Abstract: Taking into account the non-linearity and non-stationarity of the carbon price series in carbon trading, this paper proposes a combined prediction model which is based on the multi-mode decomposition, sample entropy, the whale optimization algorithm and the LSTM neural network for predicting carbon trading price.Firstly, the original carbon price series are decomposed by using the singular spectrum decomposition（SSD）, the variational modal decomposition（VMD） and the complete ensemble empirical mode decomposition with adaptive noise（CEEMDAN） respectively to reduce the complexity and non-stationarity of the original data and realize the complementation of the modal components regular pattern of different modes.Secondly, the sample entropy algorithm is used to reconstruct the entropy close component into a new component to improve the prediction efficiency.Finally, the WOA-LSTM combined prediction network is used to establish the time characteristic relationship between historical carbon trading prices, and the final prediction results are obtained based on the spatio-temporal correlation analysis.The experiment results show that the combined prediction model based on multi-mode decomposition-sample entropy-WOA-LSTM can improve the accuracy of carbon trading price prediction effectively.