受扰时滞系统抗干扰故障重构研究

The systems are often affected by a variety of different types of disturbances and time delays. These existed disturbances and time delays will affect the design of fault reconstruction method, so disturbances estimation and attenuation, the effective use of delay information is a key issue to improve the accuracy of fault reconstruction. However, the current methods mostly focused on plants with a single disturbance, which did not fully consider the characteristics of multi-source disturbances on system performance.And the system input time delay and output time delay information is not fully utilized. Thereby, they hinder further improvement of fault reconstruction accuracy.. In this project, fault reconstruction problem is studied for a class of nonlinear time-delay systems. Firstly, the various features of the system with disturbances are analysed, and these disturbances are modelled via different classification methods. Secondly, for the modelled different disturbances, the design of new online observers estimated to improve the accuracy and speed of response disturbances observed. Then, for a single fault, multiple faults, faults and disturbances coupling, etc., some adaptive fault reconstruction methods are established. Meanwhile, in order to investigate the influences on time delays to fault reconstruction accuracy, time delays also are considered in the designed methods. Finally, fault reconstruction of the six-rotor aircraft is given to demonstrate the effectiveness of the proposed methods. The research tasks of this project will further provide effective theory and methods for high-precision disturbance rejection fault reconstruction of systems with disturbances and time-delays.

系统往往受到多种不同类型干扰及时滞的影响，而干扰、时滞的存在会影响故障重构方法设计，于是干扰的抵消和抑制、时滞信息有效利用是提高故障重构精度的关键问题。然而目前抗干扰方法大多针对单一类型干扰，没有充分考虑多源干扰的特征对系统性能的影响，同时对系统的输入、输出时滞信息考虑也不充分，因而阻碍了故障重构精度的进一步提高。. 本项目以一类非线性时滞系统为研究对象，首先分析系统中存在的各种干扰模型特征，并对其进行分类建模。其次针对不同干扰，设计新型观测器组合进行在线估计，提高干扰观测的精度和响应速度。然后，针对单一故障，多故障，故障与扰动耦合等情况，建立自适应抗干扰故障重构方法。并在此基础上进一步分析时滞对故障重构精度的影响。最后将研究结果应用于小型六旋翼飞行器的姿态控制。该项目的研究将为受扰时滞系统的高精度抗干扰故障重构提供切实有效的理论和方法。

本项目针对一类受扰非线性时滞系统，研究其在干扰、时滞情况下，基于观测器理论框架的故障诊断新方法。主要内容包括：1）针对含单一执行器故障、建模误差、时滞的随机分布系统，提出了一种基于概率密度函数、建模误差界和时滞界等多源信息的故障检测、诊断方法，提高了故障诊断的时滞敏感性。2）针对执行器和传感器同时发生故障的随机分布系统，提出了一种新型的基于组合自适应观测器的在线故障估计方案。基于虚拟补偿技术的增广控制器用于实现故障补偿并同时实现概率分布形状跟踪。3）针对故障和多源干扰同时发生的马尔科夫跳跃系统，提出了未知输入干扰和输出噪声共同影响下未知输入观测器的设计方法。4）以挠性航天器为例，研究了执行器部分失效、完全失效情况下的故障估计问题。对于部分失效情况，提出了失效因子鲁棒被动学习法和自适应学习法。对于执行器完全失效时，在某些干扰特性已知的情况下，提出了一种基于干扰观测器与自适应控制相结合的姿控系统抗干扰方法。以六旋翼无人机为例，对执行机构失效时，构建辅助输出，设计观测器，实现故障的在线估计。项目共发表期刊论文26篇，其中SCI收录论文8篇，JCR一区论文6篇。获授权专利5件，转让2件。其中发明专利2件，实用新型专利3件，软件著作权登记2项。获江苏省教育科学研究成果二等奖1项，项目负责人入选江苏省333工程中青年科技领军人才。