Abstract | ||
---|---|---|
Short-term stream flow forecasts are required for simulation, optimization, and decision-making purposes in applications ranging from hydropower planning to flood prevention. The particular case of one-day ahead stream flow forecasting is an important but difficult problem that has been increasingly studied using hybrid computational intelligence and machine learning techniques. However, these studies present several limitations. In this work we attempt to address those limitations by (1) replicating and validating previous works; (2) using more objective evaluation criteria; (3) applying several computational intelligence techniques to datasets representative of diverse geographic areas; (4) preprocessing data and performing an extensive parameter optimization in order to improve previous results. |
Year | Venue | Keywords |
---|---|---|
2015 | PROCEEDINGS OF THE 2015 CONFERENCE OF THE INTERNATIONAL FUZZY SYSTEMS ASSOCIATION AND THE EUROPEAN SOCIETY FOR FUZZY LOGIC AND TECHNOLOGY | Stream flow forecasting,One step-ahead forecasting,ANFIS,Artificial Neural Networks,Support Vector Machines |
Field | DocType | Volume |
Hydropower,Data mining,Computational intelligence,Computer science,Support vector machine,Preprocessor,Ranging,Artificial intelligence,Adaptive neuro fuzzy inference system,Artificial neural network,Machine learning,Flood myth | Conference | 89 |
ISSN | Citations | PageRank |
1951-6851 | 2 | 0.40 |
References | Authors | |
8 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
João Paulo Carvalho | 1 | 110 | 17.52 |
Filipe V. Camelo | 2 | 2 | 0.40 |