Nonlinear robust observers for simultaneous state and fault estimation

A fault in the system operation is deemed to occur when the system practically experiences an abnormal condition, such as a malfunction in the actuators/sensors. This situation is happening very often in many control process such as: Chemical processes, Jet engine control, flight control, Robotics, temperature control, and etc. Faults can cause catastrophic damages to control systems. Therefore, reliability is one of the key requirements for process industries. Since many process control loops are utilized, the fault-free operation of these control loops is strictly required. For this purpose, effective model based Fault Detection and Isolation (FDI) has to be developed. On the other hand, effective control and monitoring of a system requires accurate information of internal behavior of the system. This internal behavior can be analyzed by system's states. Practically, in many real systems, state space variables are not fully available for measurements, or it is not practical to measure all of them, or it is too expensive to measure all these state space variables. Thus, one is faced with the problem of estimating system's state space variables. This can be done by constructing another dynamical system called state observer.;It is well known that two promising control strategies to cope with vastly uncertain control processes are Hinfinity Control and Sliding Mode Control. The robustness and simple implementation of these two control theories introduce them as strong practical control methods. Sliding mode observers are very successful to deal with uncertain faulty systems. Furthermore, in Robust FDI, the main objective is to design residuals that can distinguish faults from disturbances/uncertainties by reducing the effect of disturbances. However, in this thesis, we go one step further and we propose Robust Fault Reconstruction (RFR) by integrating Hinfinity filtering and Sliding Mode Control. It is also shown how adaptive control can improve the robustness of the observer based RFR by assuming that there is no information on the bound of a fault and nevertheless the observer can still reconstruct the fault effectively.;Another open problem in the context of FDI and RFR is due to systems with multiple faults at different system's components since it is often the case where actuators and also sensors suffer from faults during the course of the system's operation. Both actuators and sensors can suffer from faults either alone, at separate times or simultaneously. In this case, detection and reconstruction of all faults is highly important. The co-existence of unknown fault at both some sensor(s) and actuator(s) has not been addressed in any earlier design of fault reconstruction schemes. Thus, in this Thesis, inspired by the theory of singular systems, we aim at solving this open problem. Unknown Input Observers (UIOs) for estimation of unknown input and sensor fault are also studied by proposing a new UIO structure. The application of the proposed UIO for chaotic communication is also addressed. The class of system which will be considered throughout this thesis is Lipschitz nonlinear systems with fault and uncertainty. The reason behind focusing on Lipschitz system is that Lipschitz systems constitute a very important and wide class, since any nonlinear system with continuously differentiable nonlinearities can be locally expressed in this form. As a conclusion, we design novel observers (estimators) which benefits from the following main features: (1) are robust and insensitive to faults; (2) minimize the effect of disturbances on the state and fault estimation; (3) are able of detecting unknown behavior-type faults via adaptive gain adjustment; (4) can simultaneously estimate sensor(s) and actuator(s) fault; (5) have sensor fault reconstruction ability via the use of a reduced-order UIO.;The two critical problems stated above have motivated significant research work in the area of robust state and fault estimation. Fault reconstruction and estimation is regarded as a stronger extension to FDI since accurate fault estimation automatically implies fault detection. Fault reconstruction is excellent for directly detecting and isolating the malfunctions within a system by reviling which sensor or actuator is faulty and is useful for diagnosing incipient and small faults. Moreover, Fault reconstruction finds solid applications in Fault Tolerant Control Systems (FTC). Therefore, in this PhD thesis, we restrict our attention to design observers (estimators) that can simultaneously estimate the system states and faults. It is worth mentioning that both faults and disturbances considerably affect the state observation (estimation) and designing a robust observer which is insensitive to faults and disturbances is of great interest for achieving an accurate state estimation.