A novel adaptive single-phase reclosure scheme based on improved variational mode decomposition and energy entropy. (June 2020)

• Record Type: Journal Article
• Authors: Sun, Qiuqin; Lin, Fei; Jiang, Rong; Wang, Feng; Chen, She; Zhong, Lipeng
• Is Part Of: International journal of electrical power & energy systems. Volume 118(2020)
• Publication Date: 2020-06
• Subjects: Secondary arc -- Extended grading capacitor -- Single phase adaptive reclosure -- Variational mode decomposition -- Energy Entropy; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2019.105771
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A new technique for suppressing secondary arc is described. A novel SPAR scheme using an improved VMD algorithm combined with energy entropy is proposed. The quality factor is raised to determine the optimal decomposition number, and thus avoid modal mixing. The proposed method is almost not affected by the shunt reactor bank, the fault location, the fault resistance, etc. Abstract: The extended grading capacitor (GC) for multi-break circuit breaker is a cost-effective method to suppress secondary arc. Since the integration of GC alter the topology of the system, the existing adaptive auto-reclosure scheme should be changed. Considering the characteristics of line voltage waveforms under various faulty conditions, a novel adaptive single-phase reclosing scheme is proposed based on an improved variational mode decomposition (VMD) and energy entropy (EE). The voltage signal extracted from capacitive voltage transformer is decomposed into a series of finite bandwidth intrinsic mode functions (IMF), and then the energy entropy (EE) of each IMF component is computed. To avoid modal mixing, the concept of quality factor is raised. In the case of permanent fault, the natural component of the voltage attenuates fast due to the continuous discharge through the fault point. As time goes, such component approaches zero, and the total voltage becomes more smooth and regular. The EE value also decreases gradually. The EE value is considerably large for the transient fault due to the repetitive re-strike of electric arc, and the EE value has an oscillating trend. The fault type can be distinguished by setting a reasonable threshold. The performance of the proposed method is almost not affected by the installation mode of shunt reactor bank, the fault location, the shunt compensation degree, fault resistance, etc. Both the theoretical analysis and simulation results demonstrate that the proposed method can quickly and accurately identify the fault type.