LAUSR

Abstract The complex random characteristics in smart grid lead to inaccuracy of smart electricity metering. This is caused by the power filter and energy accumulation algorithm under dynamic signal conditions. By analyzing the typical intrinsic characteristics of large power electrical loads, this paper proposes a distorted m-sequence dynamic test (DmDT) signal model to reflect the characteristics and summarizes the parameter set that relates to the characteristics. In addition, based on the compressive measurement (CM) theory, a novel non-overlapping moving compressive measurement (NOLM-CM) algorithm is proposed to accurately estimate electrical energy. The performance of the non-overlapping moving compressive measurement (NOLM-CM) algorithm is tested under representative distorted dynamic random conditions. Simulation and experimental results indicate that the non-overlapping moving compressive measurement (NOLM-CM) algorithm achieves accurate estimation of electrical energy. Furthermore, the comparisons with five popular window-based estimation algorithms by simulations verify the higher performance of the non-overlapping moving compressive measurement (NOLM-CM) algorithm.