Design and Implementation of a Real-time IoT Solution for Smart Meter Data Analysis in Microsoft Azure

This research presents an end-to-end IoT solution designed as a project, not a traditional lab setup. It simulates high-velocity data generation from smart meters and focuses on real-time data visualization within Azure’s cloud ecosystem. An integral part of this project is an AI-based spike and dip anomaly detection system. The project creates an IoT ecosystem that replicates smart meter telemetry data. It utilizes Microsoft Azure’s robust services to cover the entire IoT data lifecycle, from data ingestion to real-time visualization. Key project components involve Azure resource provisioning, IoT Hub setup, and the deployment of a Device Provisioning Service. The project implements a lambda architecture, filtering and processing a subset of telemetry data in real-time. This data undergoes immediate visualization, showcasing Azure’s agility in handling high-velocity streams. This paper is a valuable reference for interests in implementing IoT solutions with Microsoft Azure, particularly emphasizing real-time data processing and visualization within a lambda architecture while incorporating AI-based anomaly detection for spikes and dips.