Solar Energy (photovoltaics)

Technical Symposium

Accurate Performance Prediction of Large Scale PV Power Plants with Complex Layouts

Thursday, September 27
11:20 AM - 11:40 AM
Location: 203AB

Abstract Content : Accurate performance prediction of large scale PV power plants with complex layouts has been a challenge for project developers, designers, and independent engineers, because complexities, such as uneven terrain, non-uniform shading, and variable layout pitch and row length, have made analysis of these sites difficult and uncertain. These uncertainties add considerable risk that can impact project financing. Common approaches to mitigating this risk, such as inducing backtracking by applying an artificial ground-coverage ratio, may reduce the amount of available potential energy collected at a site. Modeling simplifications, such as assuming flat terrain, applying a fixed shade impact factor, or only calculating string level shading, result in simulations that don’t reflect actual conditions and introduce significant errors [“Accurate Modeling of Partially Shaded PV Arrays,” B. Meyers, IEEE 44th PVSC, 2017]. To address these risks, DNV GL has developed an advanced model, that includes shading calculations at the submodule level, to simulate energy production on complex terrain with realistic PV system layouts. In analyzing projects with the advanced model, DNV GL estimates that project energy production forecasts could be improved by as much as 5-10% versus less complex models with simplifications and assumptions. In designing projects with the advanced model, DNV GL predicts 5-10% more energy with potential improvements in layout and operation. In this presentation DNV GL will provide guidance for developers and designers seeking to mitigate risks in simulation of sites with complex terrain and explain how to use simulations to improve site design to optimize solar energy output.


Mark Mikofski, PhD

Principal Engineer

Mark Mikofski is a solar energy analyst at DNV GL. Mark has worked in the field since 2007, predicting system performance and degradation. Before joining DNV GL in 2017, Mark worked at SunPower for 7 years. Prior to SunPower, Mark worked at the solar thermal startup Ausra which was acquired by AREVA in 2010. Mark earned his PhD in Mechanical Engineering from the University of California for his study of soot and carbon monoxide formation in under-ventilated fires. Mark is also active in the PV modeling community and has contributed to open source software to model electrical mismatch in PV systems.


Send Email for Mark Mikofski


Accurate Performance Prediction of Large Scale PV Power Plants with Complex Layouts

Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Accurate Performance Prediction of Large Scale PV Power Plants with Complex Layouts