Classification for transient overvoltages in offshore wind farms based on multi-scale mathematical morphology. (March 2022)

• Record Type: Journal Article
• Authors: Tang, W.H.; Gu, Y.C.; Xin, Y.L.; Liang, Q.H.; Qian, T.
• Is Part Of: International journal of electrical power & energy systems. Volume 136(2022)
• Publication Date: 2022-03
• Subjects: Multi-scale mathematical morphology -- High/low-frequency energy ratio -- O shore wind farms -- Internal transient overvoltages; 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.2021.107157
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: A multi-scale MM method is proposed for extracting the features of transient overvoltages. A multi-scale MM method is proposed for extracting the features of transient overvoltages. Separated nonlinear HF/LF components are designed as recognition features of SVM. The proposed method has been verified for extensive simulation and experiment data. The classification ability of the proposed method is superior to that of WT algorithm. Abstract: At present, the transient overvoltages in o shore wind farms caused by faults or frequent operations of electrical equipment are particularly severe. The quick and accurate classi cation of transient overvoltages is important for protecting electrical equipment from damage. In order to classify the internal transient overvoltages in o shore wind farms, this research rstly proposes a feature extraction method based on mathematical morphology, which proposes a new morphological structure element and utilizes a multi-scale mathematical morphology to extract the high/low-frequency components of transient overvoltages. Then a high-frequency feature and a high/low-frequency energy ratio feature are constructed as identi cation features. Finally, based on the constructed features, a support vector machine is employed to identify di erent types of internal transient overvoltages. Extensive simulations and experiments are performed to verify the superiority of the proposed feature extraction method based on mathematical morphology, which is also compared with the widely used conventional wavelet algorithms. Results show the proposed mathematical morphology feature extraction method is capable of classifying and discriminating among various types of internal transient overvoltages with a simple procedure and an improved accuracy. The proposed method provides a valuable reference for the protection and insulation coordination of electrical equipment in o shore wind farm substations.