In view of the issue of harmonic interference detection and reactive power compensation of battery for medical organization

a detection method based on flexible analytic wavelet analysis is adopted in this paper. Firstly

the rectified voltage and current data of the battery is obtained as different scales of component signal by way of flexible analytic wavelet decomposition

then

feature extraction and selection are carried out for the scale signal with combination of fuzzy entropy. Finally

least squares support vector machine is used for classification and identification so to detect accurately harmonic contents of the batter energy signal. It is shown by the experimental test result that the average accurate rate of harmonic detection isis 96%

which provides guarantee for reactive power compensation further.