Solar Plant Handover: How to Hand Over a Plant Without Losing Its Data

Solar plant handover is where utility-scale projects quietly lose the data their owners will need for the next 25 years. I have asked this question to O&M teams at the start of new plant engagements probably 40 times now.

“Do you have the commissioning IV curve test results?”

The answers I get are usually some version of: “I think so, somewhere.” “The EPC contractor is supposed to provide those.” “We have some of them.” “What does that matter — the plant is operational.”

It matters because those IV curves are your as-commissioned performance baseline. Without them, you cannot distinguish normal first-year degradation from a systematic defect developing in a module batch. You cannot make the warranty claim. You cannot price the risk when a lender refinances.

And those IV curves are exactly the kind of data that disappears at handover.

What gets lost at solar plant handover

On EPC-to-O&M handovers for utility-scale projects, a meaningful share of project data ends up lost or inaccessible. Not “not digitised yet.” Actually gone, in a site container, in a former employee’s inbox, or on a USB stick that never made it into the data room.

Here is what disappears most often:

• Equipment serial numbers: Often missing for a large share of equipment items. They were written on handwritten site sheets that nobody digitised. Your asset register has a module count, not a module identity.
• As-built drawings: Many “as-built” drawings on utility-scale projects are red-lined design drawings, not drone- or survey-verified actual positions. The cable route that deviates 4 metres from the drawing because of a drainage channel is not captured until you excavate for a repair in year 8.
• Commissioning records: IV curves, insulation resistance measurements, and protection relay settings are frequently recorded on paper during commissioning and never transcribed.
• Warranty registrations: Module, inverter, and transformer warranties are too often not registered or are registered incorrectly at handover. You discover this when you try to make a claim.

Why handover data always deteriorates

The structural problem is timing and incentives.

EPC teams compile handover data under time pressure at the end of the project, when key personnel are already mentally on the next job, when subcontractors have demobilised, and when documents are scattered across a dozen tools and inboxes. Handover is one of the EPC phases most project teams don’t coordinate.

The O&M team that will operate the plant for 25 years was not in the room when most of the decisions were made. They receive a Dropbox link and a ZIP file and are expected to reconstruct operational intelligence from scratch.

The conventional fix, more thorough handover checklists and more contractual obligations on the EPC, addresses the symptom, not the cause. The cause is that handover is a manual compilation exercise performed by the wrong people at the wrong time.

For broader operational context, SolarPower Europe and the IEA PVPS O&M guidelines reinforce the same point: plant data only creates value when it survives the handoff into operations.

A solar plant handover framework that works

Let me give you the three-principle framework that eliminates handover data loss, because I have seen it work and the logic is straightforward.

Principle 1: Capture at source, not at handover

Every piece of operational data should enter a permanent, structured system at the moment it is created, not compiled into a data room six months later. Equipment serial numbers should be captured during delivery confirmation. IV curve test results should be uploaded from the commissioning tool, not transcribed from paper. Drone surveys should create permanent georeferenced records, not PDFs. For a worked example of capture-at-source on a live build, see this 40 MW EPC orchestration case study.

Principle 2: The handover is an agent-to-agent transfer, not a document dump

The SolarEPC Hub Orchestrator accumulates a living project model throughout the EPC lifecycle. At COD, it transfers that structured knowledge graph, not a folder of files, directly to the Asset Manager Intelligence agent.

The receiving agent does not get a ZIP file to interpret. It gets a fully structured digital twin: queryable, linked, and ready to answer operational questions from day one.

ClearSpot’s mesh combines autonomous AI agents with autonomous drones. Agents see what drones see, and drones execute what agents decide. That is how the drone surveys taken during construction become permanent O&M baseline geometry, not orthomosaic PDFs. They become live spatial data that the Asset Manager can query.

Principle 3: Drone as-builts, not red-line as-builts

Every construction milestone survey — earthworks, piling, racking, module installation, cabling — should generate a drone-verified record. Not only to check quality in the moment, though that is valuable, but to create a permanent, surveyor-accurate record of what was actually built and where.

When the O&M engineer needs to know the exact cable route through Zone 4B in year 12, the answer should be in the digital twin, not a mystery.

What the O&M team gets on day one

When a plant is handed over through ClearSpot’s agent mesh, the Asset Manager Intelligence agent receives:

• Complete equipment register with serial numbers, specifications, and warranty expiry dates for every device.
• Drone-verified as-built geometry at centimetre resolution, meaning actual positions, not design drawings.
• Full commissioning test records, including IV curves, insulation resistance, and protection relay settings, structured and queryable.
• Warranty registration status for all equipment, with renewal windows pre-calendared.
• Punch list closure history, including every item, its source, and its resolution.
• Performance baseline, meaning the generation and PR benchmarks against which future performance is measured.

That is the difference between spending your first six months reconstructing what you should have been given and spending your first six months actually operating the plant.

The warranty claim you did not know you had

Here is the concrete financial consequence of doing this right.

One of our Asset Manager Intelligence deployments identified, in the first 90 days of operation, that a module batch on a recently commissioned utility-scale plant showed systematic early-stage PID degradation. The IV curve baselines from commissioning, captured and preserved through the mesh handover, proved that the degradation pattern was post-delivery, not pre-delivery. The warranty claim, supported by drone thermal evidence and the baseline IV data, recovered the cost of the affected modules from the manufacturer.

That claim was possible because the data survived handover. Without it, the degradation would have been attributed to normal operation. The warranty window would have expired. The cost would have stayed with the owner.

For practical O&M benchmarks, the NLR Solar System Operations and Maintenance Analysis and the NREL best-practices reference provide useful context on why structured lifecycle data matters.

What this changes

The main shift is simple: handover stops being a document collection exercise and becomes a live transfer of operational memory.

That is why ClearSpot’s AI agents for solar O&M and solar plant performance monitoring layers matter once the plant is live. Connected to an asset management platform built for utility-scale portfolios, they keep the data model alive after COD instead of letting it decay into a file archive.

Closing thought

What is the most critical piece of data you have had to reconstruct from scratch after a solar plant handover, and what did it cost you to do that?

Frequently Asked Questions