Analysing smart meter data to uncover energy consumption patterns
summary
Uses a representative sample of 13,000 households with half‑hourly gas and electricity from 2019, engineers features (proportion overnight, seasonal curves), reduces dimensionality with PCA and compares models before settling on KMeans clustering (also tried GMM and HDBSCAN), produces ten energy use profiles (one has ~60% EVs, two are electricity‑only), runs entirely in a secure data lab, exposes limitations from small sample size and missing contextual fields.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 4:42Project overview
- 7:12Secure lab with synthetic data
- 9:00Methodology to clustering models
- 11:27Feature engineering
- 13:19Energy use profiles
- 15:50Policy visualizations
- 19:48Limitations and challenges
- 21:53Energy transition applications
- 24:25Future work
- 25:07Acknowledgments
- 27:22Resources from Nesta
related talks
22mSustainable Analytics
40mHow to Use Python to Better Understand the Impact of Electricity Pricing on Consumption
38mTransforming Data to Unlock Its Latent Value
31mBuilding an electricity market model from scratch
36mExtracting Relevant Metrics with Spectral Clustering
33mMapping the local heat transition
32mData Transformation: A Framework for Exploratory Data Analysis
37mLocation Analytics For Smart Grid Reliability