Inter Area Oscillations In Power Systems With Uncertainties

"The steadily increasing load demand and the liberalization of electricity supply industry have resulted in heavy power trades over the weak tie lines of modern wide-area power grids. This effect, compounded by the slow addition of new transmission facilities, introduces numerous stability challenges. Among them, poorly-damped inter-area oscillations pose a serious threat to safe power system operation and may lead to cascading outages and blackouts. Nowadays, power networks are complex and experience various types of uncertainties causing model inaccuracies. Thus, it is understood that conventional model-based inter-area mode monitoring and control philosophy requires reconsideration. These ideas determine the scope of this thesis, which mainly focuses on the design of data-driven damping control strategies for inter-area modes. Recognizing the potential of recently developed Wide-Area Measurement System (WAMS) technology to provide a coherent picture of the entire network in real time based on Phasor Measurement Unit (PMU) data, this work proposes two Wide-Area Damping Control (WADC) algorithms against inter-area oscillations. Execution of the proposed schemes involves the online identification of the dynamic system state matrix from PMU measurements. A novel centralized participation factor-based WADC that can target multiple inter-area modes without affecting the rest of the modes is firstly presented. It is completely independent of the network model knowledge, while only requiring the generator inertia and damping constants as known parameters. The advantage of such control over model-based WADC is its capability to quickly adapt to operating condition variations. Additionally, the developed WADC algorithm does not require offline training, is adaptive to the selection of the PMU dataset and can be mapped to the actual power network dynamics. In order to bypass the high communication requirements and computational burden of centralized control architectures, a novel Modal Linear Quadratic Regulator (MLQR)-based sparse optimal WADC is also proposed. This methodology is purely data-driven and can directly shape the closed-loop damping features of every weakly-damped inter-area mode. Moreover, it takes into account the communication network constraints of WAMSs and demonstrates comparable performance to model-based and centralized model-free WADC. Finally, the thesis addresses the issue of small-signal stability monitoring degradation caused by high penetration of intermittent wind generation. A new data-driven Energy Storage System (ESS)-based algorithm is introduced and contributes a wind power balancing policy to improve the inter-area mode monitoring, and thus the WADC effectiveness. Case studies on the IEEE 39-bus, 68-bus and 145-bus benchmark systems validate the performance of the proposed WADC and ESS techniques"--