Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/160080
Title: Pemodelan Pertumbuhan Tanaman Selada dan Kale Berdasarkan Pengaruh Jenis Pencahayaan dalam Plant Factory Menggunakan Artificial Neural Network
Other Titles: Modeling the Growth of Lettuce and Kale Plants Based on the Effect of Lighting Type in a Plant Factory Using Artificial Neural Network
Authors: Suhardiyanto, Herry
Supriyanto
Silfiya, Anisya Dika
Issue Date: 2024
Publisher: IPB University
Abstract: Pertanian dalam lingkungan terkendali seperti hidroponik menawarkan suatu solusi alternatif untuk tantangan ketahanan pangan global, terutama di tengah perubahan iklim dan berkurangnya lahan subur. Penelitian ini bertujuan untuk mengetahui pengaruh jenis pencahayaan terhadap pertumbuhan selada romaine (Lactuca sativa var. longifolia) dan kale Curly (Brassica oleracea var. sabellica) di plant factory menggunakan white LED, grow LED merah-biru, dan grow LED kombinasi. Metode yang digunakan meliputi uji ANOVA dan analisis lanjut DMRT untuk mengevaluasi perbedaan pertumbuhan kedua jenis tanaman. Selain itu, model Artificial Neural Network (ANN) dengan algoritma backpropagation dikembangkan untuk memprediksi pertumbuhan tanaman berdasarkan parameter lingkungan dan kondisi tanaman. Hasil penelitian menunjukkan bahwa jenis pencahayaan berpengaruh signifikan terhadap parameter pertumbuhan seperti jumlah daun, luas daun, dan bobot akhir tanaman. Grow LED merah-biru memberikan hasil terbaik dengan bobot akhir selada romaine mencapai 79,01 gram dan curly kale 37,19 gram. Model ANN yang dikembangkan berhasil memprediksi pertumbuhan tanaman dengan nilai R² sebesar 0,9913 dan RMSE sebesar 2,0579 untuk selada romaine, serta R² sebesar 0,9918 dan RMSE sebesar 0,6541 untuk curly kale, menunjukkan hubungan erat antara parameter lingkungan dan pertumbuhan tanaman. Penelitian ini mengindikasikan bahwa pemilihan pencahayaan yang tepat dapat meningkatkan efisiensi produksi dalam sistem pertanian lingkungan terkendali, memberikan dasar kuat untuk optimasi budidaya tanaman hidroponik.
Agriculture in a controlled environment, such as hydroponics, offers an alternative solution to global food security challenges, especially amid climate change and reducing arable land. This study aims to determine the effect of lighting type on the growth of Romaine lettuce (Lactuca sativa var. longifolia) and curly kale (Brassica oleracea var. sabellica) in a plant factory using white LED, red-blue LED grow, and combined LED grow. The methods used include ANOVA test and DMRT further analysis to evaluate the difference in growth of the two types of plants. In addition, an Artificial Neural Network (ANN) model with a backpropagation algorithm was developed to predict plant growth based on environmental parameters and plant conditions. The results showed that the lighting type significantly affected growth parameters such as number of leaves, leaf area, and final plant weight. The red-blue LED growth gave the best results, with the final weight of Romaine lettuce reaching 79,01 grams and curly kale at 37,19 grams. The ANN model thrived predicted plant growth with R² values of 0.9913 and RMSE of 2,0579 for Romaine lettuce and R² of 0,9918 and RMSE of 0,6541 for curly kale, indicating a close relationship between environmental parameters and plant growth. This study suggested that proper lighting selection can improve production efficiency in controlled environment farming systems, providing a solid basis for optimizing hydroponic crop cultivation.
URI: http://repository.ipb.ac.id/handle/123456789/160080
Appears in Collections:UT - Agricultural and Biosystem Engineering

Files in This Item:
File Description SizeFormat 
cover_F1401201101_57233ea7a3324341acdadb9f6262ed39.pdfCover457.29 kBAdobe PDFView/Open
fulltext_F1401201101_096a1e29cf814eee8c33dc11e2fbf9c8.pdf
  Restricted Access
Fulltext2.72 MBAdobe PDFView/Open
lampiran_F1401201101_a561ac305d3f4b079a3f38f6e4fa7586.pdf
  Restricted Access
Lampiran1.79 MBAdobe PDFView/Open


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