论文信息 - Data-centric demand forecasting for utilities

Data-centric demand forecasting for utilities

Abstract Growing amounts of data being archived and maintained in process historians calls for ever more intensive use of data-mining technologies allowing to explore the data, extract useful knowledge, and turn it into business advantage. Utility producers and utility distributors can take significant benefits from application of database-intensive techniques. This paper describes how exploratory analysis of historical datasets can help utility companies to identify typical demand patterns, and how these can be further used for forecasting of future consumption, and consequently for making decisions on optimal operation of the utility plant. The technical concept as well as all steps of the forecasting process are described along with summarization of the experience gained from development and practical implementation of a proprietary forecasting tool.

Petr Stluka | Karel Mařík | Zdeněk Beran

[1] Tony Perchard,et al. Short Term Gas Demand Forecasting , 2000 .

[2] Magdy M. A. Salama,et al. Application of the decomposition technique for forecasting the load of a large electric power network , 1996 .

[3] D.W. Bunn,et al. Forecasting loads and prices in competitive power markets , 2000, Proceedings of the IEEE.

[4] Nima Amjady,et al. Short-term hourly load forecasting using time-series modeling with peak load estimation capability , 2001 .

[5] W. Cleveland. Robust Locally Weighted Regression and Smoothing Scatterplots , 1979 .

[6] John McEnroe,et al. You Cannot Be Serious , 2002 .

[7] John Piggott. Accurate Load Forecasting - "You Cannot Be Serious" , 2003 .

[8] A. Doucet,et al. Sequential MCMC for Bayesian model selection , 1999, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics. SPW-HOS '99.

[9] J. W. Taylor,et al. Short-term electricity demand forecasting using double seasonal exponential smoothing , 2003, J. Oper. Res. Soc..

[10] John V. Ringwood,et al. Forecasting Electricity Demand on Short, Medium and Long Time Scales Using Neural Networks , 2001, J. Intell. Robotic Syst..