Model Peramalan Banjir Di Das Bengawan Solo

Mularto, Listiya Hery (2009) Model Peramalan Banjir Di Das Bengawan Solo. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 3107205711-Master_thesis.pdf Restricted to Repository staff only Download (19MB)

Abstract

Sejak tahun 1863 Bengawan Solo telah menimbulkan banjir di daerah hulu, bahkan saat ini banjir sudah masuk kawasan hilir. Kejadian banjir yang terjadi pada akhir Desember 2007 telah banyak menyebabkan kerugian bagi masyarakat di sekitar sungai, baik kerugian harta benda maupun nyawa. Salah satu upaya yang penting saat akan terjadi banjir adalah bagaimana cara mengetahui kapan waktu datangnya banjir jauh sebelum banjir tersebut terjadi, sehingga penduduk sekitar sungai memiliki kesempatan untuk menyelamatkan diri dan harta bendanya. Tujuan dari penelitian ini adalah membangun sistem peramalan yang dapat meramalkan waktu datangnya banjir sebelum kejadian banjir tersebut datang.Penelitian diawali dengan melakukan studi literatur dan mengumpulkan data-data elevasi muka air di beberapa stasiun pengamatan yang memiliki korelasi terbesar terhadap daerah yang akan dijadikan fokus peramalan, dalam hal ini stasiun Bojonegoro, Babat dan Kuro. Data-data yang diperoleh digunakan sebagai input dalam membangun model peramalan. Pemilihan variabel input yang berpengaruh terhadap variabel output dilakukan menggunakan analisa korelasi. Metode peramalan menggunakan data driven model antara lain: M5 Model Tree dan Artificial Neural Network (ANN), dimana proses pembelajarannya (learning) menggunakan program bantu Weka Knowledge Explore. Kelayakan performa kedua model tersebut melalui uji verifikasi.Dari kedua model yang digunakan, model peramalan yang terpilih untuk peramalan muka air di Bojonegoro untuk 1, 3 dan 6 jam ke depan adalah model M5 Model Tree dengan nilai RMSE (Root Mean Square Error) saat verifikasi berkisar 0,278 - 0,717, sedangkan untuk di daerah Babat dan Kuro, model yang terpilih untuk peramalan elevasi muka air saat 6 jam ke depan adalah model ANN dengan nilai RMSE 0,432 dan 0,154.
===================================================================================================================================
Since 1863, Bengawan Solo has caused floods in the upstream area, and even nowadays the flood has already entered the downstream area. The flood that occurred at the end of December 2007 caused a big loss of property and life as well for the people near it. One of the essential efforts before the flood happens is how to know when the flood will come, far before it really happens, so the citizens who are settled around the river can have a chance to save themselves and their belongings. The purpose of this research is to build a forecasting system which can forecast the coming time of flooding before it really comes.This research started by doing a literature study and collecting water level data in some AWLR (Automatic Water Level Record) stations which have good correlation with the area to be forecast, in this case Bojonegoro station, Babat station and Kuro station. All the data collected will be used as input to build the forecasting model. The choice of input variables that may affect output variables was performed using correlation analysis. The forecasting method uses data driven models such as M5 Model Tree and Artificial Neural Network (ANN), with the learning process performed by Weka Knowledge Explore software. The performance capability of both models is tested with a verification test.From both of the models, the forecasting model chosen to forecast water level in Bojonegoro for the next 1, 3 and 6 hours is M5 Model Tree with an RMSE (Root Mean Square Error) verification value of about 0.278 - 0.717, while for Babat and Kuro, the model chosen for the next 6 hours is the ANN model with RMSE verification values of 0.432 and 0.154.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Model peramalan, Elevasi muka air Bengawan Solo, RMSE, M5 Model Tree, Artificial Neural Network, Data Driven Model , Forecasting model, Bengawan Solo water level, RMSE, MS Model Tree, Artificial Neural Network, Data Driven Model.
Subjects:	G Geography. Anthropology. Recreation > GB Physical geography > GB1399.2 Flood forecasting.
Divisions:	Faculty of Civil Engineering and Planning > Civil Engineering > 22101-(S2) Master Thesis
Depositing User:	magang .
Date Deposited:	24 Sep 2026 06:25
Last Modified:	24 Sep 2026 06:25
URI:	http://repository.its.ac.id/id/eprint/144870

Actions (login required)

View Item