Rainfall runoff model using probability distributed model(PDM)
Probability Distributed Model (PDM) is widely used in analyzing the hydrological behavior. One of the applications is involved in rainfall runoff modeling to forecast the occurring of flood in Johor Bahru, Malaysia. In this study, Pareto distribution is used to represent the storage capacity of PDM....
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/40569/1/YazreYusofMFS2014.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Probability Distributed Model (PDM) is widely used in analyzing the hydrological behavior. One of the applications is involved in rainfall runoff modeling to forecast the occurring of flood in Johor Bahru, Malaysia. In this study, Pareto distribution is used to represent the storage capacity of PDM. By providing the best fit between observed and simulated discharges, a Genetic Algorithm (GA) method has been applied to optimize the optimal parameter of PDM. The performance of PDM is accessed through the calibration and validation of different data with the same parameter. The model was applied to Sungai Pengeli (Station B) and the performance was assessed using the values R-squared value. A strong relationship between observed and calculated discharge was detected. In order to forecast water level days ahead, the error prediction employs the Autoregressive Moving Average (ARMA) model which is one of the model of time series. From the analysis, the trend of change in water level of station Sungai Pengeli (Station B) is quite accurately captured by using PDM with to small differences of the water level between actual and forecast values. |
|---|