The impact of load models in an algorithm for improving voltage stability via demand response

Abstract

Increasing communication and control capabilities will allow future power system operators to exploit large quantities of responsive demand. This paper discusses ongoing work that employs demand response to improve voltage stability via virtual spatial shifting of loads (i.e., altering the locational distribution of power consumption in one time period with an energy payback in a following time period). In this paper, we study the impact of load models on a previously proposed iterative linearization algorithm to determine loading patterns that maximize a voltage stability margin, namely, the smallest singular value (SSV) of the power flow Jacobian matrix. Specifically, we extend the algorithm to enable inclusion of composite load models consisting of both “ZIP” components and a steady-state squirrel-cage induction machine (IM) model. We then investigate the impact of different load models on both the stability margin and the loading pattern. Using the IEEE 14-bus system as an illustrative example, the results show that the type of load model affects the nominal system's SSV, the optimal SSV, and the optimal loading pattern. The maximum-achievable percent change in SSV is larger using IM models than using ZIP models. We also discuss the difficulty in interpreting the stability margin when the system undergoes structural changes resulting from the use of different voltage-dependent load models.