污水处理系统长期处于恶劣的环境当中，同时伴随着天气变化和未知进水等突发情况，会导致溶氧仪、曝气器等传感器和执行器故障。故障一旦发生，会严重影响污水处理的效率，造成巨大的经济损失。 为了削弱上述问题引起的出水水质恶化、工艺失效和安全隐患， 本文提出了传感器和执行器故障的估计与自愈控制方案， 保证污水处理系统安全、平稳、高效地运行。具体研究内容如下：

1.针对污水处理系统机理复杂、难以分析和控制的问题，基于活性污泥基准仿真模型，进行了数学建模研究。首先，根据污水处理过程的特点进行了建模分析；其次，利用加权递推最小二乘的方法进行系统辨识，得到了污水处理系统的离散T-S模糊模型；最后，通过仿真实验对模型进行验证，为后续故障估计与自愈控制奠定基础。

2.针对单观测器在进行传感器和执行器故障估计时稳定性条件的保守性高、适用范围小的问题，基于建立的离散T-S模糊模型，提出了双观测器的故障估计方法。首先，设计了广义观测器和自适应卡尔曼观测器分别对传感器故障和执行器故障进行估计，并进行了稳定性证明；其次，考虑溶氧仪和曝气器故障中的四种情况，基于仿真结果设计了误估计校正算法；最后，根据误估计校正算法进行了仿真实验，结果表明了本文所提方法能够实现传感器和执行器故障的准确估计。

3.针对当前控制方法仅考虑单一传感器或执行器故障的问题，基于建立的离散T-S模糊模型和双观测器得到的故障信息，同时考虑了传感器和执行器故障，进行了自愈控制研究。针对系统状态可观测的情形：分别设计了基于积分滑模和基于状态反馈的自愈控制器，通过设计模糊规则得到了模糊加权自愈控制器的输出，并进行了对比仿真实验，结果表明该自愈控制器具有更好的综合性能。针对系统状态不可测的情形：设计了基于动态输出反馈的自愈控制器 并证明了该自愈控制器的稳定性，通过对比仿真实验验证了该自愈控制器的有效性。

本文通过基于模型的故障估计和自愈控制方法，实现了对污水处理系统传感器和执行器故障的估计与自愈控制。本文的研究成果为自愈控制在污水处理系统的实际应用提供了理论基础，扩充了自愈控制的研究内容，具有重要的现实意义。

The wastewater treatment system has been in a harsh environment for a long time, accompanied by weather changes, unknown influent, and other emergencies, which will lead to the failure of sensors and actuators such as the dissolved oxygen meter and aerator. Once the failure occurs, it will seriously affect the efficiency of wastewater treatment and cause huge economic losses. In order to reduce the deterioration of effluent quality, process failure, and safety risks caused by the above problems, this paper proposes a sensor and actuator fault estimation and self-healing control scheme to ensure the safe, stable, and efficient operation of the wastewater treatment system. The specific research contents are as follows:

1.Aiming at the problem that the mechanism of wastewater treatment system is complex and difficult to analyze and control,  mathematical modeling research is carried out based on the activated sludge benchmark simulation model. Firstly, the modeling analysis is carried out according to the characteristics of wastewater treatment process. Secondly, the discrete-time T-S fuzzy model of the wastewater  treatment system is obtained by using the weighted recursive least squares method to identify the system. Finally, the  model is verified by simulation experiments, which lays a foundation for subsequent fault estimation and self-healing control.

2.Aiming at the problem that the stability condition of the single observer is highly conservative and the applicable range is small when the sensor and actuator faults are estimated, a fault estimation method for the double observers is proposed based on the established discrete-time T-S fuzzy model. Firstly, a descriptor observer and an adaptive Kalman observer are designed to estimate the fault of the sensor and the fault of the actuator , and the stability of observers is proven. Secondly, the  misestimate correction algorithm is designed based on the simulation results, considering the four  situations  involving the faults  of the dissolved oxygen meter and aerator. Finally, the simulation experiment is carried out according to the misestimate correction algorithm. And the results show that the proposed method can accurately estimate the sensor and actuator faults.

3.Aiming at the problem that the current control method only considers the single sensor or actuator fault, based on the established discrete-time T-S fuzzy model and the fault information obtained by double observers, the self-healing control is studied by considering both sensor and actuator faults. For the case where the system state can be observed, the self-healing controllers based on integral sliding mode and state feedback are designed respectively. The output of the fuzzy weighted self-healing controller is obtained by designing fuzzy rules. The comparison simulation results show that the self-healing controller has better comprehensive performance. For the case where the system state cannot be observed, a self-healing controller based on dynamic output feedback is designed and its stability is proven. The effectiveness of the self-healing controller is verified by comparison simulation experiments.

In this thesis, fault estimation and self-healing control of sensor and actuator in wastewater treatment system are realized  using model-based fault estimation and self-healing control methods. The research results of this paper provide a theoretical basis for the practical application of self-healing control in wastewater treatment system, and  expand the research content of  self-healing control,  which has important practical significance.