Why is accurate performance assessment of decentralized PV systems a must?

Assessing solar photovoltaic (PV) performance is crucial to ensure optimal plant operation and financial feasibility. This is relevant for any scale - on the wide spectrum of grid connected PV systems from residential size to solar fields. A whole field of applied mathematics has developed to address this issue, with multiple researchers engaging in model optimizations to find the best possible tools and models to assess PV performance. There is a variety of tools for the task - such as onsite weather sensors, satellites, regional weather stations and novel algorithmic solutions - all essentially utilized for calculating PV performance at various levels of accuracy.

Fig.1 The performance assessment map is based on the work by SunSpec Alliance and the San Jose State University (2014).

Basic-level performance assessment for Solar PV: At the basic level, solar PV performance can be assessed using commonly available tools. Manual review of inverter or utility energy meter readings and comparing them with historical monthly averages of the same system is one approach. Alternatively, using performance modeling tools such as PVSyst, PVSol, and PVWatts offers a better independent benchmark for monthly performance. These methods typically achieve 80-85% accuracy on a monthly resolution and about 90-95% accuracy on an annual resolution, sufficient for financial reports.

Another straightforward technique is peak system power measurement, a simplified way to measure the Power Performance Index (PPI). This method provides a fair estimate of system performance during the month with about 85% accuracy and can even be utilized daily, albeit with lower accuracy. Lastly, it is common to compare normalized energy output of a system with nearby PV systems - which can bring about 80-90% accuracy on monthly scale. These methods are typical for residential and small commercial segments due to their low cost and simplicity, making them adequate for basic financial reports but not for professional Operation and Maintenance (O&M) requirements.

Advanced-level PV performance assessment can be achieved with independent tools as benchmarks, including onsite weather sensors, satellites, and regional weather sensors. These tools calculate performance indicators such as Performance Ratio (PR), Power Performance Index (PPI), and Energy Performance Index (EPI) with varying accuracy levels. High-tier onsite weather sensors enable 95% accuracy on daily resolution (with temperature correction), while lower-tier sensors and satellites offer around 85-90% daily accuracy. Regional weather stations may be less accurate but still provide daily resolution necessary for professional operation & maintenance (O&M). Incorporating these advanced tools and real-time weather data into energy management systems is crucial. Accurate PV performance assessment ensures that energy production forecasts are reliable, which in turn supports effective decision-making for grid management and energy distribution. By leveraging advanced PV performance assessment tools, energy management systems can better predict and manage solar energy contributions, ensuring consistent and dependable energy production.

Expert-level PV performance assessment involves state-of-the-art hardware for weather sensing combined with professional software tools, utilizing proprietary algorithms and often artificial intelligence. This hardware is meticulously calibrated and corrected with software in real-time to achieve 95-98% accuracy on daily resolution. These advanced tools are highly sensitive, enabling real-time detection of minor anomalies in solar facilities. However, this high-tier solution is costly, requiring both expensive hardware and professional software. As a result, we see such solutions mainly within the utility-scale solar PV segment.

Novel methods for high accuracy performance measurement are also being developed. Soltell is excited to contribute to the global effort of optimizing solar PV energy finance, operations, and integration with its Sensorless technology for measuring PV performance. This proprietary technology allows for quantitative performance assessment on any PV plant without relying on external weather sensors. Key metrics such as Energy Performance Index (EPI), Power Performance Index (PPI), Soiling Ratio (SR), and Availability can be obtained with high accuracy.

The proprietary "Intactness" index offers a superior quantitative approach to assessing PV performance. It combines technical assessment for components of a photovoltaic system with quantitative performance indicators using proprietary algorithms. This method is significantly more accurate than monthly performance models based on seasonal averages, which can have a monthly deviation of ±20% due to weather variations and model inaccuracies. The Intactness index is also superior to other performance indexes used across the decentralized energy resources (DERs) industry.

Fig.2 Energy Performance Index via Sensorless technology detecting chemical staining, which can't be identified with hardware sensors. Measurement of a 150kW rooftop PV system in an industrial area, during and after a chemical staining event affecting the modules.

In summary, Soltell's Sensorless technology enables the cost-effective application of proprietary algorithms to residential and commercial PV plants, achieving an uncertainty level of 2-6% on daily resolution. This results in a performance assessment accuracy level of 94-98%, enabling superb industrial automation with priority-enabling ratings and predictive maintenance capabilitie comparable to high-tier automation tools.

Interested in Sensorless technology for measuring PV performance and integrating it into an energy management software or a service application? Fill-in the demo request to get more details.