基于ASTSMO和UIO的故障估计方法

针对系统在有未知干扰情况下的故障估计问题，提出一种基于自适应Super-Twisting滑模观测器（ASTSMO）和未知输入观测器（UIO）的故障估计方法。不需要已知故障导数的上界，避免了现有自适应算法存在的滑模增益过估计问题，并且能够处理多执行器同时发生故障的情况。首先，通过非奇异变换将原系统降阶为两个子系统，其中一个子系统只受故障的影响，另一个子系统同时含有故障和不确定干扰。对两个子系统分别设计ASTSMO观测器和UIO观测器，并对误差系统有限时间内收敛的条件进行了证明，同时给出了滑模增益初始值和时变增益的设计方法。然后，基于等效控制的概念对故障进行检测和估计。最后，通过仿真算例验证了所提故障估计方法的有效性。 

Fault estimation method based on adaptive super-twisting sliding mode observer and unknown input observer

For the problem of fault estimation with unknown disturbance, a fault estimation method based on Adaptive Super-Twisting Sliding Mode Observer (ASTSMO) and Unknown Input Observer (UIO) is proposed. This method does not require that the upper bound of fault derivative is known, which avoids the problem of sliding-mode gain overestimation in the existing adaptive algorithms, and is able to handle simultaneous faults of multiple actuators. First, the original system is degraded to two subsystems by non-singular transformation, one of which is only affected by the fault, and the other subsystem contains both fault and uncertain interference. ASTSMO and UIO are designed for two subsystems respectively, and the error of the system of finite time convergence condition has been carried on the proof. At the same time, the initial value of sliding-mode gain and the design method of time-varying gain are given. Then, based on the concept of equivalent control, the fault is detected and estimated. Finally, a simulation example is given to verify the effectiveness of the proposed fault estimation method.