IoTeeHee

VIT VEllore

Team ID : IH059RT06B

HACK ID : RT06

Identifying appliance-level energy consumption without intrusive sensors enables better energy efficiency and cost savings. Traditional monitoring methods lack granularity and actionable insights.

Smart Energy Fingerprinting for Buildings

PS41

IOT & AR / VR

PROPOSED SOLUTION & WORKING

• ESP32-based device measures voltage and current in real time using a CT sensor
• Energy data is sent to a Go backend API and stored in MongoDB
• ML-based NILM (KNN + clustering) analyzes power patterns to identify appliances
• Next.js dashboard visualizes live data, history, and insights

​

FEATURES & TECH STACK

Features:

1. Non-intrusive appliance-level energy monitoring
2. Real-time power consumption visualization
3. ML-based appliance identification
4. Historical usage analytics and insights
5. Scalable deployment using Docker and Kubernetes

​

​

Tech stack:Hardware: ESP32, SCT-013 CT sensor, passive components

• Backend: Go (Golang), MongoDB
• Frontend: Next.js
• AI/ML: KNN, clustering (NILM)
• Deployment: Docker, Kubernetes

​

​

Novelty and USPs

Novelty:

• Non-intrusive energy fingerprinting using a single sensing point
• Hybrid ML approach combining KNN and clustering for appliance identification
• Learning-based system that adapts to different appliance usage patterns

​

USPs:

• Low-cost, retrofit-friendly hardware (ESP32 + SCT-013)
• No per-appliance sensors required, reducing installation complexity
• Scalable, deployment-ready architecture using Docker and Kubernetes

​

SDGs

SDGs (Sustainable development goals: The sustainable development goals (Set by WHO) that our product/innovation meets or solves.

• [SDG 7] – Affordable and Clean Energy
• [SDG 11] – Sustainable Cities and Communities
• [SDG 12] – Responsible Consumption and Production
• [SDG 13] – Climate Action
• [SDG 9] – Industry, Innovation and Infrastructure�

​

​

​

​

THANK YOU