Optimization for Data-Driven Preventive Control Using Model Interpretation and Augmented Dataset

Transient stability preventive control (TSPC) ensures that power systems have a sufficient margin by adjusting flow before faults occur. The generation of TSPC measures requires accuracy and efficiency. In this paper, novel model interpretation-based multi-fault coordinated data-driven optimization strategy is proposed. First, an augmented dataset covering the fault information constructed, enabling transient assessment (TSA) to discriminate system under different scenarios. Then, adaptive synthetic sampling (ADASYN) method implemented deal with imbalanced instances systems. Next, instance-based machine interpretation tool, Shapley additive explanations (SHAP), embedded explain TSA model’s predictions find out most effective objects, thus narrowing number objects. Finally, differential evolution deployed optimize measures, taking into account security economy TSPC. proposed method’s efficiency robustness are verified on New England 39-bus IEEE 54-machine 118-bus system.