基于传统的扩展卡尔曼滤波器（ＥＫＦ），本文提出了一种带次优渐消因子的ＥＫＦ用于非线性时变随机动态系统状态与参数的联合估计。应用于液体火箭发动机健康监控算法的仿真研究表明，本文所提出的联合估计器具有较好的收敛性、实时性和动态跟踪能力。此外，文中还讨论了联合估计器应用于实际系统的有关问题。 Based on the traditional Extended Kalman Filter (EKF), this paper proposes a joint EKF with suboptimal fading factor for the joint estimation of states and parameters of a nonlinear time-varying stochastic dynamic system. The simulation research on the health monitoring algorithm of liquid rocket engine shows that the joint estimator proposed in this paper has better convergence, real-time and dynamic tracking ability. In addition, the article also discusses the joint estimator applied to the actual system of the relevant issues.