Research On Termination Algorithm Of Time-domain Simulation For Power System Transient Stability

With the rapid developing of the power system, every country pays more attention to the security of power system. As a result, it has become an urgent task to find a method of power system transient stability assessment which is quick and accurate, in order to ensure the safe and stable operation of the power system.In this paper, the existing transient stability assessment method is summarized briefly. The unity of speed and accuracy is the difficulties and focus in the study of transient stability analysis. The determination of time-length in the time-domain simulation relies on the experience of experts in advance. However, this time-length cannot be determined accurately in order to ensure the reliability of the analysis. Thus, the researchers have to set integration interval very large, causing a lot of unnecessary computation. This paper focuses on the problem of excess integration time-step produced by the whole process of traditional numerical integration, try to find an accurate and efficient termination algorithm of simulation. By this algorithm, the process of time-domain simulation can be terminated in advance, and the amount of unnecessary calculation can be reduced significantly with the assurance of the accuracy of the transient stability qualitative analysis.Accordingly, this paper presents three kinds of simulation termination criteria algorithm to predict the transient stability of power system with the short-time-domain simulation:(1) the method based on comprehensive performance index. It gets transient stability criterion threshold by the sort of index at different times. During the process of time-domain simulation, part of the contingency examples are terminated in advance through criterion threshold, as a result, the efficiency of the simulation get improved.(2) the method based on Artificial Neural Network. Building an ANN classifier for transient stability, the contingency examples are divided into three subsets:stable contingency, unstable contingency and uncertain contingency.(3) the method based on IEEAC theory. The paper presents the index for evaluating coherency in the cluster of trajectories, to reflect transient stability of the disturbed trajectories after the observation interval.Finally, the application the power system in certain place and IEEE New England10-machines and39-bus system proves the effectiveness and reliability of above-mentioned algorithms.