Article - Greenbyte

January 6, 2021 — Product News

‘Potential Power Learned’ - a crucial feature to assess lost production in inverters

Utilizing historical data from your assets in combination with machine learning, a novel and more accurate approach to assessing the performance of your sites is now available. Potential Power Learned enables an individual model for each inverter and outperforms traditional potential power calculations.

Johanna Renman
Data Scientist, Greenbyte

Incorrect assessment of the potential power of your inverters can hide important information regarding the performance of your sites. This could, in turn, mean that you are missing out on potential opportunities to increase production.

The way forward is data driven

Working towards reduced downtime and increased revenue requires informed decision-making as even the smallest issue can have a large impact. Potential Power Learned acknowledges this by providing an accurate estimate of the potential power of your assets, which is key when calculating important KPI:s like Performance Index (PI) and Lost Production. Utilizing historic data from your assets, the algorithm behind Potential Power Learned uses machine learning to model the power produced. The model learns the best performance of the individual inverters, considering the resource and environmental conditions on the site. The result is a potential power tailored to each individual inverter, that outperforms traditional potential power calculations in accuracy.

What is wrong with my current estimation calculations?

Unlike wind turbines, where guaranteed power curves are provided by the manufacturers, power from solar panels is estimated using standard equations. These equations use metadata like installed DC capacity and panel specific details like power temperature coefficient, etc. These equations are a best estimate of the behavior of panels placed in standard conditions and fail to take into account site-specific environmental conditions that have a considerable impact on production.

Potential Power Learned is based on the historical behaviour of your assets as well as the resources and environmental conditions specific to your sites, creating a level of accuracy that is simply not possible with the generic standard equations.

Is it for me?

Greenbyte Platform is designed to help you analyze production losses, pinpoint issues, and reveal hidden opportunities for cost savings and optimization. Providing you and your team the ability to focus on performance analysis rather than data collection and lengthy processing.

An accurate estimate of possible production from your inverters is a prerequisite to reveal actionable insights, which is where Potential Power Learned becomes crucial. Furthermore, it does not require any metadata, which means as long as there is 1 year of historical data, it is plug-and-play in Greenbyte platform.

Interested to see how Potential Power Learned could possibly leverage your solar pv asset management? Contact your Customer Success Manager!

Get the latest news in your inbox

Save the date! Greenbyte Forum 2019 is coming!

Hear from and get inspired by leaders in our industry, attend educational breakout sessions, and get up to speed on everything Greenbyte while meeting your brilliant peers who, like yourself, leverage data analytics to drive innovation and business growth.

Nov 26-27th
Gothenburg, Sweden

Learn more