Performance analysis of utility-scale solar PV power plant from SCADA Data

Abstract

The subject of this Master Thesis is to analyze the performance of utility scale solar power plants using SCADA data transmitted from the different components of a plant. This thesis as done while doing an internship at Innergex, a company that owns solar, wind and hydro parks. The main power plants on which the work was done are Phoebe, a 300 MW power plant being constructed in Texas, Stardale, a 27 MW in Ontario and Kokomo, a small 6 MW in Missouri. Phoebe is the first large scale solar power plant of Innergex, from which came the need to build a solar analysis team. The SCADA had to be built from scratch, and it was part of the work to find which parameters Innergex wanted to measure in order to be able to perform performance analysis. The data already available of Stardale and Kokomo, which don’t have yet performance analysis because of their small size, was used to develop the performance analysis methods with MATLAB. Through academic and industry literature reviews, various performance indicators have been found, such as Performance Ratio, Weather Corrected Performance Ratio, Effective Availability, Soiling Losses and Degradation. Methods such as linear regression were also used in order to restore lost energy due to unavailability of inverters. MATLAB and Excel were the main tools used for this work, as filtering the data was a big part of the work. This document presents graphs and tables presenting the application of these performance analysis tools.