A Novel Method for Generating Typical Scenarios of Power Systems with a High Proportion of Renewable Energy Considering Fault Scenarios

The randomness of renewable energy sources, combined with multiple faults, results in a wider variety and increased complexity of fault scenarios, posing significant challenges for the design of damping controllers. Therefore, a method for generating typical scenarios is proposed to categorize these diverse and complex fault scenarios. First, a Gaussian Mixture Model (GMM) is employed to represent the variability of renewable energy, generating numerous operating scenarios. The power flow data from these scenarios are then extracted to characterize system stability and clustered using a spectral clustering algorithm. Next, a cluster of stable operating scenarios is employed to generate N-1 fault scenarios. Oscillation amplitudes and frequencies of the fault responses are computed using the Prony algorithm to reflect the system’s oscillatory behavior after faults. A GMM is then used to cluster these features and determine the Gaussian probability distributions for the fault scenarios. Finally, The proposed method’s effectiveness is cross-validated by electromagnetic transient simulations using standard cases on the CloudPSS platform.