造纸机网前箱系统鲁棒故障诊断及容错控制

Robust Fault Diagnosis and Fault Tolerant Control of Paper Machine Headbox System

在造纸机网前箱控制系统中，传感器失效时，控制系统的性能严重恶化，而常规故障诊断方法又要求准确数学模型，针对这一问题本文提出将自适应技术引入常规故障诊断方案，形成鲁棒故障诊断策略，实现了存在被控对象建模误差及参数慢时变情况下，造纸机网前箱控制系统的可靠故障诊断及状态估计．仿真实验证实了所给方法的有效性．

For the paper machine headbox control system, once a sensor failure occurs, the performance of control system will degrade, and the usual dedicated observer scheme (DOS) is restricted by the accurate mathematics model. For this problem, a robust fault diagnosis scheme combining adaptive observer with DOS is proposed. The proposed method can achieve a reliable fault diagnosis and state estimation of paper machine headbox control system under circumstances of existing plant model error and parameter drift. Simulation Results shows that the method is effective.