PM.
Energy — Case study blueprint

UK Energy Intelligence

A decision-focused view of electricity demand, generation mix and carbon intensity across Great Britain.

APIPythonSQLPower BIPower QueryDAX
Business problem

Energy-intensive activity is often scheduled without a clear view of when grid electricity is cleanest or when demand pressure is highest.

Objective

Create a repeatable pipeline and an interactive dashboard that makes half-hourly carbon and generation data useful for operational planning.

Decision

Shift flexible loads into forecast low-intensity periods.

01

Data source

Official Great Britain carbon-intensity forecasts, estimated intensity and generation mix.

Visit NESO Carbon Intensity API ↗
02

Data preparation

  • Request national and regional endpoints at half-hour granularity.
  • Normalise nested JSON, standardise UTC timestamps and retain forecast revisions.
  • Validate missing periods and load curated facts into SQL.
03

Data model

  • FactCarbonIntensity at settlement-period grain
  • FactGenerationMix by fuel type and period
  • DimDate, DimTime, DimRegion and DimFuel
04

Analysis

  • Time-of-day and day-of-week intensity profiles
  • Forecast error and high-emission event detection
  • Generation-mix contribution and demand correlation
05

Dashboard

Embed-ready dashboard area

Designed for Power BI, Tableau, Looker Studio or Plotly.

The live report will sit here after publication. Until then, the portfolio keeps a clear, truthful build status.

Discuss the project ↗
06

Key questions & KPIs

  • Identify recurring low-carbon windows by weekday and settlement period.
  • Compare forecast and estimated intensity to understand planning confidence.
  • Separate changes driven by demand from changes driven by generation mix.
Carbon intensity (gCO₂/kWh)Low-carbon generation sharePeak demand windowForecast variance
07

Business recommendations

  • Shift flexible loads into forecast low-intensity periods.
  • Create alerts when intensity moves above an agreed operating threshold.
  • Review forecasting error monthly before automating operational decisions.
08

Limitations

  • The API reports generation-related emissions rather than full lifecycle emissions.
  • Forecasts change as real-time conditions and generation availability change.
09

Next steps

  • Build the ingestion notebook and SQL schema.
  • Publish the first Power BI report with a 30-day refresh window.
  • Test postcode-level recommendations against the national view.

From blueprint to published analysis.

Build the pipeline, validate the model, document the findings, then add the live dashboard and code.

Explore another project →