01
Collect the operating signals
Bring solar generation, demand, battery state, and distributed-resource data into a usable operating view.
Independent, source-backed renewable-energy case study · Solar-grid operations
i GRID LAB: AI control for surplus solar power.
i GRID LAB built a cloud platform that predicts surplus solar generation and creates control scenarios for batteries, electric vehicles, and other distributed energy resources.
Starting point
i GRID LAB wanted to increase renewable-energy use by forecasting local solar surpluses and coordinating how distributed assets respond.
Solution implemented
The company built the R.E.A.L. New Energy Platform on AWS to collect generation, demand, battery, and EV data, predict excess solar power, and create automatic device-control scenarios.
Reported result
AWS reports that the proof of concept was built in six months, produced its first demonstration results one month after launch, and was later adapted for around 500 solar facilities.
How it worked
i GRID LAB connected prediction to device control.
The platform collected operating data from distributed equipment, combined it with external information, predicted surplus generation, and calculated charging and discharging scenarios.
Headline result
6 months
to build the proof of concept, reported by AWS
02
Forecast the surplus
Estimate when and where solar production is likely to exceed local demand or network capacity.
03
Calculate the response
Compare charging, discharging, curtailment, and other control options against the operating constraints.
04
Apply and monitor
Send the approved control scenario to connected resources and keep the outcome visible to operators.
Reported business impact
The project moved from development to field evidence quickly.
AWS's customer story describes both the delivery timeline and the scale reached after the initial proof of concept.
01
Six-month proof of concept
The initial platform was built in six months from the start of the project.
02
Results after one month
AWS reports that the demonstration produced specific findings one month after it began.
03
Around 500 solar facilities
The i GRID Group later began adapting the platform for approximately 500 solar facilities.
What this case shows
The evidence is specific.
These lessons come from the implementation and reported results above. The source and attribution are provided below.
- 01 Design the action path at the same time as the forecasting model.
- 02 Keep equipment limits, operator approval, and fail-safe behaviour explicit.
- 03 Measure forecast quality together with device behaviour and grid outcomes.
- 04 Use a bounded proof of concept to identify operating effects and problems before wider deployment.
Source and attribution
Where these claims come from.
Independent industry case study summarised by Remova Tech from AWS's published i GRID LAB customer story. Remova Tech did not deliver this project; all timelines, scale figures, and capabilities are attributed to the original source.
Verify the reported results in the original sourceHave a similar challenge?
Let’s discuss the actual project.
Show us the current process, systems, data, and target result. We will help you determine what can be built and how success should be measured.