There is a selection of mathematical methods to assess the performance of solar PV facilitiies for financial and operational purposes. In line with the more familiar performance indicators such as Performance Ratio (PR) and the modeled Expected Yield (EY) one should also acknowledge the importance of Technical Availability (Technical PV Availability) and Contractual Availability (Contractual PV Availability) for those goals. In fact, availability indicators can be more accurate to determine PV system performance and allow comparison of PV systems on an identical scale. However, calculating availability indicators is a tricky business and companies must be aware of this complexity and technical demands to make this work. Here we are going to dive into advantages of the availability performance indicators, in line with explanation on the accompanying technical complications.
First of all, let's talk about the above mentioned Performance Ratio and modeled Expected Yield - the generally accepted performance indicators in the solar PV business. The hardware-dependent Performance Ratio is more suitable for the utility and industrial-scale PV segments, while modeled Expected Yield is relevant to the small and medium commercial scales. PR assessment requires a calibrated weather station on the site and a reasonable performance model to give an accuracy of 10-15% on a daily timespan and 5-10% on a monthly timespan. A more accurate version of the PR - the Temperature-adjusted Performance Ratio (T-adjusted PR) can do better, enabling performance assessment accuracy of 5-10% on a daily timespan and 2-5% on a monthly timespan. Modeled Expected Yield is relying on a performance model coupled with historical weather measurement averages. Expected Yield method can't give a reasonable result on a daily timespan due to weather variability, but does produce performance assessment accuracy of 15-20% on a monthly timespan. PV systems can't be compared directly in terms of PR indicators (though possible via PR ratio), but can be in terms of modeled Expected Yield performance indicators.
Performance assessment is typically considered irrelevant to residential scale PV due to associated hardware and modeling costs, though some innovative companies do offer PV performance assessment via remote sensing combined with low-cost performance models. The accuracy of such performance assessment applied on residential PV is typically mediocre, but can provide reasonable accuracy values on a daily timespan and may reach an accuracy of 10-15% on a monthly timespan. Without better alternatives in terms of cost-efficiency - this is certainly something to consider, but there is much room for improvement.
The PV market is hence quite segmented in terms of performance indicator utilization. The multiple standards for Performance Ratios, inaccurate Expected Yield-based assessments and the general lack of proper performance indicators for the residential scale PV imply of a need to find a common ground for the entire sector. Interestingly, such an option exists and is utilized by some players in the industry - those are PV system availability indicators, which can be equally applied to all scales and be calculated through several methods. Moreover, Solar Power Europe, the most notable European professional solar association does recommend utilizing availability indicators rather than the less accurate modeled Expected Yield indicators.
Now, let's define what is PV system availability - according to IEEE standard 762-2006 it is the fraction of time in which a unit is capable of providing service and accounts for outage frequency and duration, while according toInverter mfg. 2010, it is the probability that a system will be operable when called upon. Obviously, PV system availability is relevant upon solar irradiance hours and hence is inherently energy input adjusted during the day. However, when looking at the longer time-scale one can define a time-adjusted system availability (T-adjusted Uptime) - when each day is weighted equally during a month or a year. Alternatively, an energy-adjusted system availability (E-adjusted Uptime) when each day is weighted proportionally during a month or a year. Though the latter does certainly require a weather station on the site, the first option of time-adjusted availability can be reasonably calculated for all scales regardless hardware.
Despite the promise of applying T-adjusted Uptime to all solar PV scales to achieve accurate and comparable performance assessment, this performance indicator does typically require manual corrections on daily basis per system in order to adjust it to performance monitoring results. This is due to the fact that contractors do utilize Contractual Availability rather than the more broadly defined Technical Availability. Contractual Availability rightfully excludes grid blackouts, as well as Force majeure events from contractor's responsibility, but the contractor is required to properly record and calculate the impact of those occurrences to make the contract work. It is typical to commit on a Contractual Availability of 97-99%, with 98% monthly or annual PV system availability being the most common use case.
To summarize, availability indicators are an interesting alternative to assess PV performance all across the solar scale and type spectrum. While some definitions of availability do require an on-site weather station, the T-adjusted Uptime method can reasonably be applied without it and allow a cross-scale and cross-type comparison of PV performances. In theory, remote sensing or a similar weather data sourcing method can also be applied to achieve the more accurate E-adjusted Uptime. In conculsion, asset management software providers should consider embedding availability indicators to achieve accurate financial reporting on monthly and quarterly timespans, while PV contractors should consider availability indicators to be included in contract commitments instead of Performance Ratio (PR) or modeled Expected Yield (EY). However, T-adjusted Uptime and E-adjusted Uptime indicators require either industrial automation or proper weather data sources to make this happen.
Interested to know more on cost-effective technological solutions to accurately assess residential, commercial and industrial PV system performance indicators? Fill-in the demo request form to try our solutions or email us to get details.