Energy Consumption Tracker and AI Driven Framework for Real-Time Tracking and Forecasting of Solar Energy Utilization
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Solar power is an integral part of long-term energy planning, but it has the problem of changing naturally. With the rather random as well as natural daily variation ion weather, this means that power generation can be extremely unpredictable. Addressing this problem, we developed a smart real-time prediction system based on Internet of Things and AI. First, we developed an IoT sensor node which has the capability to report a significant aspect of operation such as the voltage, current and the light intensity. We carefully tested three different predictive models: Linear Regression, SVM with RBF, and time-series oriented ARIMA, using the rich time series dataset that we collected. The results of our experiment clearly showed that ARIMA was the better algorithm. The model worked excellently for this application, with less than 5% MAPE, and R2 predictive accuracy of more than 0.95.
Fields of Study & Subjects
Solar Radiation and Photovoltaics
Energy Load and Power Forecasting
Photovoltaic System Optimization Techniques
Publication details
- Journal / source
- International Conference on Power Electronics & IoT Applications in Renewable Energy and its Control (PARC)
- Pages
- 182–186
- Publication date
- 20 February 2026
- Language
- EN