| Work Detail |
Researchers have developed a three-step technique to estimate power generation losses in floating photovoltaic plants due to tilt changes and misalignment losses. The new module was tested in an experimental setup and demonstrated that power losses depend primarily on both the system design and its location. A group of scientists led by the Norwegian Institute of Energy Technology (IFE) has developed a novel wave-induced loss (WIL) model for floating photovoltaic (FPV) installations. The model consists of three main steps, namely wave-structure interaction, irradiance, and electrical modeling. In floating photovoltaic systems, waves, if present, cause movements in the floating structures on which the photovoltaic panels are mounted. First, the movement will deflect the panels from their fixed orientation. Second, if the panels are mounted so they can move independently, the different irradiance conditions at each panel in the string cause a mismatch loss, the researchers explain. In this work, we propose a method to model the total loss of these two combined components, and we call it WIL. To create the model, the team used in-house software 3DFloat to simulate the motion of the photovoltaic panels and Python libraries for irradiance and electrical modeling. The first step, wave-structure interaction, analyzes how the FPV moves due to waves. It uses a standard ocean wave model, assuming the photovoltaic panels accurately follow the local wave motions without moving horizontally. The second part, which models irradiance, uses the panel tilt angle obtained from the previous stage. From there, the third stage of electrical modeling uses the PVMismatch simulation library to calculate electrical output. The WIL is then calculated as the percentage difference between that result and the simulated result of a static setup. Step 2 of irradiance modeling was tested and verified using an experimental model. This experiment was conducted on an FPV prototype provided by Sunlit Sea and located in the Oslo Fjord. It consisted of two arrays with four photovoltaic panels each. Four irradiance sensors, an accelerometer, and a 3D gyroscope were attached to the edge of one of the photovoltaic panels for data collection. The analysis showed that the modeled irradiance matches the measured irradiance well and provides acceptable agreement with the measurements, with a root mean square error (RMSE) between the modeled and measured irradiance of 2.8 W/m2. However, the model tends to slightly underestimate the extreme values ??of the measured irradiance, the scientists emphasize. Following successful validation, the group demonstrated the model with several scenarios. Among the changing parameters tested were the locations (0?N, 0?E) and (50?N, 0?E), with waves of 0.25 m, 0.5 m, 1 m, and 3 m. Waves were tested with a direction of 0°, traveling along the string; 90°, traveling across the PV string; and 45°, approaching the PV string at a diagonal angle. “Higher WIL was observed for systems further from the optimal orientation and under conditions with steeper and higher waves. WIL also increased with increasing PV string length and when the wave direction changed from being orthogonal to the string length to being parallel to it,” the results showed. “For example, WIL ranged from 3.3% for a significant wave height of 0.25 m to 6.7% for a significant wave height of 1 m. This indicates that WIL can be significant and should be taken into account for energy performance analysis.” The novel technique was presented in “ Modeling wave - induced losses for floating photovoltaics: Impact of design parameters and environmental conditions,” published in Solar Energy . Researchers from Norway’s IFE, the University of Oslo (UiO), and the University of Himachal Pradesh in India participated in the study. |
