Stranded Power Finder
Matching underused electricity with computational loads that can travel to it, turning stranded generation into usable capacity.
- Program
- Research
- Started
- Aug 2026
- Field test
- Not yet
- Commercial
- No
01 Problem
Power is wasted where loads cannot see it.
Curtailed renewables, weak-grid sites and behind-the-meter generation waste power that productive loads elsewhere would pay for.
The waste comes from matching, not technology: the electricity and the loads that would pay for it never learn of each other.
02 Why existing approaches fail
The two halves never meet.
Neither the market, the loads, nor the data currently connect stranded supply to mobile demand.
- Market structure
Power markets clear on the grid's terms, not the load's.
- Load-side blindness
The loads that can relocate, such as compute, cold storage, pumping and processing, rarely have visibility into where stranded power exists.
- Fragmented data
The underlying data is scattered across operators, regulatory filings and market reports, with no single place to ask the question.
03 System
From public data to candidate pairings.
The model works from public data toward specific pairings: sites screened for stranded power, loads screened for mobility, and the economics of putting the two together.
- 01Generation and curtailment data
- 02Site screening model
- 03Load matching: compute, storage, processing
- 04Site economics
- 05Candidate pairings
04 What exists today
An early model on desk research.
What exists today is desk research and an early screening model, together with data-source mapping across public market and curtailment records.
No screening has been published and no pairing has been made.
05 Evidence
What we can show.
No published evidence yet. This section fills as experiments produce results worth showing.
06 Status
The program, stated plainly.
This table is the actual state of the program. It changes when the state changes, not before.
- Status
- Research
- Started
- Aug 2026
- Field test
- Not yet
- Commercial
- No
- Open components
- No
07 What comes next
One region, screened and published.
The next milestone is a published screening of one region, with the model's candidate sites and their economics.
08 Work with us
Sitting on underused power, or operating loads that could move to it?
Describe the site or the load. The interesting projects start where the two halves have never been introduced.