The denoising and detection of transient disturbances are two important subjects for power quality monitoring and analysis. To effectively denoise and detect transient disturbances under noisy conditions, an improved ...The denoising and detection of transient disturbances are two important subjects for power quality monitoring and analysis. To effectively denoise and detect transient disturbances under noisy conditions, an improved iterative adaptive kernel regression method is proposed in this paper. The proposed method has advantages that itdoes not need to estimate the noise variance or a filter threshold, and has both denoising and detection capabilities for transient disturbances. Simulation results demonstrate that the proposed method provides excellent denoising effects, which can not only suppress noise effectively but also preserve disturbance features of sudden change points well. Additionally, it provides good detection and location performance for single and combined transient disturbances, even under strong noise conditions. Finally, the effectiveness of the proposed method is further verified by using real disturbance data.展开更多
基金supported in part by the NationalKey R&D Program of China (No. 2016YFB1200401, No. 2017YFB1201103)in part by the Program for Application of Cophase Power Supply Technology (No. 2018002)
文摘The denoising and detection of transient disturbances are two important subjects for power quality monitoring and analysis. To effectively denoise and detect transient disturbances under noisy conditions, an improved iterative adaptive kernel regression method is proposed in this paper. The proposed method has advantages that itdoes not need to estimate the noise variance or a filter threshold, and has both denoising and detection capabilities for transient disturbances. Simulation results demonstrate that the proposed method provides excellent denoising effects, which can not only suppress noise effectively but also preserve disturbance features of sudden change points well. Additionally, it provides good detection and location performance for single and combined transient disturbances, even under strong noise conditions. Finally, the effectiveness of the proposed method is further verified by using real disturbance data.
