Abstract
The expansion of renewable energy sources, battery energy storage systems, electric vehicles charging stations, power electronics interfaced loads and the integration of microgrids operating in connected or island modes have caused voltage and current signals to be distorted from their conventional sinusoidal model [1]. The transient phenomena over short timeframes (less than 1s) cannot be detected by conventional means of measurement devices, which focuses more on information concentrators of electric quantities, rather than analyzing the entire variation of the signal. Therefore, new methods of measuring and modelling energy transfer are needed to understand these phenomena in detail and to adapt electric networks to the new requirements of the energy transition. This paper proposes a new method for monitoring the deviation of the current signal from the sinusoidal model at low voltage distribution level through sampled values [2]. A system composed of a sampling unit and an edge computing capacity is proposed to receive the aggregated current signal from multiple loads and analyze its deviation from the sinusoidal model over a predefined timeframe. The deviation from the sinusoidal model is quantified using two statistical indicators, Root Mean Square Deviation (RMSD) and Goodness of Fit (GoF), which provide a different view about the deviation of the signal. This way, the Distribution System Operator (DSO) can receive aggregated information about the degree of distortion of current signals in a system node.