Identification of Inoperative Skid Prevents 81 MWh in Projected Solar Losses

Context

Have you ever wondered if planned maintenance activities are truly successful?

In a photovoltaic plant operated by one of our clients, a severe availability anomaly began on May 5, 2026. Starting at 3:50 PM, the “Abnormal grid voltage” alarm was systematically triggered, simultaneously compromising three inverters associated with the same conversion skid.

The productive impact was drastic. The Performance Ratio (PR) index of the affected block, previously operating within normal standards, suffered an abrupt reduction, recording continuous negative deviations (around -30.61 points) compared to the expected average for the plant. In practice, this meant that the three pieces of equipment completely stopped injecting energy into the grid.

The major aggravation in the scenario lay in an invisible monitoring point: the local Operation and Maintenance (O&M) team was unaware that the skid was inoperative, allowing the fault to remain active and invisible to the daily routine of the park.

Implemented Solution

The ability to reveal unmapped losses relied on the platform's data intelligence and the proactivity of the Performance Engineering team, following the steps below:

• Performance Ratio (PR) Monitoring: Through the daily PR tracking charts, the platform highlighted the immediate detachment of the skid from the plant's average shortly after May 5.
• Alarm Timeline: The tool isolated the root cause by demonstrating that, in the exact minute of the performance reduction, all three inverters registered the uninterrupted "Abnormal grid voltage" alert.
• Prescriptive Action: With the diagnosis of the shutdown and the quantification of losses in hand, the case was reported in a structured manner to the client's team during the periodic Performance Engineering meeting, held on May 11, 2026.

Results Achieved

The provided report transformed an invisible operational point into an immediate and highly targeted corrective field action:

1. Just one day after the report, on May 12, 2026, the O&M team went into the field, located the affected skid, and corrected the problem, resuming normal equipment operation.
2. During the exact 7 days the fault silently persisted (from 05/05 to 12/05), the skid failed to produce 5.53 MWh.
3. Given that the local team was unaware of the shutdown, the fault had the potential to drag on until the next monthly or quarterly physical inspection. The platform calculated that, if the skid remained inoperative for 3 months, the client would have accumulated a loss of 81 MWh, a loss that was completely avoided by the analytical action.

Conclusion

This case reinforces the premise that the operational availability of equipment is not always clearly reflected in local control centers. The shutdown of three inverters, masked by the O&M team's lack of visibility, perfectly illustrates the “invisible losses” in solar generation. By acting as an intelligent monitoring layer, crossing Performance Ratio reductions with the timeline of electrical alarms, the system ensured not only the identification of the error but also the proof of return on investment by protecting the asset from a loss of tens of Megawatt-hours over the months.

Key Data

• 81 MWh of projected and avoided loss (if the 3-month fault had consolidated);
• 5.53 MWh lost in just 7 days of silent unavailability;
• 3 inverters inoperative on the same skid due to grid voltage failure;
• Immediate intervention: Operation reestablished just 1 day after the technical alarm, following the alert from the Performance Engineering Team.

‍