Title :
ANN-based risk assessment for short-term load forecasting
Author :
Mori, H. ; Iwashita, D.
Author_Institution :
Dept. of Electr. & Electron. Eng., Meiji Univ., Kawasaki
Abstract :
A new risk assessment method is proposed for short-term load forecasting. The proposed method makes use of the RBFN (radial basis function network) to forecast loads due to the good performance. Sufficient realistic pseudo-scenarios are required to carry out quantitative risk analysis in the Monte-Carlo simulation. The multivariate normal distribution with the correlation between the input variables of RBFN is used to give more realistic results. In addition, this paper employs the moment matching method to improve the accuracy of the multivariate normal distribution. The peak over threshold (POT) approach is used to evaluate the risk that exceeds the upper bound of generation capacity. The proposed method is successfully applied to real data of one-step ahead daily maximum load forecasting
Keywords :
Monte Carlo methods; load forecasting; neural nets; normal distribution; power generation control; radial basis function networks; risk management; Monte-Carlo simulation; artificial neural network; extreme value theory; load forecasting; moment matching method; multivariate normal distribution; peak over threshold; radial basis function network; risk analysis; risk assessment; Artificial neural networks; Bayesian methods; Input variables; Load forecasting; Power markets; Power system modeling; Predictive models; Radial basis function networks; Risk management; Uncertainty;
Conference_Titel :
Intelligent Systems Application to Power Systems, 2005. Proceedings of the 13th International Conference on
Conference_Location :
Arlington, VA
Print_ISBN :
1-59975-174-7
DOI :
10.1109/ISAP.2005.1599305
