Probing-Based Load Modeling and Optimal Low-Level Signal Design

As a crucial aspect of analyzing power system dynamics, load modeling has been continually aimed at increasing the accuracy of models and attaining the ability to track time-varying dynamics. In this paper, a probing-based load modeling approach is proposed. This innovative method actively injects probing signals into power systems through controllable devices to better excite load dynamics and has become increasingly applicable due to the increasing integration of controllable devices. To investigate this method and leverage its signal controllability advantages, an optimal probing signal design method is further proposed. First, theoretical identification errors are derived based on a reformulated frequency-domain identification model, providing a basis for evaluating identification performance. Subsequently, an optimization model for probing signal design tasks is established to minimize the theoretical variance, and a two-step optimization approach is developed to solve this problem. Case studies demonstrate that the optimal probing signals achieve superior identification performance while inducing fewer disturbances, which verifies the effectiveness of the proposed method. Furthermore, the consistency between the theoretical analysis and simulation results also provides additional validation for the proposed identification error model.