| Summary: | 碩士 === 中華大學 === 土木工程學系 === 105 === The objective of this research is to apply the Adaptive Network-Based Fuzzy Inference System, called ANFIS, to predict and understand the River Bed Migration which is caused by Erosion and Sedimentation. After doing on-site investigation and related data collection (River Project Discharge, River Bigger Cross-Section, Pre and Post Harness Plan), we can learn about the main factors which cause the change of river flow (flow rate, rainfall, inclination and sediment concentration). This research will focus on Lan-Yang River midstream (Section 15〔Luo Dong River Confluence〕~ Section 35〔Cai-Lien Dike〕); It is calculated using ANFIS module which is integrated in the MATLAB Program, using Fuzzy Inference data base to establish flood peak flow rate forecasting mechanism and to estimate the river flow transition simulation under normal and heavy rainfall condition.
This research result utilizes ANFIS Uncertainty Analysis Methods in predicting the river migration Correlation Coefficient which (A) under normal rainfall condition will result in 92.8%; relatively (B) under heavy rainfall condition, Section 25, 31 and 34 will result in bigger change. This will need to be enhanced with retaining embankment. At the same time, we will need to collect more data in order for ANFIS to perform more accurate result, which is the objective of applying Artificial Intelligence for Prediction on River Channel Migration under heavy rainfall condition.
|
|---|
