Power quality disturbance detection based on morphology singular entropy

This paper proposes a novel method for power quality (PQ) disturbance detection, based on morphology singular entropy (MSE). MSE consists of three techniques, i.e., mathematical morphology (MM), singular value decomposition (SVD) and entropy theory. The proposed method firstly utilizes MM to obtain the filtered outputs of the original signal at different levels. Then a matrix composed by the outputs obtained is decomposed by SVD to get the singular values. Finally, the entropy of the signal is calculated using the singular values obtained according to the entropy theory to monitor the occurrence of a power disturbance. Simulation studies have been conducted using MATLAB and PSCAD/EMTDC. The graphic results have demonstrated that the proposed method can detect the beginning and end of the power quality disturbances correctly. Moreover, the quantitative results have further validated the accuracy of the proposed method.