Power System Post-contingency Frequency Prediction under Incomplete Information with Data-Model-Hybrid Method

Power system post-contingency frequency prediction can help operators make reasonable decisions quickly and regulate the impact of unpredictable contingencies. It is essential that post-contingency prediction models give consideration to both accuracy and speed, and should be able to operate under incomplete information. A data-model-hybrid method is proposed where a simplified model-based method is used to obtain the rough prediction, which is then optimized by a data-based model. Case study is presented over New England 39-bus system, and the result shows that the method proposed can give reasonable result within a short period of time.