In order to improve the fuel economy of fuel cell hybrid electric vehicles during short range driving, a vehicle speed prediction model structure VBSnet based on hybrid deep neural network was constructed. This structure not only further improves the convolutional network based on the VGGNet structure, but also introduces a bidirectional long shortterm memory neural network to effectively learn the spatiotemporal dependencies of the entire vehicle speed prediction sequence. Simultaneously considering the influence of prediction time domain and input sequence length on the prediction accuracy of shortrange vehicle speed problems, Bayesian optimization hyperparameters are used to further improve the prediction accuracy of VBSNet. To address the online optimization and computational efficiency issues of energy management strategies, a multiobjective optimization based on model predictive control (MPC) energy management strategy was designed. This strategy can achieve a balance and optimization of hydrogen consumption, lithium battery state of charge (SOC) maintenance, and fuel cell utilization efficiency. Finally, under actual vehicle conditions, the proposed strategy was compared with rulebased strategies, resulting in fuel economy improvements of 7.25%, 9.94% and 19.23%, and better SOC maintenance characteristics.