Design and Implementation of an IoTBased Smart Energy Meter with Machine LearningBased Appliance Identification and Usage Optimization

This study proposes an Internet-of-Things (IoT) based smart energy metering system enhanced with machine learning techniques for real-time monitoring, appliance recognition, and energy optimization. The developed system acquires electrical and environmental measurements continuously, assigns accurate timestamps using a real-time clock, and transmits the data to a cloud platform through wireless communication. To ensure reliable operation during connectivity interruptions, measurements are temporarily stored in local memory and automatically synchronized once the network is restored. Machine learning algorithms process the collected time-series data to detect active appliances and analyze consumption behavior. A context-aware optimization module further integrates ambient temperature information to generate personalized daily recommendations for improving energy efficiency. The proposed framework demonstrates a practical, scalable, and economical approach for intelligent energy management in smart home environments and future smart grid systems.