Application of adaptive learning in generalized neuron model for short term load forecasting under error gradient functions. (English) Zbl 1213.68486
Ranka, Sanjay (ed.) et al., Contemporary computing. Third international conference, IC3 2010, Noida, India, August 9–11, 2010. Proceedings, Part I. Berlin: Springer (ISBN 978-3-642-14833-0/pbk; 978-3-642-14834-7/ebook). Communications in Computer and Information Science 94, 508-517 (2010).
Summary: Artificial Neural Networks (ANN’s) have huge difficulties such as large training time, large number of nodes, hidden nodes can cause training difficulties, more training patterns, more complexity of model, least flexibility. Generalized Neuron Model (GNM) has less training time, no hidden layer, and more flexibility, less complexity. In this paper non adaptive learning, adaptive learning in GNM for short term load forecasting (STLF) is trained and tested for different error gradient functions.
For the entire collection see [Zbl 1201.68009].
For the entire collection see [Zbl 1201.68009].
MSC:
| 68T05 | Learning and adaptive systems in artificial intelligence |
