Perancangan Program Peramalan Kanal Banjir Barat Jakarta Menggunakan Autoregresi Multivariant

Abstract

This paper discusses the design of computer programs that is able to discern the characteristics description of water surface elevation data in Manggarai water gate, which variable is the most influential on the water surface elevation model and find a proper flood forecasting model using multivariate autoregressive model. The result of this study is able to assist the water gate officer in delivering early warning, prevention and anticipation of flood countermeasure. The forecast equation model obtained is Yt = 109,.7828 + 0,9291 CHt-6 – 24,484 T t-2 – 0,06245 PM t-2 + 1,4706 KB t-2 in which temperature and water surface elevation is a variable that owns the strongest correlation. This variable owns negative correlation which means that if the temperature falls, the water levels will rise. The coefficient of determination has a value of R2 = 0.4056 and the Durbin Watson statistics for DW = 0.7429.

Keywords

• flood forecasting, Manggarai water gate, multivariate autoregressive model