基于调制宽频模态分解和局部保持投影特征融合的光伏直流电能质量扰动识别OA

Nonlinear loads in photovoltaic(PV)direct current(DC)systems may introduce disturbances such as ripples,transients and noise in the DC power signal.Existing time-frequency analysis methods,such as variational mode decomposition,often lead to errors when decomposing PV DC power signals.This paper,building upon the foundation of broadband mode decomposition,employs modulated broadband mode decomposition(MBMD)with a modulation difference operator to denoise PV DC power signals,aiming to reduce decomposition errors.The proposed approach first utilizes MBMD for adaptive signal decomposition,incorporating a local preserving projection(LPP)algorithm for feature fusion.Finally,a back propagation artificial neural network model is employed for intelligent recognition of DC power quality.Simulation and experimental analysis demonstrate that the proposed method can accurately identify various types of disturbances in PV DC power.