Outliers' detection and filling algorithms for smart metering centers

This paper summarizes diverse outliers' detection and filling algorithms applied to metering centers of smart grids. The paper first shows typical problems in data measuring, especially considering smart metering infrastructure. The objective is to discuss and test detection and filling algorithms in several load time series of a real distribution utility. The impact of these algorithms is evaluated under a data mining approach, and results demonstrate the benefits that can be obtained in other applications like, load forecasting.