PV metric

Solar Performance Ratio (PR): formula, averages and what is good

The headline efficiency number for a PV plant, independent of weather.

Quick answer

Performance Ratio (PR) is the ratio of a PV plant’s actual energy yield to the yield it would have produced at its nameplate efficiency under the irradiation it actually received. It normalises out the weather, so it is the single number that says how well the plant is converting available sunlight. Average PR for modern plants is roughly 80 to 90 percent (Fraunhofer ISE); measured weather-adjusted fleet averages in the US run 91 to 94 percent, while unmanaged distributed fleets can sit near 79 percent.

Definition

PR divides measured AC energy by the theoretical energy from the plane-of-array irradiation at STC efficiency. Because it cancels the irradiance the plant received, a falling PR points to losses inside the plant — soiling, degradation, downtime, clipping, thermal losses — rather than to a cloudy month. It is the EN 61724 benchmark metric for plant health.

Formula

PR = actual energy yield ÷ (POA irradiation × nameplate DC ÷ STC irradiance)

Typical range

Well-run utility-scale plants: PR ≈ 0.80–0.85. New plants can exceed 0.85 in mild climates; hot desert sites sit lower because of temperature losses. A PR below ~0.75, or a downward trend, signals recoverable loss.

Why it matters

PR is the term in nearly every O&M contract and performance guarantee, and the metric lenders track. A 2-point PR slip on a large plant is six figures a year in lost generation. Because it strips out weather, a falling PR is the earliest honest signal that something inside the plant — not the sky — is costing yield.

How NuraVolt tracks it

NuraVolt computes PR continuously from POA irradiance and AC output, decomposes the gap to nameplate into named loss buckets (soiling, temperature, availability, clipping, degradation), and trends each — so a PR decline is attributed to a cause and a euro figure, not just flagged.

Average performance ratio: published benchmarks

There is no single "average PR" because studies measure different flavours of the metric: raw PR includes temperature losses, weather-adjusted PR corrects them out, which is why fleet studies report higher numbers than raw rules of thumb. The table below only contains published, sourced figures.

Published PR benchmarks. PR definition varies by study (raw vs weather-adjusted); each row cites its source.
Fleet / eraPerformance ratioSource
Modern plants, industry-wide rule of thumbAbout 80 to 90 percent (raw)Fraunhofer ISE, Photovoltaics Report
Plants built before 2000About 70 percentFraunhofer ISE, Photovoltaics Report
US fleet study, 250 systems, 157 MW93.5 percent average (adjusted)Deline et al. 2020, via US DOE performance report 2022
California fleet, 2,200 reporting systems91.7 percent weather-adjusted averageWalker et al. 2019, via US DOE performance report 2022
Germany, 100 systems, 201070 to 90 percent, median 84 percentReich et al. 2012, Progress in Photovoltaics
US federal buildings, 75 mostly unmanaged systemsAbout 79 percent average when runningUS DOE performance report 2022
1980s-era systemsAround 70 percentIEA-PVPS Task 13 long-term performance report

Two more published numbers frame what "good" means. NREL’s O&M best-practices report (3rd edition) estimates that comprehensive O&M could lift the average age-and-temperature-adjusted PR of US systems from 91.7 percent to at least 95 percent, which is the recoverable-loss argument in one sentence. And kWh Analytics’ Solar Risk Assessment 2025 found US PV sites underperforming their P50 production estimates by 8.6 percent on average across roughly 34,000 system-months, meaning the typical plant has more recoverable loss than its owner assumes. Module degradation compounds this slowly: NREL’s Jordan and Kurtz compendium puts median crystalline-silicon degradation at 0.5 to 0.6 percent per year, with the mean nearer 0.8 percent.

PR in contracts: guarantees for IPPs and utilities

PR guarantees used to be the standard performance clause in EPC and O&M contracts, and they still appear, most often where the EPC and the O&M provider are the same company. SolarPower Europe’s O&M Best Practice Guidelines now recommend availability and response-time guarantees instead, citing a minimum guaranteed contractual availability of 98 percent over a year as best practice, because an O&M contractor controls uptime but not irradiance sensors, soiling regimes, or degradation. For an IPP the practical consequence cuts both ways: if your contract carries a PR guarantee, the metering and sensor basis of the PR calculation decides disputes, and if it carries an availability guarantee instead, PR becomes your own internal health metric rather than the contractor’s liability, which makes independent PR tracking more important, not less.

Methodology & sources: IEC 61724-1 performance monitoring · Fraunhofer ISE, Photovoltaics Report · US DOE / NREL, Understanding Solar Photovoltaic System Performance, February 2022 · NREL, Best Practices for Operation and Maintenance of Photovoltaic and Energy Storage Systems, 3rd edition (NREL/TP-7A40-73822) · IEA-PVPS, Analysis of Long-Term Performance of PV Systems, T13-05:2014 · Jordan and Kurtz, Compendium of Photovoltaic Degradation Rates, NREL 2016 · kWh Analytics, Solar Risk Assessment 2025 · SolarPower Europe, O&M Best Practice Guidelines v5.0 · nuravolt/digitaltwin/hybrid_model.py

Frequently asked questions

See also

See this on your own plants

NuraVolt turns your SCADA and BMS data into early fault detection, degradation-aware BESS analytics, and audit-ready reporting. A fixed-scope audit shows you what we’d find on your portfolio.