| In recent years,in order to ensure the safety of power supply in public power areas,people have paid more and more attention to the power quality management of low-voltage power supply and distribution systems,and the detection of malignant loads is one of its important contents.This paper proposes an intelligent malignant load detection method,and conducts simulation research and experimental verification.First,a slope fitting method based on an improved variable step size LMS algorithm is proposed.After filtering the power waveform,add a sliding data window to fit the slope.While filtering unnecessary noise,it retains data characteristics,and can accurately locate and mark the start and end time of load events,facilitating data analysis.Secondly,based on waveform feature analysis,the improved wavelet packet harmonic detection algorithm is used to perform harmonic analysis on the marked current waveform to detect the types of electrical appliances.Based on the five-layer wavelet packet transform of db40,the Hilbert transform is used to shift the frequency of each harmonic to the sideband for signal decomposition and reconstruction,avoiding high-frequency aliasing and spectrum leakage,achieve high-precision detection of various harmonics.Build a database of typical malignant loads in advance.If the harmonic value comparison is successful,it is judged as a malignant load,which successfully improves the accuracy of low-power malignant load identification.At the same time,the area phase method is used to determine the high-power malignant load,which saves the storage space of the electric meter,reduces the system cost,and improves the calculation speed.Finally,a malignant load intelligent detection system was built based on the SPM9511 guide rail prepaid meter,and the malignant load intelligent detection experiment was carried out,which realized the rapid and accurate detection of malignant load. |
