Excitation System Parameter Identification And PSS Parameter Optimization Using Ant Colony Algorithm

The China Southern Power Grid is the first long-distance, bulk-power, high-voltage AC/DC power grid in China which transmits large amounts of power from west to east. It covers five provinces and spans almost 2,000 kilometers hence both angle stability and voltage stability are in great concern in this system.Excitation systems are part of synchronous generators. They have significant inpact on system safety and stability. Therefore, model accuracies are important for powe system analysis, planning, design, operation, management and so on.In the CSG, ac excitation systems and static excitation systems are widely used. Their mathematical models and the common processure used to get the models are first described. Then, advantages and disadvantages of the processure are analyzed. Because some model parameters are sensitive to generator small-signal voltage-step response whereas the others are sensitive to the large-signal voltage-step response, a layored identification method using the Maximun-Minimun Ant System algorithm is proposed in this paper. The parameters that dominate generator small-signal step response are first identified and then the others. In this way, the over-all system model is obtained.The low-frequency oscillations and PSS control principles are described. When accelerating power is used as the input signal, PSS mechanism is analyzed.The existing PSS tuning processures are analyzed. Based on the measured excitation system phase-lag, a PSS parameter optimization method is proposed. First, parameters of PSS lead-lag compensation networks are determined to satisfy the phase requirements on the whole low-frequency band. Then, control gain of PSS is tuned by minimizing power oscillations subsequent to a disturbance. The Maximum-Minimum Ant System algorithm is used to solve the optimization model which provides a technical means for the spot tuning of PSS. Using the method and the measured phase-lag characteristics of lots of generating units, tuning of PSSs in the China Southern Power Grid is achieved which justify the effectiveness of the method suggested. This paper proposes a method for excitation system identification and an optimization tool for the tuning of PSS using the ant colony optimization algorithm which facilitates modeling and setting of excitation systems and PSSs in power systems.The research of this paper is supportted by Key Laboratory of Guangdong Regular Higher Education Institutions under award No:ZDSYS200701.