Fuzzy-rough imbalanced learning for the diagnosis of High Voltage Circuit Breaker maintenance: The SMOTE-FRST-2T algorithm

For any electric power system, it is crucial to guarantee a reliable performance of its High Voltage Circuit Breaker (HCVB). Determining when the HCVB needs maintenance is an important and non-trivial problem, since these devices are used over extensive periods of time. In this paper, we propose the use of data mining techniques in order to predict the need of maintenance. In the corresponding data, one class (minority, or positive class) is significantly less represented than the other (majority, or negative class). For this reason, we introduce a new imbalanced learning preprocessing algorithm, called SMOTE-FRST-2T. It combines the well-known Synthetic Minority Oversampling Technique (SMOTE) with a strategy of instance selection based on fuzzy rough set theory (FRST), using two different thresholds for cleaning synthetic minority instances introduced by SMOTE, as well as real majority instances. Our experimental analysis shows that we obtain better results than a range of state-of-the-art algorithms.