Smart Fan Energy Behavior Analyzer

Inspiration

Our project started from a simple question: how much energy do everyday automatic devices use without us realizing it?

Many devices like fans, heaters, and air conditioners run automatically based on environmental conditions. However, most of them are not truly smart. They react to temperature but don’t consider whether their behavior is actually efficient.

Since the theme of the project was sustainability, we started thinking about how these small automatic behaviors add up over time. Even a small device running unnecessarily can contribute to electricity costs and the CO₂ emissions associated with power generation. We wanted to build a system that makes this relationship visible and explores how smarter decision logic could reduce wasted energy.

What it does

Smart Fan Energy Behavior Analyzer is a weather-adaptive IoT energy monitoring system.

It measures ambient temperature and fan power usage using an ESP32 edge device. The system applies adaptive logic informed by real-time weather data to determine the appropriate fan operating mode. Telemetry data is streamed to a cloud database where it can be analyzed and visualized through a dashboard.

The dashboard shows how temperature, device behavior, and energy consumption are connected. It also estimates electricity cost and compares adaptive operation against a constant-power baseline to highlight potential energy savings.

How we built it

The system uses an ESP32 microcontroller as the edge device. It reads ambient temperature and fan power consumption and applies adaptive control logic to determine the operating mode of the fan.

The ESP32 streams telemetry data to Supabase, which stores time-series readings including temperature, fan mode, and power usage.

A Streamlit dashboard queries this database in real time and performs analytics such as energy integration, cost estimation, and comparison between adaptive operation and baseline scenarios. Weather information is retrieved through the Open-Meteo API, which allows the system to dynamically adjust the thresholds used for fan behavior decisions.

This architecture demonstrates a full IoT pipeline from sensor measurement to cloud storage and data visualization.

Challenges we ran into

One of the main challenges was integrating multiple layers of the system. The edge device, cloud backend, weather API, and dashboard all needed to communicate reliably.

Another challenge was implementing adaptive logic that changes dynamically with weather conditions while still producing stable and interpretable behavior. We also had to design the analytics layer to correctly compute energy usage over time rather than simply displaying instantaneous power values.

Accomplishments that we're proud of

We are proud that we successfully built a complete end-to-end IoT architecture.

The system collects real-time telemetry on an embedded device, streams it to the cloud, and visualizes meaningful insights through an interactive dashboard. We also implemented adaptive weather-informed logic and a baseline comparison model that highlights potential energy savings.

This allowed us to move beyond simple measurement and instead demonstrate how intelligent control strategies can influence energy efficiency.

What we learned

Through this project we gained experience building integrated IoT systems that combine embedded hardware, cloud databases, and data analytics.

We learned how to stream telemetry from microcontrollers, structure time-series data in a cloud backend, and build interactive dashboards that convert raw sensor data into useful insights. We also explored how environmental data can be incorporated into control logic to make devices behave more efficiently.

What's next for Smart Fan Energy Behavior Analyzer

Future versions of the project could expand from monitoring to direct device control. For example, the ESP32 could adjust fan speed through PWM or relay control instead of only computing operating modes.

We would also like to implement hysteresis-based control to stabilize switching behavior, aggregate historical data for longer-term analysis, and estimate the CO₂ emissions associated with energy usage.

Ultimately, the goal is to evolve the system into a more complete smart energy monitoring platform that helps users understand and optimize the energy behavior of everyday devices.