Variational mode decomposition and weighted online sequential extreme learning machine for power quality event patterns recognition

• Published in: Neurocomputing (Amsterdam) Vol. 310; pp. 10 - 27
• Main Authors: Sahani, Mrutyunjaya, Dash, P.K.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 08.10.2018
• Subjects: Power quality events; Power quality indices; Real-time analysis; Variational mode decomposition; Weighted online sequential extreme learning machine; Power quality events; Variational mode decomposition; Power quality indices; Real-time analysis; Weighted online sequential extreme learning machine
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2018.03.056

•Variational Mode decomposition (VMD) is used for Power Quality signal processing.•New Power Quality indices are derived from VMD for PQ event recognition.•Weighted On-line sequential ELM (WOSELM) is used for classification of PQ patterns.•Comparison with Empirical Mode decomposition based WOSELM is presented.•Hardware implementation of the new PQ monitor is used to validate its performance. In this paper, variational mode decomposition (VMD) and a newly developed weighted online sequential extreme learning machine (WOSELM) are integrated to detect and classify the power quality events (PQEs) in real-time. The feasibility of VMD is validated by applying on PQEs (such as harmonic and flicker) for the estimation of magnitude, phase,and frequency. Estimated results prove the usefulness of VMD and further four efficacious power quality indices of the band-limited intrinsic mode functions (BLIMFs) are extracted. The indices are used for the classification of single and multiple PQEs using different advanced classifiers. The recognition architecture of variational mode decomposition with weighted online sequential extreme learning machine (VMD-WOSELM) is tested and compared withother methods. The robust anti-noise performance, faster learning speed, lesser computational complexity, superior classification accuracy and short event detection time prove that the proposed VMD-WOSELM method can be implemented in electrical power systems. Finally, a PC interface based hardware prototype is developed to verify the cogency of the proposed method in real time. The feasibility of the proposed method is tested and validated by both the simulation and laboratory experiments.