Integrating Hierarchical Clustering and Pareto-Efficiency to Preventive Controls Selection in Voltage Stability Assessment

Many methods to estimate the cut-off value in order to determine the actual groups from a dendrogram given via hierarchical clustering methods have been proposed in the litetarure. However, in most of the cases, the determination of this value is critical and based on heuristics. In this context, a new method based on Pareto-optimality and on the hierarchical clustering method called Data Mine of Code Repositories (DAMICORE) to determine the most promising groups in a given dendrogram is proposed. This method is called Pareto-Efficient Set Algorithm (PESA). In order to validate the proposed method, PESA was applied find the most promising groups for the preventive control selection problem in the context of voltage stability assessment in electrical power systems. PESA was able to design a set of controllers to eliminate all critical contingencies and was successfully tested in a reduced south-southeast Brazilian system composed of 107 buses.