Claim Missing Document
Check
Articles

Found 2 Documents
Search

Perbandingan Prediksi Status Pompa Tandon Air Berbasis IoT Menggunakan Random Forest dan XGBoost Reni Veliyanti; Dani Samoko; Ummi Hanik
Joined Journal (Journal of Informatics Education) Vol 8 No 2 (2025): Volume 8 Nomor 2 (2025)
Publisher : Universitas Ivet

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31331/joined.v8i2.4294

Abstract

Pengelolaan air bersih di lingkungan rumah tangga dan instansi pemerintahan sering kali bergantung pada mekanisme manual atau saklar pelampung yang kurang efisien, menyebabkan pemborosan energi dan keterlambatan respons. Penelitian ini bertujuan mengembangkan model prediktif status pompa tandon air berbasis data sensor IoT, termasuk ketinggian air, kekeruhan, suhu, curah hujan, dan fitur temporal, untuk meningkatkan otomasi yang adaptif. Metodologi mencakup praproses data (normalisasi dengan StandardScaler, split 80:20), pelatihan enam algoritma machine learning (Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Random Forest, XGBoost, Multi-Layer Perceptron) menggunakan Python di Google Colab, dengan hyperparameter tuning via GridSearchCV dan 5-fold cross-validation. Evaluasi menggunakan akurasi, precision, recall, F1-score, serta analisis interpretabilitas melalui feature importance dan SHAP. Hasil menunjukkan Random Forest dan XGBoost mencapai performa sempurna (100%), sementara model lain di atas 96%, dengan water_level_cm sebagai fitur dominan (>55-78%). Analisis SHAP mengonfirmasi konsistensi logis model. Kontribusi utama adalah dataset IoT realistis untuk skenario tandon air, evaluasi komparatif model, dan interpretabilitas yang mendukung SDGs 6 dan 9, memungkinkan implementasi otomasi efisien di konteks tropis Indonesia.
Pompa Otomatis dan Monitoring Kekeruhan Tempat Penampungan Air: (Studi pada TPQ At – Ta’awun) Ummi Hanik; Dani Sasmoko; Nuris Dwi Setiawan; Iman Saufik Suasana; Sulartopo Sulartopo
Jurnal Informatika Dan Tekonologi Komputer (JITEK) Vol. 5 No. 3 (2025): November : Jurnal Informatika dan Tekonologi Komputer
Publisher : Pusat Riset dan Inovasi Nasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/jitek.v5i3.8122

Abstract

This research aims to design and develop an Internet of Things (IoT)-based automatic water pump monitoring and control system. The system utilizes an ESP32 microcontroller connected to an ultrasonic sensor to measure water level and a turbidity sensor to detect water clarity. The pump is automatically activated when the water level drops below a specified threshold and deactivated when the tank is full. Furthermore, when the turbidity value exceeds a certain limit (e.g., >40 NTU), the system sends notifications to the user via the Blynk application. In addition, notifications are also sent through a Telegram bot, allowing users to receive alerts instantly without needing to open the Blynk app. Key data such as water level distance, NTU value, pump status, and water condition are displayed locally on a 16x2 LCD and remotely on the Blynk dashboard.System testing was conducted using samples of clean water, tea, and milk coffee to represent varying levels of turbidity. The results showed that the system effectively detected different turbidity levels and automatically controlled the pump based on the water level. This system offers an efficient real-time solution for monitoring water quality and can be applied in places such as Islamic schools (TPQ), educational institutions, or households.