A comparison between normal ratio method and a data-driven, expectation-maximization algorithm for filling In missing rainfall values | |
| Author | Mudiyanselage, Viyath Araliya R. |
| Call Number | AIT Caps. Proj. no.CIE-17-11 |
| Subject(s) | Artificial intelligence Rain and rainfall--Mathematical models |
| Note | A capstone project report submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Engineering Civil and Infrastructure Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Caps. Proj. ; no. CIE-17-11 |
| Abstract | Rainfall precipitation is an important parameter in hydrology and climatology. Rainfall pattern recognition plays an important role in irrigation structures such as damns, hydrological power stations, canals and reservoirs. Therefore maintaining a continuous rainfall data record is essential. However due to issues such as instrument malfunctioning, lack of observers, rainfall data records contain gaps(missing data). Since continuous data records is an essential for weather prediction models and runoff model, numerous techniques have been developed over time to fill in gaps and ignore the missing data. They include both Statistical and Artificial Intelligence techniques. The accuracy of the predicted data differs accordingly. The suitability of a given method may differ according to the climatologically and geographical details of the selected area. Therefore it is important to select a method which have a higher correlations with the selected area. This study focuses on comparing the reliability two methods which have been adopted to calculated missing rainfall data, to obtain missing values for Badullu Oya catchment area. The selected catchment area belongs to the upcountry wet zone of Sri Lanka, a mountainous area with higher altitudes which varies with the location. And it contributes for various number of hydrological projects such as Hydro Power stations and irrigation canals and reservoirs. One of the methods which will be studied in this study is a statistical method known as Normal Ratio Method and other one is an artificial intelligence technique, known as Expectation Maximization. Ten rainfall gauges have been selected for this study and daily rainfall data from year 2000 to year 2015 will be used for analysis. |
| Year | 2017 |
| Corresponding Series Added Entry | Asian Institute of Technology. Caps. Proj. ; no. CIE-17-11 |
| Type | Capstone Project |
| School | School of Engineering and Technology (SET) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Civil and Infrastructure Engineering (CIE) |
| Chairperson(s) | Andriyas, Sanyogita |
| Examination Committee(s) | Shrestha, Sangam |
| Degree | Capstone Project (B.Sc.)-Asian Institute of Technology, 2017 |