Automatic Identification of Different Types of Consumer Configurations by Using Harmonic Current Measurements

Power quality (PQ) is an increasing concern in the distribution networks of modern industrialized countries. The PQ monitoring activities system operators (DSO), and consequently amount measurement data, continuously increase, new automated tools are required for efficient analysis. Time characteristics parameters (e.g., harmonics) usually show characteristic daily weekly cycles, mainly caused by usage behaviour electric devices. In this paper, methods proposed classification harmonic emission profiles typical consumer configurations public low voltage (LV) using a binary decision tree combination with support vector machines. performance was evaluated based on 40 different sites German LV grids. This method can network identification early detection fundamental changes behaviour. enables, example, assistance resolving customer complaints or supporting planning managing levels profiles.