作者: Fahimeh. Sarafraz , Hassan. Ghasemi , Hassan. Monsef
DOI: 10.1109/EEEIC.2011.5874828
关键词:
摘要: Price forecasting in competitive electricity markets plays a crucial role for any decision making. This is difficult task since price time series are non-stationary, and with variable mean variance, also have periodic monthly seasonal behavior. paper introduces an approach to forecast several-hours-ahead locational marginal (LMP) using locally linear neuro-fuzzy (LLNF) model the PJM market. The autocorrelation method used make appropriate input vectors. LLNF leads more accurate results compared Multi Layer Perceptron (MLP) neural network.