PV system performance evaluation by clustering production data to normal and non-normal operation

The purpose of this study is the development of an algorithmic tool to cluster the points of a scatterplot of "System Yield vs. Reference Yield" of photovoltaic systems, which tends to follow a linear regression. From such clustering one can distinguish data points that comply with the linearity (inliers) and the data points which do not (outliers). The tool then is applied to the scatterplot for two main purposes: a) to detect and exclude any data input anomalies or b) to detect and separate measurements where the PV system is functioning properly from the measurements that show that the photovoltaic (PV) system is malfunctioning.