Wavelet Based Artificial Intelligence Approaches for Prediction of Hydrological Time Series

Autor:	Vahid Nourani, Gholamreza Andalib
Rok vydání:	2015
Předmět:	Support vector machine Engineering Wavelet Artificial neural network Series (mathematics) business.industry Wavelet transform Artificial intelligence business
Zdroj:	Lecture Notes in Computer Science ISBN: 9783319148021 ACALCI
DOI:	10.1007/978-3-319-14803-8_33
Popis:	In this paper, the efficiency of a Wavelet-based Least Square Support Vector Machine (WLSSVM) model was examined for prediction of monthly Suspended Sediment Load (SSL) of the Aji-Chay River. First the SSL was predicted via ad hoc Least Square Support Vector Machine (LSSVM) and Artificial Neural Network (ANN) models. Thereafter in hybrid models, streamflow and SSL time series were decomposed into sub-signals via a wavelet transform and the decomposed subseries were fed into LSSVMs and ANNs to simulate a discharge-SSL relationship. The results showed that ANNs led to better outcomes with Determination Coefficient (DC)=0.62 than ad hoc LSSVMs with DC=0.59. On the other hand, WLSSVMs performed better than wavelet-based ANN (WANN) models in monthly SSL prediction and wavelet data pre-processing could lead to more accurate results.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::6c6e79c87c5cfd13b8d70031c91e884b https://doi.org/10.1007/978-3-319-14803-8_33 Zobrazit plný text záznamu