论文信息 - Adaptive Short Term Prediction of Power Load

Adaptive Short Term Prediction of Power Load

Abstract In this paper an adaptive k-step prediction algorithm is applied to shortterm prediction of the hourly load on a Power network. The load is Partitioned in a nominal and a residual Part. The residual load is modelled as an ARMA-process and different representations of the nominal load are considered. It is shown that the choise of nominal load representation as well as the adaptive mechanisms of the algorithm are crucial for the performance of the prediction method. The results from predictions with proposed algorithm compare favourably with other published prediction results.

Gudmundur R. Jonsson | J. Holst

[1] Francisco D Galiana. An application of system identification and state prediction to electric load modelling and forecasting. , 1971 .

[2] H. A. Dryar,et al. The Effect of Weather on the System Load , 1944, Transactions of the American Institute of Electrical Engineers.

[3] T. Bohlin. Four Cases of Identification of Changing Systems , 1976 .

[4] G. Box,et al. On a measure of lack of fit in time series models , 1978 .

[5] G. Irisarri,et al. On-Line Load Forecasting for Energy Control Center Application , 1982, IEEE Transactions on Power Apparatus and Systems.

[6] W. R. Christiaanse. Short-Term Load Forecasting Using General Exponential Smoothing , 1971 .

[7] Jan Holst. Adaptive Prediction and Recursive Estimation , 1977 .

[8] Lennart Ljung,et al. A theoretical analysis of recursive identification methods , 1978, Autom..

[9] P. Vähäkyla,et al. Short-term forecasting of grid load using Box-Jenkins techniques , 1980 .