Carousel-shaped measurement platform for Arctic PV

<p class="p1"><span class="s1">Scientists in Finland have created a special platform to assess the performance of PV modules operating under Arctic conditions. Specifically, it provides data for efficient system design and performance modelling validation, with azimuth and tilt angle being key parameters.</span></p><p>Researchers from Finland’s <a href="https://www.pv-magazine.com/2023/08/30/scientists-analyze-seasonal-performance-of-rooftop-vertical-pv-in-the-arctic/" rel="noopener" target="_blank">University of Oulu</a> have conducted a two-year empirical evaluation of solar panels in different azimuths and tilt angles under Arctic conditions.</p>
<p>They took the measurements through a unique carousel-shaped infrastructure that records high-resolution, multi-azimuth data. Market metrics were also considered in the techno-economic analysis.</p>
<p>“This analysis assesses not only the generation potential of solar PV across different azimuths throughout a year, but also when that energy is generated and its value at various times of the day,” said the academics. “The results will also assist policymakers in making more informed decisions regarding solar PV requirements in northern regions, potentially influencing future policy directions.”</p>
<p>This solar carousel infrastructure consists of sixteen solar PV panels. Eight panels are placed at a 40° tilt angle, and the remaining eight are at a 90° tilt angle. In each tilt angle, the panels are oriented towards eight compass points, namely north, northeast, east, southeast, south, southwest, west, and northwest azimuths. The panels have a rated power output of 270 W.</p>
<p>Energy generation data were collected from September 2021 to August 2023 at 15-minute intervals. The value of electricity was based on the Finnish day-ahead electricity market prices.</p>
<p>The infrastructure is placed on a roof in Oulu, Finland, located in the transitional zone between the Arctic and subarctic regions. The solar irradiation in the region varies substantially by season, with horizontal average daily insolation ranging from 0.01 kWh/m2/day in December to 5.82 kWh/m2/day in June. The average global horizontal irradiance is around 2.4 kWh/m2/day, with an annual maximum of 863.83 kWh/m2/day. July is the hottest month with a mean temperature of 16.7 C, while February is the coldest month, with a mean temperature of -8.7 C.</p>
<figure class="wp-caption aligncenter" id="attachment_318004" style="width: 600px;"><img alt="" class="size-medium wp-image-318004" height="468" src="https://www.pv-magazine.com/wp-content/uploads/2025/09/1-s2.0-S0960148125021056-gr6_lrg-600x468.jpg" tabindex="0" width="600" /><figcaption class="wp-caption-text">Energy yield and losses <p><i>Image: University of Oulu, Renewable Energy, CC BY 4.0</i></p>
</figcaption></figure>
<p>Panels facing north, northwest, and northeast were found to generate 36–66% less energy and value of electricity than their counterparts facing southeast and south.</p>
<p>“Panels facing southeast and south consistently achieved the highest annual energy yields and value of electricity generated, with only a 1–3% variation in yearly output,&#8221; the researchers emphasized. &#8220;A clear seasonal and azimuthal trade-off was observed, with 40° panels facing south, southeast, and southwest outperforming 90° panels in summer. In winter, however, 90° panels, especially those facing south and southeast, generated up to six times more energy than 40° panels at the same azimuth due to minimal or no snow accumulation and improved albedo.”</p>
<p>The southeast 40° panel and the southwest 40° panel yielded a higher electricity generation value, despite their annual energy yield being lower than that of the south 40° panels. “This shows that, with increasing price volatility, time of use (ToU)-aware orientation can be more critical than maximising total yearly output. However, if the strategy is to maximise the annual generation, the traditional recommendation of south and southeast holds,” the academic added.</p>
<figure class="wp-caption aligncenter" id="attachment_318003" style="width: 600px;"><img alt="" class="size-medium wp-image-318003" height="318" src="https://www.pv-magazine.com/wp-content/uploads/2025/09/1-s2.0-S0960148125021056-gr2_lrg-600x318.jpg" tabindex="0" width="600" /><figcaption class="wp-caption-text">The solar carousel infrastructure <p><i>Image: University of Oulu, Renewable Energy, CC BY 4.0</i></p>
</figcaption></figure>
<p>The results were compared to those from simulations conducted using the Skelion plugin, Photovoltaic Geographical Information System (PVGIS), and the National Renewable Energy Laboratory’s (NREL) PVWatts.</p>
<p>All simulations showed good agreement for 40° panels, with a mean absolute percentage error (APE) of around 3%. However, in the case of the 90° panels, the error percentage was about 11%. “Simulation models overestimated performance for the majority of 40° panel azimuths and underestimated performance for seven out of eight 90° panel azimuths,” the team concluded.</p>
<p>The platform was presented in “<a href="https://www.sciencedirect.com/science/article/pii/S0960148125021056" rel="noopener" target="_blank">Optimising solar photovoltaic azimuth performance in Arctic conditions: An empirical techno-economic analysis</a>,” published in <em>Renewable Energy. </em></p>