基于故障树和混沌粒子群算法的锻压机床故障诊断方法

锻压机床由于生产效率高和材料利用率高的特点,被广泛应用于各领域.然而,锻压机床发生故障时,其故障种类繁多、故障数据量大,所以对锻压机床故障源的快速、准确诊断较困难.针对该问题,文章提出一种将故障树分析法和混沌粒子群算法相融合的方法,对锻压机床的故障源进行故障诊断.该方法是先通过故障树分析法对锻压机床的故障进行分析从而得到故障模式及其故障概率,然后由得到的故障模式和已知的故障维修经验分析归纳出故障模式的学习样本,再根据得到的故障概率运用混沌粒子群算法的遍历性快速、准确地诊断出锻压机床发生故障的精确位置.文章提出的方法以锻压机床的伺服系统为例进行了故障诊断实验,将该实验结果与遗传算法、粒子群算法进行对比.实验结果表明,文章的算法在锻压机床伺服系统的故障诊断中准确度更高、速度更快.     

基于多项式混沌拓展模型和bootstrap的结构可靠性分析方法

多项式混沌拓展(polynomial chaos expansion,PCE)模型现已发展为全局灵敏度分析的强大工具,却很少作为替代模型用于可靠性分析。针对该模型缺乏误差项从而很难构造主动学习函数来逐步更新的事实,在结构可靠性分析的框架下提出了基于PCE模型和bootstrap重抽样的仿真方法来计算失效概率。首先,对试验设计(experimental design)使用bootstrap重抽样步骤以刻画PCE模型的预测误差;其次,基于这个局部误差构造主动学习函数,通过不断填充试验设计以自适应地更新模型,直到能够精确地逼近真实的功能函数;最后,当PCE模型具有足够精确的拟合、预测能力,再使用蒙特卡洛仿真方法来计算失效概率。提出的平行加点策略既能在模型更新过程中找到改进模型拟合能力的"最好"的点,又考虑了模型拟合的计算量;而且,当失效概率的数量级较低时,PCE-bootstrap步骤与子集仿真(subset simulation)的结合能进一步加速失效概率估计量的收敛。本文方法将PCE模型在概率可靠性领域的应用从灵敏度分析延伸到了可靠性分析,同时,算例分析结果显示了该方法的精确性和高效性。     

Partial approximate boundary synchronization for a coupled system of wave equations with Dirichlet boundary controls

In this paper, we propose the concept of partial approximate boundary synchronization for a coupled system of wave equations with Dirichlet boundary controls, and make a deep discussion on it. We analyze the relation-ship between the partial approximate boundary synchronization and the partial exact boundary synchronization, and obtain sufficient conditions to realize the partial approximate boundary synchronization and necessary conditions of Kalman's criterion. In addition, with the help of partial synchronization decomposition, a condition that the approximately synchronizable state does not depend on the sequence of boundary controls is also given.     

Synchronization of chaotic dynamical network with unknown generally time-delayed couplings via a simple adaptive feedback control

In this paper, based on the principle of functional differential equations, a simple adaptive feedback controller is proposed to synchronize the network with unknown generally time-delayed coupling functions. Unlike the other control schemes, we design the controllers by the output of each node. The advantage of this scheme is that we need not to know the concrete structure of coupling functions or a solution of node system in advance. Moreover, an illustrative example is considered and numerical simulations are performed to demonstrate the effectiveness of this control scheme. The numerical simulation results show that our method is valid and robust against the weak noise.     

时变耦合变时滞非自治神经网络模型的全局同步性分析

利用Lyapunov泛函方法,对一类时变线性耦合神经网络模型的全局同步性进行了研究．在去掉耦合矩阵的对称性、不可约性和扩散耦合限制的基础上,得到了确保耦合时滞神经网络模型全局同步的充分性条件．所得结果仅依赖于系统中的参数,条件易于验证且不必求矩阵的特征值．     

混沌时滞随机神经网络同步控制

研究了一类混沌时滞随机神经网络同步控制问题．采用更具一般性的时滞反馈控制器,通过巧妙地构造Lyapunov数,分别得到了均方指数同步和均方渐近同步两个判别准则．仿真例子表明,新准则是有效的．     

算子空间上的左乘映射

Let X be a separable infinite dimensional Banach space and B(X) denote its operator algebra,the algebra of all bounded linear operators T : X → X.Define a left multiplication mapping LT : B(X) → B(X) by LT (V ) = T V,V ∈ B(X).We investigate the connections between hypercyclic and chaotic behaviors of the left multiplication mapping LT on B(X) and that of operator T on X.We obtain that LT is SOT-hypercyclic if and only if T satisfies the Hypercyclicity Criterion.If we define chaos on B(X) as SOT-hypercyclicity plus SOT-dense subset of periodic points,we also get that LT is chaotic if and only if T is chaotic in the sense of Devaney.     

Bifurcations of resonant solutions and chaos in physical pendulum equation with suspension axis vibrations

This paper is a continuation of ＂Complex Dynamics in Physical Pendulum Equation with Suspension Axis Vibrations＂[1].In this paper,we investigate the existence and the bifurcations of resonant solution for ω0：ω：Ω ≈ 1：1：n,1：2：n,1：3：n,2：1：n and 3：1：n by using second-order averaging method,give a criterion for the existence of resonant solution for ω0：ω：Ω ≈ 1：m：n by using Melnikov＇s method and verify the theoretical analysis by numerical simulations.By numerical simulation,we expose some other interesting dynamical behaviors including the entire invariant torus region,the cascade of invariant torus behaviors,the entire chaos region without periodic windows,chaotic region with complex periodic windows,the entire period-one orbits region;the jumping behaviors including invariant torus behaviors converting to period-one orbits,from chaos to invariant torus behaviors or from invariant torus behaviors to chaos,from period-one to chaos,from invariant torus behaviors to another invariant torus behaviors;the interior crisis;and the different nice invariant torus attractors and chaotic attractors.The numerical results show the difference of dynamical behaviors for the physical pendulum equation with suspension axis vibrations between the cases under the three frequencies resonant condition and under the periodic/quasi-periodic perturbations.It exhibits many invariant torus behaviors under the resonant conditions.We find a lot of chaotic behaviors which are different from those under the periodic/quasi-periodic perturbations.However,we did not find the cascades of period-doubling bifurcation.     

A random transport-diffusion equation

In this paper we study the integral curve in a random vector field perturbed by white noise. It is related to a stochastic transport-diffusion equation. Under some conditions on the covariance function of the vector field, the solution of this stochastic partial differential equation is proved to have moments. The exact p-th moment is represented through integrals with respect to Brownian motions. The basic tool is Girsanov formula.     

Chaos synchronization for a class of nonlinear oscillators with fractional order

In this paper, a special kind of nonlinear chaotic oscillator, the Qi oscillator, is studied in detail. Since such systems are shown to possess a relatively wide spectral bandwidth, it is considerably beneficial to practical engineering in the secure communication field. The chaos synchronization problem of the fractional-order Qi oscillators coupled in a master-slave pattern is examined by applying three different kinds of methods: the nonlinear feedback method, the one-way coupling method and the method based on the state observer. Suitable synchronization conditions are derived by using the Lyapunov stability theory, and most importantly, a sufficient and necessary synchronization condition for the case with fractional order between 1 and 2 is presented. Results of numerical simulations validate the effectiveness and applicability of the proposed schemes.