APLIKASI ARTIFICIAL NEURAL NETWORK UNTUK PERAMALAN ALIRAN SUNGAI BLEGA
| Main Authors: | Manyuk, Fauzi, Minarni , Nur Trilita |
|---|---|
| Format: | Article PeerReviewed application/pdf |
| Terbitan: |
Faculty of Civil Engineering and Planning
, 2005
|
| Subjects: | |
| Online Access: |
http://eprints.upnjatim.ac.id/2522/1/MINARNI_NUR_TRILITA.pdf http://eprints.upnjatim.ac.id/2522/ |
Daftar Isi:
- A multivariate non-linear model for synthetic generation (forecasting) of daily flow (discharge) series is presented in this dtudy. The time series is built upon a multilayer feedfoward neural network with an added multivariate random component normally distributed. The usual error backpropagation algorithm is used to train the network, with a sequaential training scheme using shuffled patterns for a stochastic search in the weight space. Using neural network model, study case of Blega River-Madura island. The results by calculating show that the solution based on backpropagation algorithms are consistent with those based multivariate regression model. They also indicate that backpropagation model in this paper is reasonable and feasible.