Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/166484| Title: | Integrasi Monitoring Suhu, Kelembapan, dan pH pada Sistem Komposter Anaerob berbasis IoT |
| Other Titles: | |
| Authors: | Mindara, Gema Parasti Ramdani, Gusti |
| Issue Date: | 2025 |
| Publisher: | IPB University |
| Abstract: | Sampah organik masih menjadi tantangan serius di Indonesia. Pengomposan anaerob merupakan salah satu solusi potensial untuk mengurangi sampah organik. Penelitian ini merancangan sistem Internet of Things (IoT) untuk memantau proses pengomposan anaerob secara realtime menggunakan sensor suhu DS18B20, sensor kelembapan DHT11, dan sensor pH tanah. Data dari sensor diproses oleh ESP32 dan ditampilkan melalui website. Sistem ini dilengkapi dengan implementasi model machine learning Random Forest untuk memprediksi kebutuhan penyiraman secara otomatis berdasarkan pola perubahan parameter lingkungan yang terdeteksi oleh sensor. Hasil pengujian menunjukan sistem monitoring suhu, kelembapan, dan pH pada komposter anaerob berjalan dengan baik dan dapat dipantau melalui website. Mekanisme otomatisasi penyiraman berfungsi sesuai kondisi yang diperlukan. Model machine learning Random Forest menunjukan akurasi 99.28% untuk model klasifikasi, sementara Mean Absolute Error (MAE) sebesar 0.6728 L/min dan Root Mean Squared Error (RMSE) sebesar 0.8272 L/min pada data uji. Organic waste presents a significant challenge in Indonesia, and anaerobic composting offers a potential solution. This research designs an Internet of Things (IoT) system to monitor the anaerobic composting process in real-time using a DS18B20 temperature sensor, a DHT11 humidity sensor, and a soil pH sensor. Data from the sensors is processed by an ESP32 microcontroller and displayed via a website. The system is equipped with a Random Forest machine learning model to automatically predict watering needs based on patterns in the environmental parameters detected by the sensors. The test results show that the monitoring system for temperature, humidity, and pH functions well and can be monitored through the website, with the automated watering mechanism operating according to the required conditions. The machine learning model demonstrated an accuracy of 99.28% for the classification task, while the regression task achieved a Mean Absolute Error (MAE) of 0.6728 L/min and a Root Mean Squared Error (RMSE) of 0.8272 L/min on the test data. |
| URI: | http://repository.ipb.ac.id/handle/123456789/166484 |
| Appears in Collections: | UT - Computer Engineering Tehcnology |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_J0304211118_f213a5cccea54da895ff63af9641b10c.pdf | Cover | 1.52 MB | Adobe PDF | View/Open |
| fulltext_J0304211118_e6f4aaf7519845c49a764babd5f89aa8.pdf Restricted Access | Fulltext | 9.45 MB | Adobe PDF | View/Open |
| lampiran_J0304211118_99dea835d47645e6a779922a5c98d46f.pdf Restricted Access | Lampiran | 838.85 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.