Fault-tolerant wide-area control for power oscillation damping

In this paper, the effectiveness of using both local and remote (wide-area) feedback signals for power oscillation damping (POD) controllers is shown. However, the challenge is to guarantee a minimum level of dynamic performance with only the local signal following sudden loss of remote signals. A case study on the Nordic equivalent system is presented to show that the closed-loop response could deteriorate once the remote signals are lost. A fault-tolerant control (FTC) design methodology is presented to solve this problem and ensure an acceptable performance level even in case of loss of remote signals. The FTC design methodology is based on simultaneous pole-placement for normal and loss of (remote) signals conditions along with minimisation of control effort. The problem is solved non-iteratively using Linear Matrix Inequalities (LMIs). Under the normal condition (when both local and remote signals are present) the fault-tolerant controller (FTC) requires more control effort as compared to a conventional controller (CC) in order to achieve the same performance. However, case studies on the Nordic equivalent system confirm that the proposed FTC is able to produce acceptable performance in case of loss of the remote signals while the response with a CC is unacceptable.

I. Introduction

USE of feedback signals from geographically remote locations could improve the effectiveness of power oscillation damping (POD) [1] control. With the state-of-the-art wide-area measurement systems (WAMS) infrastructure, power oscillation damping using remote (or wide-area) signals is certainly feasible. This could potentially improve the stability limits and allow operation of the transmission lines closer to their thermal capacity, without compromising security. Utilities however, are concerned about the consequences of unacceptable delay or complete loss of one or more of the remote feedback signals which could jeopardize the dynamic performance of their system.