基于二阶鲁棒滑模观测器的民用涡扇发动机气路故障诊断

针对状态估计器在航空发动机气路参数估计中响应迟缓、鲁棒性不强等问题,以未知输入重构的思路,提出了一种基于Super-twisting滑模观测器的航空发动机气路故障诊断方法。通过将健康参数考虑为未知输入,设计滑模切换项重构健康参数的变化量,由于避免了状态估计器设计中健康参数导数为零的假设,本文的方法在处理突变故障时拥有更快的响应速度。针对鲁棒性问题,提出了一种新的故障向量增广形式,通过将扰动项增广至健康参数向量中,观测器的重构信号能够同时估计出健康参数变化量以及扰动项的大小,实现扰动与部件故障的解耦,从而避免了不确定项对健康参数估计结果的影响。本文建立了民用涡扇发动机包线范围内的线性变参数模型,通过不同故障模式下的数值仿真,并与状态估计器比较,验证了方法的有效性。结果表明,设计的滑模观测器具有小于0.5%的估计误差,有效地提高了气路健康参数的估计速度,增强了对不确定性的鲁棒性。

Gas Path Fault Diagnosis Based on Second-Order Robust Sliding Mode Observer for Civil Turbofan Engine

In view of the state estimator problems in aero-engine gas path parameter estimation such as slow response and weak robustness, a sliding mode observer design method based on the structure of super-twisting was proposed for aero-engine gas path fault diagnosis by reconstructing unknown inputs. By considering health parameters as unknown inputs, the switching term was designed to reconstruct the variation of health parameters. Since the assumption that the derivative of health parameters is zero in the state estimator design is avoided, the method proposed in this paper has a faster response speed when dealing with abrupt faults. A new form of an augmented fault vector is proposed for robustnes. By augmenting the disturbance term to the health parameter vector, the observer can estimate the variation of health parameters and also the magnitude of the disturbance by reconstructed signal at the same time and realize the decoupling of disturbance and component fault, thus avoiding the influence of uncertainty on the estimation results of health parameters. In this paper, a linear parameter varying model is established within the envelope range of a civil turbofan engine and the effectiveness of the method is verified by numerical simulation under different fault modes and comparison with the state estimator. The results show that the designed sliding mode observer has an estimation error of less than 0.5%, which effectively improves the estimation rate of gas path health parameters and enhances the robustness against uncertainties.