Diagnosing utility grid disturbances in photovoltaic integrated DC microgrid using adaptive multiscale morphology with DFA analysis

Abstract

In this study, a novel approach is proposed to distinguish both single and mixed power quality (PQ) disturbances in the network connected to the utility grid in a photovoltaic (PV) integrated DC microgrid. To analyze PQ disturbances in the proposed system, AC voltage signals are captured at the PCC of the utility grid. The captured voltage signals contain strong noise which impacts the system robustness. To resolve this problem, an adaptive multiscale improved combination morphological filter (AMICMF) technique is introduced in the proposed system. In this technique, an improved combination morphological filter (ICMF) is first proposed by using the basic four morphological operations. Then multiscale operation (MICMF) is done by different structural element (SE) lengths chosen by sparse weighted Kurtosis index (SKI w ). The obtained optimized signal is processed through the detrended fluctuation analysis (DFA) which gives the scaling exponent ( λ ) values based on window length. These scaling exponent values are used to classify the PQ disturbances in terms of scatter plots. The performance of the proposed study is calculated by classification accuracy (CA) and relative computational time (RCT) and the efficacy of the proposed technique is validated by comparing the CA and RCT with existing techniques. The entire study is tested on MATLAB/Simulink environment. The complete study is tested and validated using a high-speed multifunction National Instrument USB6008 device for the online monitoring of PQ disturbances on the MATLAB/Simulink environment. • A new adaptive multiscale morphological filter is used for PQ analysis of DC microgrid. • The adaptation of the filter is accomplished using a sparse weighted kurtosis index. • Detrended fluctuation analysis is used to process the optimized signal to yield scaling exponents. • These scaling exponent values are used to classify the PQ disturbances in terms of scatter plots.