Abstract: The time series curve of slope displacement has complex nonlinear characteristics, and the traditional prediction model is not accurate enough to meet the current prediction requirements.In this paper, a prediction model of bird swarm optimization kernel extreme learning machine based on variational mode decomposition is proposed, and the slope displacement of a cement plant in Hebei province is predicted.Firstly, the slope displacement sequence is decomposed into a series of finite bandwidth subsequences by VMD program.Then, the phase space reconstruction is used for each sub sequence, and the kernel limit learning machine is used to predict the sub sequences.The bird swarm algorithm is used to optimize the embedded dimension of phase space reconstruction, penalty coefficient and kernel parameter in KELM to obtain the optimal prediction model.Finally, the prediction values of each sub series are superimposed to obtain the final prediction value of slope displacement.The results show that compared with KELM model, BSA-KELM model and EEMD-BSA-KELM model, the accuracy of BSA-KELM model based on VMD program is higher, which provides a more effective method for slope displacement prediction.