN models reduce prediction root mean squared errors by more than half when compared with linear regression models. The ANN predictions are compared with predictions made by WGC gas controllers for the first 97 days of the 1994-1995 heating season. The ANN prediction errors are 82.2% and 69.7% of the WGC estimate errors for the two regions.