Title :
Next day peak load forecasting using a multilayer neural network with an additional learning
Author :
Morioka, Y. ; Sakurai, K. ; Yokoyama, A. ; Sekine, Y.
Author_Institution :
Kansai Electric Power Co., Hyogo, Japan
Abstract :
A multilayer neural network with additional learning is applied to next day peak load forecasting. First, the performance of the neural network is studied by using time series data of a sinusoidal curve added on top of an increasing time function. The authors discuss what kind of additional learning method is effective when new time series data are obtained every day. Based on the above results, simulations of the next day peak load forecasting by the neural network are conducted using actual load data.
Keywords :
backpropagation; feedforward neural nets; learning (artificial intelligence); load forecasting; power engineering computing; additional learning; backpropagation; learning method; multilayer neural network; next day peak load forecasting; sinusoidal curve; time series data; Casting; Economic forecasting; Learning systems; Load forecasting; Multi-layer neural network; Neural networks; Power generation economics; Predictive models; Root mean square; Weather forecasting;
Conference_Titel :
Neural Networks to Power Systems, 1993. ANNPS '93., Proceedings of the Second International Forum on Applications of
Conference_Location :
Yokohama, Japan
Print_ISBN :
0-7803-1217-1
DOI :
10.1109/ANN.1993.264349
