Estimating Missing Precipitation to Optimize Parameters for Prediction of Daily Water Level Using Artificial Neural Network
This study proposes the application of Artificial Neural Network (ANN)in predicting missing precipitation to predicting daily water level for Sg. Bedup station located in Batang Sadong Basin, Sarawak.ANN is undoubtedly a strong tool for forecasting various non- linear hydrologic proc...
| Main Author: | |
|---|---|
| Format: | Final Year Project Report / IMRAD |
| Language: | English English |
| Published: |
UNIMAS
2006
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/7088/ http://ir.unimas.my/id/eprint/7088/1/Estimating%20missing%20precipitation%20to%20optimize%20parameters...%2824%20pgs%29.pdf http://ir.unimas.my/id/eprint/7088/4/Dayang%20Suhaila%20Awang%20Suhaili%20ft.pdf |
| Summary: | This study proposes the application of Artificial Neural Network (ANN)in predicting missing precipitation to
predicting daily water level for Sg. Bedup station located in Batang Sadong Basin, Sarawak.ANN is undoubtedly a strong tool for forecasting various non-
linear hydrologic processes, including the missing
precipitation and water level prediction. ANN was chosen based on its ability to extract the relation between the inputs and outputs of a process without the
physics known explicitly.In this study, the ANN was
developed specifically to predict the daily missing precipitation and data simulated are utilized to optimize
prediction accuracy for daily water level. Typical
networks were trained and tested using daily data obtained from the Drainage and Irrigation Department
(DID) Kota Samarahan.Various training parameters were considered in order to gain the best prediction possible. The performances of the ANN were evaluated based on the coefficient of correlation, R. The back propagation algorithm was adopted for this study. The optimal model for predicting missing data
found in this study is the network with the combination of learning rate and the number of neurons in the hidden layer of 0.2 and 60. This model generated the highest coefficient of correlation value of 0.964 when trained with the The Resilient Back
propagation(trainrp). It has been found that the ANN has the potential to solve the problems of estimation missing precipitatio in predicting daily water level. After
appropriate trainings, they are able to generate satisfactory results during both of the training and testing phases. |
|---|