A passivity based, system reference frame approach for decentralized stability analysis and control design in future power grids

Abstract

Over the last decades, power systems have been through critical changes as a result of the worldwide efforts to decelerate climate change and global warming. Such changes were the introduction of new generation and storage technologies, and the rapid increase of the share of Renewable Energy Sources (RES) in power generation. Although these advances contributed to technological and economic development, they have introduced numerous issues that were not previously encountered in traditional power grids. Specifically, the gradual replacement of the large fossil-fueled plants with a large number of small sparselylocated RES resulted in the significant decrease of system rotational inertia and the emergence of serious stability-related problems. Despite the latest decisive steps in the area of stability analysis and control design, existing power systems are still in danger due to the continuously increasing challenges they encounter. An effective way to overcome these problems is the adoption of more accurate dynamical models for both the network and the power system components within stability studies. Such accurate modeling will not only assist in the design of more effective control mechanisms, but it will provide useful insights regarding the stability and the reliability of the system. The current thesis aims to address the above problems by introducing a novel approach for decentralized stability analysis and control design in existing and future power grids wherein more detailed dynamical models are employed. The proposed approach relies upon the transformation of both the network and the bus dynamics into the system reference frame instead of each bus local dq coordinates. In particular, this transformation allows the formulation of the network equations as an input-output system which we show it is passive even if the network’s lossy and dynamic nature is taken into account. The passivity property of the adopted network model along with the local passivity conditions imposed on a broad class of bus dynamics guarantee the asymptotic stability of the whole power network in a completely decentralized manner. The use of such a general representation also facilitates the incorporation of more accurate dynamical models for the power system components and their control mechanisms, even though their inclusion in such a decentralized analysis has been difficult. A further detailed discussion regarding the advantages of the presented approach for the reliable and robust operation of the future low-inertia power grids as well as the design of more effective distributed control mechanisms is provided. The proposed stability analysis framework is finally verified through realistic applications and simulations on several testbed systems such as the Two Area Kundur, the IEEE 68 Bus test systems and the IEEE 37 Node test feeder.