Probabilistic Assessment Method of Time-Varying Sub–/Super-Synchronous Oscillation Risk for On-Grid Wind Power Systems

Abstract

As the proportion of new energy integrated into the grid rises annually, it is of great significance to quickly and accurately evaluate the risk of complex characteristic sub–/super-synchronous oscillations. Therefore, a probabilistic assessment method of time-varying oscillation risk for the on-grid system of wind power is proposed. First, based on the measured wind speed data, a credible mathematical model of the system's random fluctuation is established. The total least squares estimation of signal parameters via rotational invariance techniques is used in combination with the singular value difference ratio method to realize the automatic determination of the model order, and then the oscillation modal parameters of the system are accurately identified from the real-time measurement signal. Second, based on the time-variables nonlinear approximation method of variable forgetting factor recursive least-squares (VFFRLS), the time-varying forgetting factor is introduced with the change of uncertainty factors, which overcomes the modeling error and reduces the amount of calculation. Based on the probability distribution of uncertainty factors, the Latin hypercube sampling technique is used to stratify the sampling in the random variable space and fuse the risk matrix to realize the probabilistic assessment. Finally, the accuracy and reliability of the proposed method are verified by arithmetic examples. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.