Operating under harsh and complex environments, the high performance requirement of aircraft engines ask for stable and efficient controllers. Model predictive control (MPC) is a potential and recently emerging technique for aero-engine control system because of its excellent multivariable control ability and prediction ability of future information. However, MPC weight parameter tuning has significant influence on control effect and is usually selected based on human experience or simply trial-and-error. In this paper, based on the linear time-invariant (LTI) model of a typical aeroengine, deep deterministic policy gradient (DDPG) algorithm is combined with MPC, in which the MPC weight matrix is adjusted through DDPG to optimize the performance of MPC controller in real time to adapt to the changing flight conditions and engine states. This method of DDPG adjusting MPC weight matrix can not only improve the performance and efficiency of aero engine under various operating conditions, but also enhance the robustness and adaptability of the system. Simulation results verified the proposed approach with improved control performance compared with the typical MPC controller