Research On Power System Stability Control Based On Wide Area Measurement System

With the development of the society, the capacity of the power grid has been growing rapidly and various new technologies have been introduced to the system. As a result, some traditional control strategies may not be good enough to meet the system requirements under the new circumstances. In recent years, researches on traditional method improvement and seeking of new practical robustness method have attracted more and more attentions.In this dissertation, three major researches of low-frequency oscillation and under-frequency load shedding (UFLS) have been introduced respectively.Firstly, a probability-one homotopy method for PSS parameter tuning is proposed. Both standard and least-square partial pole assignment problems are presented. Implementation innovations, including the development of an efficient algorithm for computing the second order derivatives of eigenvalues, eigenvalue tracking and parameter rescaling, are reported in detail. Simulation results of ten test systems including the East China power system with465machines and3296buses are described, showing that the suggested method is a reliable parameter tuning tool and it has the potential for solving real-world stabilizer parameter tuning problems.Secondly, generalized phase compensation based on a simple eigenvalue representation formula is presented. This formula can justify the celebrated phase compensation principle in the single-machine-infinity-bus from algebraic point of view. The generalizes phase compensation principle, it appears as a stability criterion, is applicable to multi-machine systems. It can also be used in PSS siting, find compensation angle for machines individually and tune the gain of the PSSs. Furthermore, WAMS based PSS is applied and compared to tradition ones. Studies of five different test systems verify the correctness and feasibility of the prosed method.Thirdly, a time-based adaptive UFLS scheme is presented. Based on the system frequency response (SFR) model, the dynamic response of the system is predicted. Then considering some basic constrains of the system and the mechanical power limit, an "Implicit Enumeration with Adaptive Discretization" algorithm is used to get the optimal results. Simulations on a real, small system weakly connected to the Eastern China system have been presented, it shows that the proposed method can optimize total load shed in small disturbance and has higher robustness in serious one.