Modeless Streaming Synchrophasor Data Recovery in Nonlinear Systems

This paper develops amodel-free approach to recover the missing points in streaming synchrophasor measurements obtained in nonlinear dynamical systems. It can accurately recover simultaneous and consecutive data losses across all channels for some time consecutively without modeling the nonlinear dynamics at all. The idea is to lift the nonlinear system to an infinite-dimensional linear dynamical system and exploit the low-rank Hankel in the lifted dimension to characterize the system dynamics. The kernel technique is employed to handle the implicit lifting function. Compared with existing model-free synchrophasor data recovery methods, our approach drops the assumption of linear systems and applies to general nonlinear systems. The algorithm has low computational complexity and can be implemented in real time. The method is validated through numerical experiments on recorded synchrophasor datasets.