# Grid forecasting

> Planning platform for an electricity distribution network, covering substations, feeders and switches across the grid.

**Sector:** Electricity distribution
**Year:** 2023
**Role:** Team of six, in a 15-person programme
**Stack:** Angular, C#, .NET, PostgreSQL
**What it does:** Demand 20 years out — against assets in the ground

## The problem

Planning infrastructure development, maintenance, and resource allocation for a power grid requires seeing years ahead of current demand, and a clear picture of the assets already in the ground. Without that, decisions on where to invest and when to maintain are made reactively rather than ahead of need.

## What we built

An internal system combining demand forecasting models over a 20-year horizon with asset management tooling, so infrastructure development, maintenance scheduling, and resource optimisation can be planned proactively rather than after the fact. The system provides visualisations and data analysis giving stakeholders a clear overview of current assets and projected needs.

## Outcome

- Planning moved off shared Excel files onto one platform, serving around 100 internal users: planners, management and technical staff.
- Covers hundreds of substations and more than a thousand feeders, switches and other assets.
- Runs many more forecast simulations than the previous process could, and produces more accurate ones. The forecast cycle is still measured in months.
- Twenty-year equipment-change predictions shown on a map, with approvals routed through the hierarchy by email notification.

## The long version

The core problem this system addresses is horizon: a power grid has to be planned years, not months, ahead, and the people making those calls need forecasting and current-state asset data in the same place rather than reconciled by hand across separate tools.

The forecasting side models demand over a 20-year window. The asset management side tracks what infrastructure exists now. Together they let planners see where the grid is headed and where it currently stands, and make maintenance scheduling and infrastructure investment decisions against that combined picture instead of guesswork or siloed spreadsheets.

Delivery included visualisation and data analysis layers on top of the underlying models, so the forecasts and asset data are usable by stakeholders directly rather than requiring a data team to interpret them on every request.


[Overcode](https://overcode.io/) — AI workflows and custom systems. This study as HTML: https://overcode.io/work/electric-forecast-distribution-system/