Firdaus, Muhammad Zidan (2025) Rancang Bangun Sistem Monitoring Faktor Lingkungan dan Nutrisi Untuk Prediksi Pertumbuhan Tanaman Pakchoy Putih Pada Hidroponik Nutrient Film Technique Menggunakan Algoritma Support Vector Regression Berbasis Internet of Things. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5009211111-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5009211111-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy |
Abstract
Inovasi dalam pemantauan dan prediksi pertumbuhan tanaman menjadi dasar utama dalam pertanian presisi. Penelitian ini merancang sistem monitoring berbasis internet of things (IoT) dan model prediksi estimasi berat tanaman pakchoy putih (brassica rapa) menggunakan algoritma support vector regression (SVR) pada sistem hidroponik nutrient film technique (NFT). Sistem menggunakan mikrokontroler ESP32 untuk memantau suhu, kelembapan udara, intensitas cahaya, electrical conductivity (EC), dan pH secara real-time. Data yang dikumpulkan digunakan untuk membangun model SVR dengan pendekatan pemetaan statis antara parameter lingkungan dan berat tanaman. Hasil kalibrasi menunjukkan akurasi dan linearitas sensor yang baik dengan ketidakpastian sebesar 0.655 °C, 4.22%, 328.45 lux, 20.24 µS/cm, dan 0.192. Model SVR dengan kernel polinomial mencapai nilai RMSE, MSE, MAE, dan koefisien determinasi (R²) masing masing sebesar 0.4299, 0.3124, dan 0.97 pada data uji. Evaluasi sistem IoT menunjukkan performa yang andal dengan throughput 16.75 kbps, delay rata-rata 203.9 ms, jitter 23.8 ms, dan packet loss 0.37%. Hasil ini menunjukkan bahwa estimasi berat tanaman dapat dilakukan secara akurat berdasarkan data lingkungan yang dimonitor secara otomatis.
=======================================================================================================================================================
Innovation in monitoring and predicting plant growth is the main basis for precision agriculture. This research designs an Internet of Things (IoT)-based monitoring system and a prediction model for estimating the weight of white pakchoy (Brassica rapa) plants using the Support Vector Regression (SVR) algorithm in the Nutrient Film Technique (NFT) hydroponic system. The system uses an ESP32 microcontroller to monitor temperature, air humidity, light intensity, Electrical Conductivity (EC), and pH in real-time. The collected data was used to build an SVR model with a static mapping approach between environmental parameters and plant weight. The calibration results showed good accuracy and linearity of the sensors, with uncertainties of 0.655 °C, 4.22%, 328.45 lux, 20.24 µS/cm, and 0.192, respectively. The SVR model with polynomial kernel achieved RMSE, MSE, MAE, and coefficient of determination (R²) values of 0.4299, 0.3124, and 0.97 on the test data, respectively. The IoT system evaluation showed reliable performance with a throughput of 16.75 kbps, average delay of 203.9 ms, jitter of 23.8 ms, and packet loss of 0.37%. These results show that plant weight estimation can be done accurately based on automatically monitored environmental data.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Estimasi Berat Tanaman, Hidroponik Nutrient Film Technique, Internet of Things, Monitoring Lingkungan, Support Vector Regression, Environmental Monitoring, Internet of Things, Nutrient Film Technique Hydroponics, Plant Weight Estimation, Support Vector Regression. |
| Subjects: | T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. |
| Divisions: | Faculty of Industrial Technology > Physics Engineering > 30201-(S1) Undergraduate Thesis |
| Depositing User: | Muhammad Zidan Firdaus |
| Date Deposited: | 04 Aug 2025 09:31 |
| Last Modified: | 04 Aug 2025 09:31 |
| URI: | http://repository.its.ac.id/id/eprint/125383 |
Actions (login required)
 |
View Item |