Out-of-step oscillation splitting criterion based on bus voltage frequency

Under the background of complicated interconnected network,the splitting criterion for accurately capturing the electrical center in real time is the prerequisite of power grid splitting.This paper studies the features of electric quantity in the electrical center in aspect of the instantaneous frequency,and proposes the out-of-step splitting criterion for power systems based on bus voltage frequency.Firstly,through the establishment and solution to the out-of-step model of the power grid,the analytical expression of the voltage frequency at any position is obtained in the out-of-step oscillation,and the voltage frequency features of electrical center and non-electrical center are analyzed in details.Then,this paper constructs the typical scene of migration of electrical center to study the change rules of voltage frequency.Finally,the splitting criterion based on bus voltage frequency is proposed as well as the instruction for use.This criterion is easy to be realized and can adapt to the migration of electrical center.Also it is free from the limits of power network structure and operational mode.Simulation results of CEPRI-36 system and interconnected network example of one actual region verify the accuracy and the effectiveness of the proposed criterion.

One-against-all-based Hellinger distance decision tree for multiclass imbalanced learning

Since traditional machine learning methods are sensitive to skewed distribution and do not consider the characteristics in multiclass imbalance problems,the skewed distribution of multiclass data poses a major challenge to machine learning algorithms.To tackle such issues,we propose a new splitting criterion of the decision tree based on the one-against-all-based Hellinger distance(OAHD).Two crucial elements are included in OAHD.First,the one-against-all scheme is integrated into the process of computing the Hellinger distance in OAHD,thereby extending the Hellinger distance decision tree to cope with the multiclass imbalance problem.Second,for the multiclass imbalance problem,the distribution and the number of distinct classes are taken into account,and a modified Gini index is designed.Moreover,we give theoretical proofs for the properties of OAHD,including skew insensitivity and the ability to seek a purer node in the decision tree.Finally,we collect 20 public real-world imbalanced data sets from the Knowledge Extraction based on Evolutionary Learning(KEEL)repository and the University of California,Irvine(UCI)repository.Experimental and statistical results show that OAHD significantly improves the performance compared with the five other well-known decision trees in terms of Precision,F-measure,and multiclass area under the receiver operating characteristic curve(MAUC).Moreover,through statistical analysis,the Friedman and Nemenyi tests are used to prove the advantage of OAHD over the five other decision trees.