Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning Export

@article{citeulike:2808880, abstract = {The evaluation of surface water resources is a necessary input to solving water management problems. Neural network models have been trained to predict monthly runoff for the Tirso basin, located in Sardinia (Italy) at the S. Chiara section. Monthly time series data were available for 69 years and are characterized by non-stationarity and seasonal irregularity, which is typical of a Mediterranean weather regime. This paper investigates the effects of data preprocessing on model performance using continuous and discrete wavelet transforms and data partitioning. The results showed that networks trained with pre-processed data performed better than networks trained on undecomposed, noisy raw signals. In particular, the best results were obtained using the data partitioning technique.}, author = {Cannas, Barbara and Fanni, Alessandra and See, Linda and Sias, Giuliana}, booktitle = {Time Series Analysis in Hydrology}, citeulike-article-id = {2808880}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.pce.2006.03.020}, citeulike-linkout-1 = {http://www.sciencedirect.com/science/article/B6X1W-4M645B2-2/1/02a5b9c6a193d30e67e2c632aad04548}, doi = {10.1016/j.pce.2006.03.020}, journal = {Physics and Chemistry of the Earth, Parts A/B/C}, keywords = {ai, ann, wavelet, wnn}, number = {18}, pages = {1164--1171}, posted-at = {2008-05-18 12:58:04}, priority = {2}, title = {Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning}, url = {http://dx.doi.org/10.1016/j.pce.2006.03.020}, volume = {31}, year = {2006} }

TY - JOUR ID - citeulike:2808880 L3 - citeulike-article-id:2808880 N2 - The evaluation of surface water resources is a necessary input to solving water management problems. Neural network models have been trained to predict monthly runoff for the Tirso basin, located in Sardinia (Italy) at the S. Chiara section. Monthly time series data were available for 69 years and are characterized by non-stationarity and seasonal irregularity, which is typical of a Mediterranean weather regime. This paper investigates the effects of data preprocessing on model performance using continuous and discrete wavelet transforms and data partitioning. The results showed that networks trained with pre-processed data performed better than networks trained on undecomposed, noisy raw signals. In particular, the best results were obtained using the data partitioning technique. IS - 18 JF - Physics and Chemistry of the Earth, Parts A/B/C EP - 1171 TI - Data preprocessing for river flow forecasting using neural networks: Wavelet transforms and data partitioning VL - 31 T2 - Time Series Analysis in Hydrology SP - 1164 KW - ai KW - ann KW - wavelet KW - wnn AU - Cannas, Barbara AU - Fanni, Alessandra AU - See, Linda AU - Sias, Giuliana PY - 2006/// UR - http://dx.doi.org/10.1016/j.pce.2006.03.020 ER -