Machine Learning for Smart Meter Data

A Structured Path from Meter Signal to Business Insight

A structured, project-based path through smart meter data: disaggregation, forecasting, anomaly detection, and grid-edge value, from raw signal to business insight.

A practical book for energy analytics and ML practitioners

Machine Learning for Smart Meter Data

From raw meter signal to business decision: disaggregation, forecasting, and grid-edge value, built to be run, not just read.

By Anthony Faustine    

Read the preface View on GitHub

Meter Signal

Data & ML

Business Value

Meter Signal

→ Disaggregation →

Forecasting

→

Grid-Edge Value

Your learning path

Part 1 · Foundations

Smart meter data fundamentals, signal processing, and the data pipelines every later chapter builds on.

• 01Reading a meter signalSoon

Part 2 · Disaggregation

Non-intrusive load monitoring: from the aggregate signal to appliance-level insight.

• 01What Is Non-Intrusive Load Monitoring

Part 3 · Forecasting

Probabilistic load and PV forecasting, built on Twiga: point forecasts and calibrated uncertainty.

• 03Forecasting with uncertaintySoon

Part 4 · Grid-Edge Value

Clustering, anomaly and theft detection, and LV hosting-capacity questions as DER growth reshapes the grid.

• 04Value at the grid edgeSoon

Chapters are being written part by part. Follow progress on GitHub.