Exponential stability for uncertain neutral systems with Markov jumps

This paper deals with the global exponential stability problems for stochastic neutral Markov jump systems （MJSs） with uncertain parameters and multiple time-delays. The delays are respectively considered as constant and time varying cases, and the uncertainties are assumed to be norm bounded. By selecting appropriate Lyapunov-Krasovskii functions, it gives the sufficient condition such that the uncertain neutral MJSs are globally exponentially stochastically stable for all admissible uncertainties. The stability criteria are formulated in the form of linear matrix inequalities （LMIs）, which can be easily checked in practice. Finally, two numerical examples are exploited to illustrate the effectiveness of the developed techniques.     

基于间歇控制的不确定主从Lur’e系统的指数同步

研究了不确定主从Lur’e系统的问歇控制同步问题．运用Lyapunov稳定性理论和S-过程,首先推导出保证标称主从Lur’e系统指数同步的一个充分条件．在此基础上进一步得到了具有范数有界不确定性主从Lur’e系统鲁棒指数同步的结果．这些同步判据以矩阵不等式的形式给出．另外我们还提出了一种最小化性能指标的优化策略．最后,通过两个数值例子验证了结果的有效性．     

Event-triggered control for stochastic systems with multiple delays

This paper investigates the stabilization problem for continuous-time stochastic systems with multiple delays under continuous event-triggered mechanisms, of which both static case and dynamic case are considered individually. In order to avoid zeno phenomenon in every sample path, a suspension time after each successful execution is forced for our event-triggered mechanisms, resulting in intermittent detection of system states. Under such control strategy, we deduce mean square exponential stability of stochastic systems with multiple delays by means of Hanalay-type inequality and obtain a delay-dependent-based and less-conservative stabilization criterion without involving the upper bound of time delays. Besides, a co-design procedure is proposed for linear controller and event-triggered mechanisms. In the end, an illustrative example is presented to show effectiveness of the proposed co-design procedure and contrasts the system performance under static and dynamic event-triggered mechanisms.     

基于多特征融合的航空发动机剩余寿命预测

针对航空发动机剩余可用寿命(RUL)预测任务中代表性特征提取不充分导致RUL预测精度较低等问题, 提出了一种基于多特征融合的航空发动机RUL预测方法. 利用指数平滑法(ES)降低原始数据中的噪声干扰, 得到相对平稳的特征数据. 使用双向长短期记忆网络(Bi-LSTM)提取特征数据的时序特征, 利用多头注意力机制(Multi-attention)为时序特征赋予权重; 设计卷积长短期记忆网络(Conv-LSTM)提取特征数据的时空特征; 提取特征数据的手工特征并使用Softmax函数计算权重. 设计一个特征融合框架将上述特征进行融合, 然后通过全连接网络回归实现最终RUL预测. 使用C-MAPSS数据集对模型进行仿真验证, 与Bi-LSTM等模型进行对比, 模型RUL预测精度更高, 适应性更好.     

基于频谱增强和卷积宽度学习的音乐流派分类

针对频谱图对于音乐特征挖掘较弱、深度学习分类模型复杂且训练时间长的问题,设计了一种基于频谱增强和卷积宽度学习(CNNBLS)的音乐流派分类模型.该模型首先通过SpecAugment中随机屏蔽部分频率信道的方法增强梅尔频谱图,再将切割后的梅尔频谱图作为CNNBLS的输入,同时将指数线性单元函数(ELU)融合进CNNBLS的卷积层,以增强其分类精度.相较于其他机器学习网络框架, CNNBLS能用少量的训练时间获得较高的分类精度.此外, CNNBLS可以对增量数据进行快速学习.实验结果表明:无增量模型CNNBLS在训练400首音乐数据可获得90.06%的分类准确率,增量模型Incremental-CNNBLS在增加400首训练数据后可达91.53%的分类准确率.     

一类具比例时滞细胞神经网络反周期解的指数稳定性

本文针对一类具比例时滞细胞神经网络反周期解的全局指数稳定性进行研究.首先利用非线性变换将一类具比例时滞的细胞神经网络等价变换成一类具常时滞变系数的细胞神经网络.然后通过构造合适的时滞微分不等式和利用不等式技巧,得到了保证该系统反周期解的存在性和全局指数稳定性时滞依赖的充分条件.最后数值算例结果验证所得结论的正确性和与以往文献相比较低的保守性.     

求解变系数对流扩散反应方程的指数型高精度紧致差分方法

本文给出了一种数值求解变系数对流扩散反应方程的指数型高精度紧致差分方法.我们首先将模型方程变形,借助常系数对流扩散方程的指数型高精度紧致差分格式,采用残量修正法得到变系数对流扩散反应方程的指数型高精度紧致差分格式;并从理论上分析了当Pelect数很大时,本文格式达到四阶计算精度时网格步长的限制条件;离散得到的代数方程组可采用追赶法直接求解.数值实验结果与理论分析完全吻合,表明了本文格式对于边界层问题或大梯度变化的物理量求解问题具有的高精度和鲁棒性的优点.     

时变时滞NCS控制器的改进设计

建立时滞差分不等式,得到网络化控制系统(NCS)的指数稳定判据,基于该判据设计NCS状态反馈控制器。该设计以线性矩阵不等式(LMI)形式给出,不包含任何自由矩阵,通过Matlab中的LMI工具箱进行求解。数值仿真结果表明,该设计具有计算量小、求解方便、保守性弱的特点。     

不确定切换奇异时滞系统鲁棒指数容许性分析

讨论一类连续时间不确定切换奇异区间时变时滞系统的鲁棒指数容许性问题. 通过定义衰减率依赖李亚普诺夫函数并利用平均驻留时间法, 给出一个时滞区间依赖充分条件保证标称系统正则、无脉冲且均方指数稳定. 同时该准则也被推广至不确定系统. 本文获得的结论为连续时间切换奇异时滞系统的基本问题提供了一个解, 即识别切换信号使得切换奇异时滞系统正则、无脉冲且均方指数稳定. 数值例子说明本文结果的有效性.     

Exponential stability and exponential stabilization of singularly perturbed stochastic systems with time‐varying delay

In this paper, the problems of exponential stability and exponential stabilization for linear singularly perturbed stochastic systems with time‐varying delay are investigated. First, an appropriate Lyapunov functional is introduced to establish an improved delay‐dependent stability criterion. By applying free‐weighting matrix technique and by equivalently eliminating time‐varying delay through the idea of convex combination, a less conservative sufficient condition for exponential stability in mean square is obtained in terms of ε‐dependent linear matrix inequalities (LMIs). It is shown that if this set of LMIs for ε=0 are feasible then the system is exponentially stable in mean square for sufficiently small ε?0. Furthermore, it is shown that if a certain matrix variable in this set of LMIs is chosen to be a special form and the resulting LMIs are feasible for ε=0, then the system is ε‐uniformly exponentially stable for all sufficiently small ε?0. Based on the stability criteria, an ε‐independent state‐feedback controller that stabilizes the system for sufficiently small ε?0 is derived. Finally, numerical examples are presented, which show our results are effective and useful. Copyright © 2010 John Wiley & Sons, Ltd.