时滞随机模糊系统的有限时间故障估计与容错控制研究

故障在生产生活中往往不可避免。当故障发生时，需要快速实现精准故障估计与有效容错控制，进而保证系统仍能稳定运行。本项目针对实际中常见但目前研究较少的多时滞随机模糊半马尔科夫跳变系统，研究了有限时间间歇性故障估计与容错控制，主要研究内容如下：为提高故障估计性能，构建多积分观测器，在有限时间内同时估计间歇性传感器和执行器故障，利用间歇性执行器故障微分信息多次迭代故障估计值；为提高稳定性能，利用间歇性故障补偿器和干扰补偿器设计容错控制器，完成对原闭环系统的有限时间容错控制；在状态和输入受限下构建观测器和控制器，分别通过多次迭代故障估计值以及设计间歇性故障补偿器与干扰补偿器提高有限时间故障估计与容错控制性能。本项目构建了多时滞随机模糊半马尔科夫跳变系统理论体系，提升了有限时间故障估计与容错控制在实际中的应用性。通过本项目研究，实现了目标系统的稳定运行，进一步保障了人员生命安全，避免了经济损失。

Faults are often unavoidable in production and life. When the faults occur, it is necessary to quickly achieve accurate fault estimation and effective fault-tolerant control to ensure that the system still runs stably. In this project, finite-time intermittent fault estimation and tolerant control are studied for stochastic fuzzy semi-Markovian jump systems with multiple time delays which are common in practice but less studied at present. The main researches are as follows: A multi-integral observer is built, which can perform finite-time estimation of intermittent sensor and actuator faults simultaneously. And the differential information of the intermittent actuator faults is utilized to iterate the fault estimation values for many times to improve the performance of fault estimation. A fault-tolerant controller is designed by using intermittent fault compensators and disturbance compensators, and the finite-time fault-tolerant control of the original closed-loop system is completed to improve the stability performance. Under the condition of state and input constraints, an observer and a controller are constructed. The performance of finite-time fault estimation and tolerant control is improved by iterating fault estimation values and designing intermittent fault compensators and disturbance compensators, respectively. This project constructs theoretical system of stochastic fuzzy semi-Markovian jump systems with multiple time delays to improve the application of finite-time fault estimation and tolerant control in practice. The stable operation of the objective system is realized, the life safety of personnel is further guaranteed, and the economic loss is avoided by the research of this project.