The hybridization of existing wind power plants with PV energy can increase their market value by up to 5%, researchers from Portugal's National Laboratory of Energy and Geology (LNEG) have shown in new research. “This study focuses on the hybridization of existing wind power plants with different shares of solar photovoltaic capacity and investigates how these power plants can reduce their combined forecast errors and thus, increase profitability in electricity markets,” the scientists explained.
Talking with pv magazine, one of the scientists behind the research Dr. António Couto said that their “methodology was designed for real-time operation and was validated using historical real data allowing to compare our forecasts with observed results.” One major drawback, he added, is that the projections were based on only one statistical method. “In the future, we will include more methods to choose the best one for each situation,” he said.
The research team used data from three existing wind farms in Portugal – one in the south of the country, a second one in the north and a third one in the west – and assumed their hybridization with either 5 MW or 10 MW of PV power. In their calculations, the total power of the hybrid plant was 20 MW, meaning either 15 MW or 10 MW of wind energy.
Then, the scientists used technical and economic metrics for different hybrid project architectures and scenarios to analyze their impact power forecasts. “Although no consistent and significant differences were observed, the results suggest that using separate forecasts for each technology appears to be the most suitable approach,” the academics added.
This approach yielded the best market value and remuneration for a specific project located in northern Portugal, for which the market value of the wind facility increased by 5.31%. This project also saw its remuneration increase by 26.42% compared to a wind-only site. The project in western Portugal had an increase in market value by 2.24% and experienced a 21.18% higher remuneration. As for the project in the south, these values were at 3.6% and 30.63%, respectively.
“The work uses a forecast methodology based on a sequential forward feature selection algorithm which employs two different objective functions and an artificial neural network approach,” the researchers explained. “The methodology uses as input data from a numerical weather prediction model and iteratively selects meteorological features to achieve the different objective functions implemented, namely minimization of the root mean square error; or maximization of the market remuneration.”
The researchers have also found that, by utilizing the feature selection algorithm, the forecast parameters such as wind gusts, mean sea level pressure gradient, planetary boundary layer height, and wind shear are crucial for hybrid wind-solar projects.
“The findings emphasize the importance of properly selecting the meteorological features and objective functions to calibrate the forecast approach based on user requirements,” the academics concluded.
Their findings are available in the study “Wind power plants hybridized with solar power: A generation forecast perspective,” published in the Journal of Cleaner Production. “We are working on enhancing the methodology by incorporating more advanced statistical techniques comparing to the one presented in the article based on artificial neural networks,” Dr. Couto added. “Preliminary results indicate these research lines can, in several cases, reduce the forecast errors compared to the original methodology.”
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: firstname.lastname@example.org.