城市污水处理过程机理复杂、工作环境恶劣且需要连续运行，这导致设备容易发生故障，从而影响出水水质、造成经济损失和安全事故。因此，为了保证污水处理系统安全、连续地运行，本文以活性污泥法城市污水处理过程为研究对象，针对执行器和传感器加性故障发生时故障估计与自愈控制问题展开研究。具体研究内容如下：
  1.针对污水处理系统机理模型难于分析的问题，采用T-S模糊方法建立数学模型。首先，确定操作变量和被控变量分别为第5反应池内氧传递系数和溶解氧浓度；其次，结合模糊网格法和递推最小二乘法建立了T-S模糊模型；最后，设计了测试信号对模型进行仿真验证，结果表明模型精度高且具有泛化能力，可以用于表征污水处理系统。
  2.针对污水处理过程中执行器老化、堵塞等加性故障，基于T-S模型进行了故障估计与自愈控制研究。首先，将执行器故障增广为系统状态的一部分，设计了自适应比例积分观测器并利用LMI方法求解增益矩阵；其次，基于故障补偿思想，采用滑模控制方法，设计了自愈控制器并利用Lyapunov原理证明了系统稳定性；最后，基于MATLAB/Simulink平台设计了污水处理执行器中曝气器故障时的仿真实验，结果表明该方法能够在执行器故障发生时及时进行故障估计和自愈控制，保证系统稳定运行。
  3.针对污水处理过程中传感器漂移、腐蚀等加性故障，基于T-S模型进行了故障估计与自愈控制研究。首先，构建了广义系统将传感器故障估计问题转化为执行器故障估计问题，设计了广义滑模观测器并利用LMI方法求解增益矩阵；其次，利用故障估计值校正系统输出，设计了基于滑模鲁棒控制的自愈控制器并利用Lyapunov原理与LMI技术求解状态反馈增益；最后，设计了传感器中溶氧仪故障时的MATLAB仿真实验，结果表明该方法能够在传感器故障发生时及时进行故障估计和自愈控制。
  本文通过设计观测器和自愈控制器等方法，实现了对城市污水处理过程执行器故障或传感器故障的估计与自愈控制。本文的研究成果可以进一步完善城市污水处理系统自愈控制的理论基础，推动自愈控制在污水处理系统中的实际应用，为因执行器或传感器故障造成出水水质恶化的问题提供了解决方案，具有重要的现实意义。

  The municipal wastewater treatment process has complex mechanism, harsh working environment and continuous operation, which makes equipment prone to failure, thus affecting the effluent quality, causing economic losses and safety accidents. Therefore, in order to ensure the safe and continuous operation of the wastewater treatment system, this paper takes the activated sludge method municipal wastewater treatment process as the research object, and studies the problem of fault estimation and self-healing control when additive faults of actuators and sensors occur. The specific research contents are as follows:
  1.Aiming at the problem that the mechanism model of the wastewater treatment system is difficult to analyze, the T-S fuzzy method is used to establish the mathematical model. First, according to the system characteristics, the input and output variables are selected and the excitation signal is designed; secondly, the T-S fuzzy model of the wastewater treatment system is established by combining the fuzzy grid method and the recursive least square method; finally, the test signal is designed to carry out the identification model. Simulation results show that the model has high accuracy and generalization ability, and can be used to characterize the wastewater treatment system.
  2.Aiming at the problem that the mechanism model of the wastewater treatment system is difficult to analyze, the T-S fuzzy method is used to establish the mathematical model. First, according to the system characteristics, the input and output variables are selected and the excitation signal is designed; secondly, the T-S fuzzy model of the wastewater treatment system is established by combining the fuzzy grid method and the recursive least square method; finally, the test signal is designed to carry out the identification model. Simulation results show that the model has high accuracy and generalization ability, and can be used to characterize the wastewater treatment system.
  3.Aiming at sensor additive faults such as drift and corrosion in wastewater treatment process, fault estimation and self-healing control are studied based on T-S model. First, a generalized system is constructed to transform the sensor fault estimation problem into the actuator fault estimation problem, a generalized sliding mode observer is designed, and the gain matrix is solved by LMI; Secondly, using the fault estimation value to correct the system output, a self-healing controller is designed and the state feedback gain is solved by Lyapunov and LMI method; Finally, the MATLAB simulation experiment when the dissolved oxygen meter fault is designed, and the results show that the method can perform fault estimation and self-healing control in time when sensor fault occurs.
  In this paper, by designing observers and self-healing controllers, the estimation and self-healing control of actuator faults or sensor faults in municipal wastewater treatment process are realized. The research results of this paper can further improve the theoretical basis of self-healing control of municipal wastewater treatment system, promote the practical application of self-healing control in wastewater treatment system, and provide a solution to the problem of water quality deterioration caused by actuator or sensor failure, which has important practical significance.