PREDICTION, MONITORING, AND EARLY WARNING SYSTEM OF WATER LEVELS IN FLOOD PROGRAM AREAS IN BANJARMASIN USING IoT AND LINEAR REGRESSION

Abstract

Banjarmasin is a city of a thousand rivers, floods often occur due to high tides and accompanied by rain. An early warning system is needed for monitoring and predicting river tides, especially in rivers that are prone to disasters. This study aims to develop an early warning system for river water levels using IoT and predicting linear regression analysis. The research method is through the manufacture of the NodeMCU V3 early warning system device , Microcontroller, and Ultrasonic A02YYUW. Monitoring to collect water level data at high tide is carried out on the banks of the river, then predictions are made using simple linear regression analysis. The results showed that an early warning system for river water levels at high tide can be read in the A02YYUW ultrasonic sensor database which is sent to the Mysql database. Water surface distance data with tools are used to predict early warning, simple regression analysis shows significant results on the t test and simple linear regression equation = 29.472 - 0.061 X + e. This means that the river water level at high tide approaches an early warning tool, namely an ultrasonic sensor. The conclusion of this study is that an IoT-based early warning system combined with simple linear regression can monitor and predict river water level rise.