Adaptive neuro-fuzzy inference system for flood forecasting in a large river system
This research investigated the application of Adaptive Neuron-Fuzzy Inference System (ANFIS) as a data-driven model for flood forecasting in the Lower Mekong, chosen as a case study of a large river system. A systematic study of the impacts of inputs such as the water levels at the station of intere...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Theses and Dissertations |
| Language: | English |
| Published: |
2012
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/50919 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This research investigated the application of Adaptive Neuron-Fuzzy Inference System (ANFIS) as a data-driven model for flood forecasting in the Lower Mekong, chosen as a case study of a large river system. A systematic study of the impacts of inputs such as the water levels at the station of interest, upstream stations and main tributaries and rainfall in the sub-basin on forecast accuracy was carried out. The ANFIS models developed in the study performed well for 1- and 3-lead-day water level forecasts when compared to the benchmark adopted by Mekong River Commission and to the results provided by the Unified Run-off Basin Simulation conceptual runoff routing model. Two approaches to improve ANFIS model results were investigated. The first approach was to employ an output updating procedure, while the second approach adopted an on-line learning algorithm in training the ANFIS model. Greater attenuation of forecast errors was achieved in the falling phase, although improvements during the rising phase of flood events remains a challenge. |
|---|