WattWise – Smart Energy Forecasting & Anomaly Detection

Inspiration

Buildings are responsible for about one-third of global energy use and emissions. In India, unpredictable surges in demand caused by heatwaves, festivals, and irregular usage patterns make energy management difficult. We wanted to create a solution that helps forecast demand accurately and detect anomalies early, enabling smarter and more sustainable energy use.

What it does

WattWise predicts short-term energy demand in residential and commercial buildings using pre-trained Time Series Foundation Models (TSFMs). It also detects anomalies in real-time, such as sudden spikes or drops, and provides insights that can help optimize energy consumption, reduce costs, and cut emissions.

How we built it

Collected and preprocessed building energy consumption datasets, cleaning missing values and normalizing inputs.
Applied IBM’s TSFMs for short-term load forecasting and fine-tuned them for different building types.
Implemented anomaly detection by flagging instances where the difference between actual and predicted consumption exceeded a threshold.
Built a dashboard to visualize forecasts, anomalies, and actionable recommendations for energy managers.

Challenges we ran into

Handling missing or irregular data points in the time series.
Calibrating anomaly thresholds to minimize false positives and negatives.
Understanding how TSFMs generalize across different building types.
Time limitations to build an end-to-end prototype during the hackathon.

Accomplishments that we're proud of

Successfully building a working prototype in a short timeframe.
Achieving reliable short-term energy forecasts using pre-trained models.
Designing a clear dashboard that makes the results actionable for non-technical users.

What we learned

The power of foundation models for time series forecasting.
Practical anomaly detection techniques using model deviations.
The importance of good data preprocessing in time series tasks.
How predictive modeling directly contributes to sustainability goals.

What's next for WattWise – Smart Energy Forecasting & Anomaly Detection

We plan to integrate renewable energy data (like solar and wind) into the forecasts, expand to large-scale grid-level predictions, and enhance the dashboard with optimization recommendations such as load shifting. In the long run, WattWise can evolve into a comprehensive AI-powered energy management assistant.