APLIKASI JARINGAN SYARAF TIRUAN UNTUK MENGESTIMASI DATA CURAH HUJAN TAHUNAN YANG TIDAK TEREKAM DI DAS CISADANE
ABSTRACT: Naturally in a
watershed rainfall distributes spatially. To know rainfall in the watershed
needs information from many installed rain gauges. However, rainfall data is
found not completely recorded. It is then important to estimate missing or unrecorded
rainfall data. This study aims to estimate annual rainfall data in stations by
using ANN (Artificial Neural Network). This study was conducted in Cisadane
watershed. This study perfomed using rainfall data for 14 periods, the location
of rainfall post (coordinates and elevation), DEM map, and watershed map. Data
processing and analyzing performed using Ms. Excel 2010, ArcGIS 10.0, and
BackPropogation Neural Network 1.0 program. Data used as input in ANN to estimates
unrecorded rainfall data were coordinates (X,Y) and elevation (Z) of each
rainfall post. ANN can be used to predict the amount of rainfall in cisadane
watershed marked with a value of determination (R2) 0,97. After all data
complete, average of rainfall in Cisadane watershed can be calculate using
arithmetic, thiessen polygon, and isohyet. The amount of rainfall watershed in
Cisadane using the arithmetic mean produce rainfall of 2.609 mm, with Thiessen
Polygon of 2.539 mm, and with Isohyets of 2.594 mm.
Penulis: Elhamida Rezkia Amien,
Roh Santoso Budi Waspodo, Budi Indra Setiawan, Rudiyanto
Kode Jurnal: jptsipildd160445
