Rizky, Muhammad (2017) Pemodelan Smart Profile Greenhouse Berbasis Neural Network. Undergraduate thesis, institut teknologi sepuluh nopember.
Preview |
Text
1112100075-Undergraduate_Theses.pdf - Published Version Download (4MB) | Preview |
Abstract
Pemodelan smart profile greenhouse merupakan salah satu hal penting untuk memaksimalkan pertumbuhan tanaman yang ditanam di dalam greenhouse. Sebelum dimodelkan, dibandingkan ketika greenhouse dengan sistem otomasi dalam keadaan mati dan ketika greenhouse dengan sistem otomasi dalam keadaan hidup. Dengan diterapkan sistem otomasi dapat meningkatkan kualitas greenhouse dengan cara menurunkan suhu dan meningkatkan kelembaban. Semakin baik kualitas greenhouse maka semakin baik pertumbuhan tanaman dalam greenhouse. Pemodelan dalam penelitian ini digunakan model neural network tipe back-propagation. Smart profile greenhouse meliputi suhu udara, kelembaban udara, suhu tanah dan kelembaban tanah. Hasilnya ialah profil terbaik dari greenhouse adalah pada suhu udara pukul 16.00 – 07.00, kelembaban udara mencapai 98%, suhu tanah pukul 18.00 – 08.00 dan kelembaban tanah mencapai 98% serta hasil pemodelan mendekati data pengukuran dengan nilai kesalahan mencapai 1%
==================================================================================
Modeling smart profile greenhouse is one of the important things to maximize the growth of plants grown in the greenhouse. Before modelled, compared to when the greenhouse with the automation system in a State of death and when the greenhouse with automation system alive. With applied automation system can improve the quality of the greenhouse by means of lowering the temperature and increase the humidity. The better the quality greenhouse then the better the growth of plants in the greenhouse. Modeling in the study used a model of neural network back-propagation type. Smart profile greenhouse includes air temperature, air humidity, soil temperature and soil moisture. The result is best profiles from the temperatures of the greenhouse is at 4 pm – 7 am, air humidity reaches 98%, ground temperatures at 6 pm – 8 am soil moisture reaches 98% as well as the results of the modelling approach to data measurement error value reaches 1%.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | greenhouse, suhu, kelembaban, neural network |
| Subjects: | Q Science Q Science > QC Physics |
| Divisions: | Faculty of Mathematics and Science > Physics |
| Depositing User: | muhammad rizky |
| Date Deposited: | 10 Oct 2017 07:41 |
| Last Modified: | 08 Mar 2019 02:27 |
| URI: | http://repository.its.ac.id/id/eprint/46350 |
Actions (login required)
 |
View Item |