基于LSSVM的卫星姿态控制系统故障诊断

提出了一种基于最小二乘支持向量机(LSSVM)非线性观测器的卫星姿态控制系统故障诊断方法。与标准的支持向量机回归算法相比,最小二乘支持向量机回归算法收敛速度快,适用于在线训练。该方法利用其回归逼近非线性函数的能力,设计基于最小二乘支持向量机的非线性系统状态观测器,在线训练最小二乘支持向量机回归,并用于估计卫星姿态控制系统故障。最后,通过仿真验证了这种方法可以快速准确地估计出卫星姿态控制系统的故障。     

最小二乘支持向量机在故障诊断中的应用

为了提高机械设备故障诊断的精度，将小波包分析与最小二乘支持向量机进行了有机的结合。首先对故障信号功率谱进行小波分解，简化了故障特征向量的提取。然后提出了一种基于最小二乘支持向量机的故障诊断模型，用二次损失函数取代支持向量机中的不敏感损失函数，将不等式约束条件变为等式约束，从而将二次规划问题转变为线性方程组的求解，用最小二乘法实现了支持向量机算法，并提出对核函数的σ参数进行动态选取，提高了诊断的准确率。仿真结果表明该模型具有较强的非线性处理和抗干扰能力。     

基于最小二乘支持向量机的故障诊断方法*

提出了一种小波包分析与最小二乘支持向量机相结合的机械设备故障诊断模型.首先对故障信号功率谱进行小波分解,简化了故障特征向量的提取,然后采用最小二乘支持向量机进行故障诊断.在该模型中,用二次损失函数取代支持向量机中的不敏感损失函数,将不等式约束条件变为等式约束,从而将二次规划问题转换为线性方程组的求解,并提出对核函数的σ参数进行动态选取.仿真结果表明:该模型可以取得较高的故障诊断准确率.     

一类非线性系统基于最小二乘支持向量机的自适应H∞控制

提出一类非线性不确定系统基于最小二乘支持向量机的白适应H∞控制方法．该方法基于最小二乘支持向量机估计对象的未知非线性函数,并给出了最小二乘支持向量机权向量和偏移值的在线学习规则．引入H∞控制器用于减弱外部干扰及最小二乘支持向量机近似误差对输出误差的影响．利用李亚普诺夫理论证明了整个闭环系统一致最终有界稳定．仿真结果表明了该方法的有效性．     

基于IMF能量矩和Bayes-LSSVM的滚动轴承诊断

针对传统方法对滚动轴承故障特征提取效果尚有局限和最小二乘支持向量机分类器的参数不易确定,从而降低了故障诊断的准确性的问题,提出基于本征模函数能量矩和贝叶斯框架下的最小二乘支持向量机实现滚动轴承的故障诊断。在该方法中,通过经验模态分解将原始信号分解为多个本征模函数,之后将本征模函数作时间轴的积分,得到本征模函数能量矩特征故障向量。采用贝叶斯推理方法进行三级分层推断,解决最小二乘支持向量机分类器的参数具有任意性和不确定性的问题,实现参数优化。对滚动轴承的仿真结果表明,该方法能对故障进行有效、准确的诊断,诊断正确率达到98.75%。     

基于支持向量机的模拟电路故障诊断

针对模拟电路的固有复杂性及其传统故障检测方法存在延时大和正确识别率低的问题,提出基于最小二乘支持向量机和Volterra级数的故障诊断方法。采用Volterra级数频域核对电路故障特征进行提取,利用最小二乘支持向量机进行模态分类,最终完成故障诊断。仿真结果表明,该方法与BP神经网络相比提高了系统故障辨识能力与系统故障诊断速度。     

支持向量机的多层动态自适应参数优化

首先提出了基于多层动态自适应搜索技术的最小二乘支持向量机参数优化方法，然后采用最小二乘支持向量机对典型非线性控制系统的辨识进行了研究．辨识结果表明，最小二乘支持向量机可以用于非线性控制系统辨识，多层动态自适应搜索方法确定了最优支持向量机参数，从而获得精确的非线性控制系统辨识结果．     

基于最小二乘支持向量机的汽轮机故障诊断

提出一种小波包分析与最小二乘支持向量机相结合的汽轮机故障诊断模型．对故障信号功率谱进行小波分解，简化了故障特征向量的提取．用二次损失函数取代支持向量机中的不敏感损失函数，将不等式约束条件变为等式约束．从而将二次规划问题转变为线性方程组的求解．选用RBF函数作为核函数。并提出对核函数的参数进行动态选取。提高了诊断的准确率．仿真结果表明该模型具有较强的非线性处理和抗干扰能力．     

基于LS-SVM的船舶航向模型预测控制

针对船舶动态性能具有较强的非线性、大惯性及时变性的特点,采用具有RBF核函数的最小二乘支持向量机(LS-SVM),利用其可以任意逼近非线性模型的良好特性实现对船舶模型有效辨识,得到船舶航向控制系统的非线性逼近模型,并将其与模型预测控制方法相结合,将最小二乘支持向量机辨识得到的系统模型作为预测模型,并将系统模型进行线性化并用线性预测控制方法求得解析的控制律,实现对船舶航向的预测控制,达到良好航向保持目的。仿真结果表明,最小二乘支持向量机降低了计算复杂度,且有较快计算速度,在小样本情况下具有良好的泛化能力;基于最小二乘支持向量机的船舶航向预测控制系统对外界干扰及模型参数摄动均具有较好的适应能力以及良好的控制性能。     

基于未知输入观测器的非线性广义系统执行器故障检测和重构

针对一类含有未知扰动广义非线性系统的执行器故障,本文提出一种重构算法。首先设计未知输入观测器对干扰鲁棒,作为故障检测观测器。检测到发生故障后,通过提出含有误差比例项和积分项的故障估计算法,形成自适应观测器,实现准确快速地估计故障,同时估计状态变量。根据李雅普诺夫稳定理论给出估计误差一致最终有界的充分条件。最后仿真验证该类观测器和重构算法的有效性。     

具有自适应调节律的非线性系统执行器故障检测和隔离方法

针对受到外部干扰的非线性系统,讨论了基于观测器的执行器故障检测和隔离方法.首先,通过引入一个对Lipschitz非线性项Lipschitz常数自适应调节的微分调节项,使得观测器具有自适应性,从而使观测器设计具有无须知道Lipschitz常数大小的优点;然后,通过一滑模控制项来抑制干扰,使观测器具有鲁棒性,并在此基础上,结合多观测器故障隔离的思想,提出了执行器故障检测和隔离方法;最后,通过对一个七阶飞行器实际模型的仿真,表明了该方法的实用性.     

Actuator Fault Reconstruction for Systems with Monotone Nonlinearities

This paper considers observer‐based fault reconstruction for systems with monotone nonlinearities. The nonlinear term in the observer error dynamics satisfies a sector property. Using Lyapunov redesign techniques, a continuous nonlinear error feedback is designed according to the slope property of the nonlinear term to stabilize the error dynamics. Hence, the convergence of observer error is independent of the Lipschitz constant. If the observer error converges to zero asymptotically, then the continuous error feedback can be used to reconstruct the faults. As an extension, an adaptive scheme is developed for systems where the arguments of nonlinear functions are perturbed by unknown parameters. Finally, simulations of an electric driving system are presented to show the effectiveness of the schemes.     

Full-order observer-based actuator fault detection and reduced-order observer-based fault reconstruction for a class of uncertain nonlinear systems

This paper considers observer-based actuator fault detection and reconstruction problems for uncertain nonlinear systems. Based on a kind of full-order observer which is robust to disturbances but sensitive to actuator faults, a single detection observer is constructed to produce a residual which can be used to alarm the occurrence of the actuator faults when at least one actuator fault occurs indeed. The full-order observer is adaptive one because an adaptation law which can adjust the Lipschitz constant of Lipschitz term is introduced. For this reason, the Lipschitz constant can be unknown in our design. After this, a kind of reduced-order observer is developed by choosing a special observer gain matrix. Based on the reduced-order observer, we provide a kind of unknown information estimating method which can be used to not only reconstruct the actuator faults but also estimate the disturbances of the system. In simulation, a real model of the seventh-order aircraft is used to illustrate the effectiveness of the proposed methods.     

基于参数估计的一类非线性系统故障诊断算法

针对系统模型的不确定性、未知输入扰动和非线性特性, 提出一类非线性系统参数估计的故障诊断算法. 构造系统故障诊断观测器, 采用Lyapunov 稳定性定理验证观测器的稳定性, 通过Barbalat 引理证明满足故障诊断观测器为渐近稳定的表征故障参数的参数估计, 并总结了设计算法流程. 仿真结果表明, 所提出算法具有快速收敛性, 对一类非线性系统诊断效果较好.

     

Fault Detection and Diagnosis Based on Modeling and Estimation Methods

This paper investigates the problem of fault detection and diagnosis in a class of nonlinear systems with modeling uncertainties. A nonlinear observer is first designed for monitoring fault. Radial basis function (RBF) neural network is used in this observer to approximate the unknown nonlinear dynamics. When a fault occurs, another RBF is triggered to capture the nonlinear characteristics of the fault function. The fault model obtained by the second neural network (NN) can be used for identifying the failure mode by comparing it with any known failure modes. Finally, a simulation example is presented to illustrate the effectiveness of the proposed scheme.     

基于广义未知输入观测器的鲁棒故障估计方法

当干扰存在时,有效地估计故障且放松故障的限制条件需要进一步的研究,为此针对含未知干扰的非线性连续系统的鲁棒故障估计问题提出一种广义未知输入观测器方法。首先,将执行器故障向量和传感器故障向量与原系统状态向量组成广义系统,放松对故障类型的限制,对此广义系统设计未知输入观测器解耦干扰,保证鲁棒性的同时估计出状态变量、执行器故障及其一阶微分和传感器故障。然后通过解线性矩阵不等式（LMI）给出估计误差渐近收敛的条件。最后,在MATLAB 的simulink平台上用三叶片水平轴风力模型仿真验证本文观测器的故障估计有效性鲁棒性。     

用B-样条神经网络设计非线性观测器

对线性部分已知、非线性部分未知的一类非线性系统,提出了一种新的状态观测器 的设计方法.首先针对线性部分设计线性观测器,随后在线性观测器中加入非线性补偿项.该 补偿项先由"反卷积"的方式确定,再用B-样条神经网络拟合.对三个非线性系统设计了观测 器,通过与已有的解析方法进行比较,说明了该方法的优越性.     

Robust fault diagnosis scheme in a class of nonlinear system based on UIO and fuzzy residual

In this study, a novel robust fault diagnosis scheme is developed for a class of nonlinear systems when both fault and disturbance are considered. The proposed scheme includes both component and sensor fault with nonlinear system that transferred to nonlinear Takagi-Sugeno (T-S) model. It considers a larger category of nonlinear system when fuzzification is used for only nonlinear distribution matrices. In fact the proposed method covers nonlinear systems could not transform to linear T-S model. This paper studies the problem of robust fault diagnosis based on two fuzzy nonlinear observers, the first one is a fuzzy nonlinear unknown input observer (FNUIO) and the other is a fuzzy nonlinear Luenberger observer (FNLO). This approach decouples the faulty subsystem from the rest of the system through a series of transformations. Then, the objective is to design FNUIO to guarantee the asymptotic stability of the error dynamic using the Lyapunov method; meanwhile, FNLO is designed for faulty subsystem to generate fuzzy residual signal based on a quadratic Lyapunov function and some matrices inequality convexification techniques. FNUIO affects only the fault free subsystem and completely removes any unknown inputs such as disturbances when residual signal is generated by FNLO is affected by component or sensor fault. This novelty and using nonlinear system in T-S model make the proposed method extremely effective from last decade literature. Sufficient conditions are established in order to guarantee the convergence of the state estimation error. Thus, a residual generator is determined on the basis of LMI conditions such that the estimation error is completely sensitive to fault vector and insensitive to the unknown inputs. Finally, an numerical example is given to show the highly effectiveness of the proposed fault diagnosis scheme.