Real Time Energy Consumption Monitoring and Prediction for University Campus
DOI:
https://doi.org/10.3126/jsce.v12i2.91419Keywords:
Smart meter, Load profile, Machine learning, Load Forecasting, Energy monitoringAbstract
The real time energy consumption monitoring system at the various buildings of university campus is quite necessary task for efficient use of energy. Such monitoring system helps to understand the proper behavior of consumers and ultimately, load demand patterns of various buildings. In this regard, a real-time energy monitoring system has been implemented at Kathmandu University. The system comprises six smart meters installed in university buildings at major energy consumption points to obtain high-resolution data on energy usage. Robust energy prediction models have been developed by combining electricity consumption data with weather information, including academic holidays. The prediction performance of the models is examined using evaluation metrics such as the root mean square error (RMSE) and execution time. A mobile application and web dashboard have also been developed to visualize the results and provide feedback to consumers and utility providers. This visualization dashboard provides real-time information, identifies energy consumption patterns, potential overloads, and ultimately supports consumers in optimizing their electricity usage.