Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/165944
Title: Penerapan Algoritma Genetika dan Jaringan Saraf Tiruan untuk Penentuan Tinggi Muka Air Optimum pada Budidaya Padi dengan Sistem Irigasi Evapotranspiratif
Other Titles: Application of Genetic Algorithm and Artificial Neural Network for Determining the Optimal Water Level in Rice Cultivation with an Evapotranspirative Irrigation System
Authors: Arif, Chusnul
Sahputra, Ivan Triadi
Issue Date: 2025
Publisher: IPB University
Abstract: Padi merupakan tanaman yang rentan terhadap kelebihan atau kekurangan air. Sehingga, peningkatan produktivitas padi serta penggunaan air yang efisien memerlukan sistem pengairan irigasi yang tepat. Penelitian ini bertujuan untuk mengetahui hubungan antara tinggi muka air setiap fase dengan total air irigasi dan produktivitas tanaman menggunakan model Jaringan Saraf Tiruan (JST) dan menentukan tinggi muka air optimum pada masing-masing fase pertumbuhan tanaman padi menggunakan Algoritma Genetika (AG). Budidaya padi dilakukan menggunakan metode System of Rice Intensification (SRI) dan terdiri dari 4 skenario tinggi muka air yaitu 5 cm (TA1), 0 cm (TA2), -5 cm (TA3), dan -10 cm (TA4). Hasil pemodelan JST menunjukkan keakuratan prediksi tinggi muka air setiap fase dengan produktivitas tanaman dan total air irigasi dimana nilai koefisien determinasi (R2) sebesar 0,9971 dan 0,9996. Berdasarkan hasil optimasi tinggi muka air dengan model AG, prediksi tinggi muka air pada setiap fase pertumbuhan tanaman sebesar 5,05 cm (WL1); -3,91 cm (WL2); -7,63 cm (WL3); dan -10 cm (WL4) dapat meningkatkan produktivitas tanaman hingga 707,35 g/m2 dengan persentase peningkatan mencapai 44,45% dan menghemat penggunaan air irigasi hingga 183,90 mm dengan persentase penghematan mencapai 54,33%.
Rice is a crop that is highly sensitive to both excess and deficiency of water. Therefore, increasing rice productivity and achieving efficient water use requires an appropriate irrigation system. This research aims to determine the relationship between water table depth at each growth phase, total irrigation water, and crop productivity using an Artificial Neural Network (ANN) model. Additionally, it seeks to identify the optimal water table depth at each growth phase of rice using a Genetic Algorithm (GA). Rice cultivation was carried out using the System of Rice Intensification (SRI) method, with four water table depth scenarios: +5?cm (WT 1), 0?cm (WT 2), –5?cm (WT 3), and –10?cm (WT 4). Results from the ANN model showed high accuracy in predicting the relationship between water table depth at each phase, total irrigation water, and crop productivity, with determination coefficients (R2R^2R2) of 0.9971 and 0.9996. Based on GA optimization, the predicted optimal water table depths for each growth phase are +5.05?cm, –3.91?cm, –7.63?cm, and –10?cm. This optimization can increase crop productivity up to 707.35?g/m² (a 44.45% increase) and reduce irrigation water use by 183.90?mm (a 54.33% saving).
URI: http://repository.ipb.ac.id/handle/123456789/165944
Appears in Collections:UT - Civil and Environmental Engineering

Files in This Item:
File Description SizeFormat 
cover_F4401211080_809cc9b4677f46b7abf479cb5392973e.pdfCover1.13 MBAdobe PDFView/Open
fulltext_F4401211080_5824de8f716943cc9b76a64a67720b12.pdf
  Restricted Access
Fulltext2.19 MBAdobe PDFView/Open
lampiran_F4401211080_d27041685247461ebe73ef9bd05bfdcf.pdf
  Restricted Access
Lampiran2.85 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.