Battery matching method

The invention relates to the technical field of generalities of batteries, and particularly discloses a battery matching method. The method specifically comprises the following steps: acquiring a timeseries curve of a battery voltage of sample battery formation; completing a missing value of the time series curve by using a KNN (K-Nearest Neighbor) algorithm; eliminating interference informationin a process of obtaining the time series curve by adopting a wavelet noise elimination method; aligning a time series of a time series curve to be compared by adopting a dynamic time wrapping method;performing primary screening on a sample battery by adopting voltage and capacitance standards; clustering the sample battery according to a static final voltage of the sample battery formation and acapacitance measured by discharging by adopting a Gaussian mixed model; and matching the clustered sample battery according to a characteristic distribution density of the sample battery. The Gaussian mixture model is used to c