Forecasting Electricity Demand on Short, Medium and Long Time Scales Using Neural Networks

This paper examines the application of artificial neural networks (ANNs) to the modelling and forecasting of electricity demand experienced by an electricity supplier. The data used in the application examples relates to the national electricity demand in the Republic of Ireland, generously supplied by the Electricity Supply Board (ESB). The paper focusses on three different time scales of interest to power boards: yearly (up to fifteen years in advance), weekly (up to three years in advance) and hourly (up to 24 h ahead). Electricity demand exhibits considerably different characteristics on these different time scales, both in terms of the underlying autoregressive processes and the causal inputs appropriate to each time scale. Where possible, the ANN-based models draw on the applications experience gained with linear modelling techniques and in one particular case, manual forecasting methods.

[1]  Masahito Yamada,et al.  Structural Time Series Models and the Kalman Filter , 1989 .

[2]  Kishan G. Mehrotra,et al.  Forecasting the behavior of multivariate time series using neural networks , 1992, Neural Networks.

[3]  Stefen Hui,et al.  Application of feedforward neural networks to dynamical system identification and control , 1993, IEEE Trans. Control. Syst. Technol..

[4]  John Ringwood,et al.  Linguistic Modelling of Short-Timescale Electricity ConsumptionUsing Fuzzy Modelling Techniques , 1995 .

[5]  Peter C. Young,et al.  Recursive estimation and forecasting of non‐stationary time series , 1990 .

[6]  Andreas S. Weigend,et al.  Time Series Prediction: Forecasting the Future and Understanding the Past , 1994 .

[7]  Ipek Guinel.  Forecasting system energy demand , 1987 .

[8]  John V. Ringwood,et al.  Integration of multi-time-scale models in time series forecasting , 2000, Int. J. Syst. Sci..

[9]  George E. P. Box,et al.  Time Series Analysis: Forecasting and Control , 1977 .

[10]  R.J. Marks,et al.  Data partitioning for training a layered perceptron to forecast electric load , 1993, [1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.

[11]  C. Chatfield,et al.  Comparative Models for Electrical Load Forecasting. , 1986 .

[12]  Saleh M. Al-Alawi,et al.  Principles of electricity demand forecasting. I. Methodologies , 1996 .

[13]  Robert J. Marks,et al.  Electric load forecasting using an artificial neural network , 1991 .

[14]  W. Pedrycz,et al.  Artificial neural network for forecasting daily loads of a Canadian electric utility , 1993, [1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.

[15]  Azzam-ul Asar,et al.  A specification of neural network applications in the load forecasting problem , 1994, IEEE Trans. Control. Syst. Technol..

[16]  John V. Ringwood,et al.  Improvement of electricity consumption forecasts using temperature inputs , 1994, Simul. Pract. Theory.

[17]  Forecasting commercial electricity sales , 1987 .

[18]  Georg Dorffner,et al.  Neural Networks for Time Series Processing , 1996 .

[19]  Kwang Y. Lee,et al.  Short-term load forecasting using diagonal recurrent neural network , 1993, [1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.

[20]  Larry D. Huugh.  Time Series Forecasting: Unified Concepts and Computer Implementation , 1989 .

[21]  Burce L Bowerman,et al.  Time series forecasting: unified concepts and computer implementation , 1986 .

[22]  Ted Jaditz.  Time series prediction: Forecasting the future and understanding the past : Andreas S. Weigend and Neil A. Gershenfeld, eds., (Reading, MA: Addison-Wesley Publishing Co., 1949) pp. xvii + 643, $29.95 , 1995 .

[23]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1971 .

[24]  Andrew Harvey,et al.  A unified view of statistical forecasting procedures , 1984 .

[25]  Adnan H. Al-Anbuky,et al.  Power demand prediction using fuzzy logic , 1995 .

[26]  P. Thomson,et al.  Forecasting load‐duration curves , 1994 .

[27]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .