Prediction of flood events in the city of Bandar Lampungusing the artificial neural network

Nurpambudi, Ramadhan and Wulandari, Eka Suci Puspita and Aziz, RZ Abdul (2023) Prediction of flood events in the city of Bandar Lampungusing the artificial neural network. Infotel, 15 (1). pp. 34-45. ISSN 24600997

Text document.pdf Download (1MB)

Abstract

The city of Bandar Lampung, Indonesia, is recently experiencing seasonal flooding, which occurs almost everyyear and results in significant losses. In the last 10 years, floods event recorded by the National Agency for DisasterCountermeasure (BNPB) in the Bandar Lampung area is 16 incidents of flooding. More than 14,000 people suffered, morethan 500 people had to be evacuated, more than 900 houses were damaged, and four public facilities were damaged. To studythe pattern of flood events in the past, the Artificial Neural Network Backpropagation learning method will be used whichwill utilize its non-linear variable learning abilities. The configuration settings for the Artificial Neural Network were carriedout experimentally without any basis for assigning values, especially for the parameters of the number of hidden layers,number of neurons, and epochs used in training and variable testing. The results obtained from this study are the results oftraining and testing of datasets that have been carried out by ANN backpropagation and can properly study patterns of floodevents and also non-flood events in the dataset, this is evidenced by the results of high model configuration accuracy andalso the results of predictive tables that able to describe actual conditions, setting the configuration model experimentallycan produce an accuracy value of 90 %-100 %, an average training correlation value of 0.96 and an average test correlationvalue of 0.89, and an average error value of 0.0089 out of 20 model configuration, and the flood prediction table are madebased on the 1 best configuration with a training and testing accuracy rate of 100 % with an error value of 0.00134, namelyconfiguration model 20, the prediction table uses an average air temperature of 27◦C with 80 % humidity. The predictiontable can produce excellent flood potential results which can represent flood events as well as non-flood events based on theresults of the dataset learning.

Item Type:	Article
Subjects:	Ilmu Komputer
Divisions:	Artikel Ilmiah Dosen > Fakultas Ilmu Komputer
Depositing User:	Dr. RZ Abdul Aziz
Date Deposited:	28 May 2024 06:27
Last Modified:	28 May 2024 06:27
URI:	http://repo.darmajaya.ac.id/id/eprint/16246

Actions (login required)

View Item