Aiming at the problem that existing multivariate grey incidence models can not be applied to different length multivariate time series (MTS), a new model was proposed based on spatial pyramid pooling (SPP). Firstly, the local features of an MTS with different lengths were pooled and aggregated by SPP to construct feature pooling matrices on the same size. Secondly, classic multivariate grey incidence models were used to measure the degree of incidence between feature pooling matrices. Finally, the effectiveness of the models was verified on two data sets compared with a variety of algorithms.