Deep learning-based multi-model ensemble method for classification of PQDs in a hydrogen energy-based microgrid using modified weighted majority algorithm. (28th February 2023)

• Record Type: Journal Article
• Authors: Bayrak, Gökay; Küçüker, Ahmet; Yılmaz, Alper
• Is Part Of: International journal of hydrogen energy. Volume 48:Number 18(2023)
• Page Start: 6824
• Page End: 6836
• Publication Date: 2023-02-28
• Subjects: Deep learning -- Microgrid -- Hydrogen energy-based distributed generation -- Power quality disturbances -- Weighted majority voting; Hydrogen as fuel -- Periodicals; Hydrogène (Combustible) -- Périodiques; Hydrogen as fuel; Periodicals; 665.81
• DOI: 10.1016/j.ijhydene.2022.05.137
• Languages: English
• ISSNs: 0360-3199

Abstract

Abstract: In this study, new multiple deep classifiers with a modified Weighted Majority Voting (WMV)-based method are proposed to identify power quality disturbances (PQDs) in a hydrogen energy-based microgrid. In the proposed approach, closed-loop deep LSTM (Long Short Time Memory), deep CNN (Convolutional Neural Network), and hybrid (CNN-LSTM) models are used for automatic feature extraction and classification. Then, a modified WMV method is employed to ensemble the outputs of the three deep learning (DL) classifier models. The enhanced WMV mechanism performs an automatically updated weighting based on the validation results of the DL classification models, unlike voting methods in the literature. The improved WMV mechanism eliminates the challenges of using multiple DL classifiers in the voting system. The mathematical data results in LabVIEW, simulation results in Matlab/Simulink, and real data results in the laboratory show that the proposed method shows superior performance in accuracy and noise immunity to state-of-the-art methods. Highlights: Deep classifiers with a modified WMV-based method are proposed for a HE-based microgrid. Challenges for using multi-DL classifiers in the voting system are addressed. Deep CNN, LSTM, and hybrid models are used for automatic feature extraction and classification. Eight single and ten complex PQD events in a HE-based microgrid were classified. Results show high performance in accuracy and noise immunity.