Application of VMD and Machine Learning for E-Bike Charging Current Detection
摘要
This paper proposes a charging current recognition algorithm for electric bicycles based on optimized data sets, designs and implements a current monitoring system. The feature extraction and recognition algorithm of electric bicycle charging current based on variational mode decomposition method is embedded in the current detection system. The variational mode decomposition method is used to extract the center frequency of the mode to construct the optimization data set, and the center frequency of several intrinsic mode functions obtained by the variational mode decomposition is used as the feature to construct the data set. The decision tree model is trained through a large amount of data to determine whether there is an electric bicycle charging current. In the process of data set construction, the optimal decomposition layer of variational mode decomposition is determined by the change rate of 99% bandwidth, and a better current recognition effect and a smaller amount of calculation are obtained.