Real-Time Compression of Measurements in Distribution Grids

Real-time lossy compression of voltage and current measurements in an electrical grid is considered. It is assumed that a compressed version of measurements needs to be transmitted to the control center to be used for line voltage compensation. This necessitates tight delay and high accuracy requirements. Acquiring high compression, while also maintaining a high-quality reconstruction is achieved via differential quantization of consecutive samples. Zero delay is made possible using sample-by-sample processing of signals. The proposed scheme consists of two adaptive quantizers on transformed versions of voltage and current signals using Park's transform. The quantizers incorporate a predictor which is based on spectral characteristics of the transformed signals. The objective is to compress the measurements as efficiently as possible, while keeping the voltage within the stipulated values by grid standards. The suggested scheme is then applied in a model of a micro-grid with a voltage source converter based controller. The trade-off between compression ratio and quality of the grid voltage and current in terms of deviation from nominal values is studied and discussed.