Khamami
Electrical Engineering Department, Politeknik Negeri Semarang, Central Java, 50275, Indonesia

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Real-time IoT-Based Monitoring and ANN-Driven Prediction of Electrical Parameters in a 1200 W Photovoltaic System Eko Supriyanto; Abu Hasan; Tulus Pramuji; Tri Raharjo Yudantoro; Khamami; Roni Apriantoro; Hanny Nurrani
JAICT Vol. 12 No. 1 (2026): JAICT
Publisher : Politeknik Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The increasing adoption of small-scale photovoltaic (PV) systems underscores the necessity for advanced monitoring that improves dependability and maintenance efficacy. Conventional photovoltaic monitoring emphasizes real-time data acquisition but is deficient in predictive capabilities, hindering early defect identification and proactive maintenance. This study introduces an IoT-enabled real-time monitoring system utilizing ANN-based predictions for a 1200 W photovoltaic configuration. Voltage and current sensors interface with an ESP32 microcontroller to quantify and analyze voltage, current, power, and energy. Data is transmitted to the cloud using MQTT, enabling users to remotely monitor and display information through web or mobile applications. A feedforward Artificial Neural Network (ANN) taught by backpropagation enhances intelligence by predicting future electrical performance based on historical data. Experimental findings indicate a consistent communication latency of 1-2 seconds and sensor inaccuracies below 3%. The ANN model attained a MAPE of 3-4%. The integration of IoT monitoring with ANN prediction facilitates early anomaly detection, enhances operational understanding, and enables scalable, intelligent energy management.