Solar Forecasting and Integration in Power Systems

Mohana Alanazi, University of Denver


Renewable energy resources are becoming critical players in the electricity generation sector, primarily due to viability in combating global warming, effectiveness in reducing pollution caused by fossil fuel based generation, and diversifying energy mix to ensure energy security and sustainability. Solar energy is one of the most common types of renewable energy that has grown rapidly over the past decade and is anticipated to grow even faster in the future. Power supply from renewable resources is forecasted to surpass other types of generation in a foreseeable future. Numerous factors, including but not limited to the dropping cost of solar technology, environmental concerns, and the state and governmental incentives, have made the path for a rapid growth of solar generation. However, increased generation from renewable resources exposes the power system to more vulnerabilities, conceivably due to their variable generation, thus highlighting the importance of accurate forecasting methods. An accurate solar forecasting method, which takes into account generation variability and is able to identify associated uncertainty, can support a reliable and cost-effective deployment. More and more large-scale solar PV farms are expected to be integrated in the existing grids in the foreseeable future in compliance with the energy sector renewable portfolio standards (RPS) in different states and countries. The integration of large-scale solar PV into power systems, however, will necessitate a system upgrade by adding new generation units and transmission lines.

This dissertation proposes a forecasting model that aims to enhance the forecasting result and reduce errors. The proposed model utilizes a new approach to overcome some of significant challenges in solar generation forecasting. The model includes different data processing stages in order to ensure the quality of the data before it is fed to the forecasting tool. The model undergoes further enhancement such as forecasting methods combination and multilevel measurements application. Numerical simulations exhibit the merits of the proposed method through testing under different weather conditions and case studies. Moreover, a co-optimization generation and transmission planning model is proposed to maximize large-scale solar PV hosting capacity. The solution of this model further determines the optimal solar PV size and location, along with potential required PV energy curtailment. Numerical simulations study the proposed co-optimization planning problem with and without considering the solar PV integration and exhibit the effectiveness of the proposed model.