THE OPTIMALITY CONDITIONS OF D-PSEUDOCONVEX PROGRAMMING WITH OPERATOR CONSTRAINTS

The problem of D-pseudoconvex programming with operator con-straints is discussed in this paper.Some new optimality conditions are presentedand an error made by V.Barbu in [4] is pointed out.     

FINITE DIMENSIONAL BEHAVIOR FOR THE DISSIPATIVE GENERALIZED SYMMETRIC REGULARIZED LONG WAVE EQUATIONS

This paper deals with the asymptotic behaviour of solutions for the generalized symmetric regularized long wave equations with dissipation term. We first show the existence of global weak attractors for the periodic initial value problem of this equations in H2 x H1. And then by an energy equation and an idea of Ghidaglia and Guo, we conclude that the global weak attractor is actually the global strong attractor for S(t) in H2(2) x H1(M).The finite dimensionality of the global attractor is also established.     

THE SPLITTING EXTRAPOLATION FOR FINITEELEMENT METHOD

1.IntroductionDespitethesignificantprogressmadeindigitalcomputersandcomputationalmethodsoaferthelasttwodecades,computationalworkaswellascomputerstoragerequiremelltintheappliedfieldissogreatthatsuper-computersarenotabletoaft'ord.111recentyears)thehotpoilltsincomputationalmathematicsarehowtodevelopanealcomPlltationalmethodwhichwouldfulfilthefollowingrequirelllents;1)ItshouldbeaparallelcomputationalIned]'>(lwhichcan11eimplementedonrnultiprocessorcomputersreqlliringminimalcommunicationanloll!{pro(…     

关于认知影响图方法的研究

在进一步发展认知图、影响图理论的基础上，提出了融合二者于一体的认知影响图理论，建立了两种构造认知影响图方法：诊断法和时间序列法，解决了实际认知过程中，按时间序列出现事件经常与人类诊断知识方向矛盾的问题：并根据在思考问题过程中，同一领域专家的认知方式不同，分别采取因果推理及诊断推理的特点，提出了一种能有效地获取上述两种知识的衍值方法     

THE MODIFIED METHOD OF CHARACTERISTICS WITH FINITE ELEMENT OPERATOR—SPLITTING PROCEDURES FOR COMPRESSIBLE MULTICOMPONENT DISPLACEMENT PROBLEM

For the three-dimensional compressible multicomponent displacement problem we put forward the modified method of characteristics with finite element operator-splitting procedures and make use of operator-splitting,characteristic method,calculus of variations,energy method,negative norm estimate,two kinds of test functions and the theory of prior estimates and techniques.Optimal order estimates in L^2 norm are derived for the error in the approximate solution.These methods have been successfully used in oil-gas resources estimation,enhanced oil recovery simulation and seawater intrusion numerical simulation.     

Fast-moving target tracking based on mean shift and frame-difference methods

The mean shift tracker has difficulty in tracking fast moving targets and suffers from tracking error accumulation problem.To overcome the limitations of the mean shift method,a new approach is proposed by integrating the mean shift algorithm and frame-difference methods.The rough position of the moving target is first located by the direct frame-difference algorithm and three-frame-difference algorithm for the immobile camera scenes and mobile camera scenes,respectively.Then,the mean shift algorithm is used to achieve precise tracking of the target.Several tracking experiments show that the proposed method can effectively track first moving targets and overcome the tracking error accumulation problem.     

神经网络对新增样本的学习算法

对新增样本的快速学习而又不损失原有样本的记忆，是自适应在线系统的要求．本文提出了一种基于对节点激励函数线性化的逐层优化学习算法，为防止由于线性化而造成较大的误差，在损失函数中加入了惩罚项．该算法在每次迭代中，权值矩阵可以显式表达出来．算例仿真表明了该方法可行有效     

New algorithm for robust H2/H∞ filtering with error variance assignment

We consider the robust H2/H∞ filtering problem for linear perturbed systems with steady-state error variance assignment. The generalized inverse technique of matrix is introduced, and a new algorithm is developed. After two Riccati equations are solved, the filter can be obtained directly, and the following three performance requirements are simultaneously satisfied: The filtering process is asymptotically stable; the steady-state variance of the estimation error of each state is not more than the individual prespecified upper bound; the transfer function from exogenous noise inputs to error state outputs meets the prespecified H∞ norm upper bound constraint. A numerical example is provided to demonstrate the flexibility of the proposed design approach.     

基于神经网络的轮廓误差控制仿真研究

运动控制系统不同轴的动态特性的差异是造成系统轮廓误差较大的主要原因。为系统每个轴设计一个高性能的前馈补偿器，可以改善每个轴的动态特性，使不同轴相互匹配，从而残小系统的轮廓误差。由于前馈补偿器的传递函数通常具有二阶或二阶以上微分环节，物理实现较困难，本文利用神经网络辨识系统的控制对象逆模型作为前馈补偿器，很好地解决了该问题；而利用在遗传算法中加入BP算法形成的混合遗传算法作为神经网络的训练算法，此算法具有收敛速度快，不易陷入局部权小的特点。     

雷达测量系统误差的残差诊断方法

针对在MISTRAM系统测量中经常可能在某些测元上出现较大系统误差的问题 ,提出了一种在雷达测量中用于识别系统误差项的残差诊断方法。其基本思路是依次去掉每个测元 ,联合其它测元求解目标参数 ,得到模型的残差 ,通过分析残差的特性可识别系统误差项。实际计算表明 ,该方法是有效的     

SINS/CNS/GPS integrated navigation algorithm based on UKF

A new nonlinear algorithm is proposed for strapdown inertial navigation system (SINS)/celestial navigation system (CNS)/global positioning system (GPS) integrated navigation systems. The algorithm employs a nonlinear system error model which can be modified by unscented Kalman filter (UKF) to give predictions of local filters. And these predictions can be fused by the federated Kalman filter. In the system error model, the rotation vector is introduced to denote vehicle’s attitude and has less variables tha...     

A NEW STABILIZED FINITE ELEMENT METHOD FOR SOLVING THE STOKES AND NAVIER-STOKES EQUATIONS PART II. THE NONLINEAR CASE

1 IntroductionIt is necessary to make some conuneots on some of the re1ated methodologies lately emp1oyedto stabilize the Ga1erkin finite e1ement method for the incompressible NavierStokes equations,including the Stokes equation and the advectiondiffusion problem. Owing to the fact that thehigh Reynold number amounts to a small diffusive term in the advection--diffusion problem,here we will repeatedly mention the diffosion, not the Reynold number or the viscosity in the.. Jincompressible Na…     

基于反双曲正切函数的跟踪微分器设计与应用

通过引入反双曲函数与终端吸引子函数,构造了一种新型跟踪微分器。反双曲正切函数在平衡点较近处优异的线性特性可以保证系统收敛的平滑性,远离平衡点处的非线性特性能保证收敛的快速性。通过引入终端吸引子函数,降低了高频信号引起的抖振,增强了噪声抑制能力。通过扫频测试总结得出了参数整定规则后,与典型的跟踪微分器进行对比仿真,测试了所设计新型跟踪微分器的性能。最后,在控制器设计时,基于所提出的新型跟踪微分器构造了干扰观测器。通过仿真对比,证明所构造的干扰观测器实现了对模型不确定项的有效估计。     

并行组合扩频非等概超宽带系统误码性能研究

基于并行组合扩频的超宽带通信系统,结合了并行组合扩频通信系统的高效通信能力和超宽带通信系统的良好保密性能,可以同时满足通信系统对高效性和安全性的要求。该系统中的并行组合扩频部分输出的信号具有非等概出现的特点,推导了在非等概情况下超宽带系统跳时M进制〖CD*2〗脉冲幅度调制（time hopping M ary pulse amplitude modulation, TH M PAM)方式的解调差错概率,并通过仿真验证了公式推导的正确性。在加性高斯白噪声信道条件下,建立系统仿真模型,对系统整体性能进行仿真。仿真结果表明,基于并行组合扩频的超宽带通信系统具有良好的误码性能,并且优于常规超宽带通信系统和基于连续波调制的并行组合扩频通信系统。     

Time Optimal Feedrate Generation with Confined Tracking Error Based on Linear Programming

In this paper,the problem of time optimal feedrate generation under confined feedrate,axis accelerations,and axis tracking errors is considered.The main contribution is to reduce the tracking error constraint to constraints about the axis velocities and accelerations,when the tracking error satisfies a second order linear ordinary differential equation.Based on this simplification on the tracking error,the original feedrate generation problem is reduced to a new form which can be efficiently solved with linear programming algorithms.Simulation results are used to validate the methods.     

RESIDUAL A POSTERIORI ERROR ESTIMATE OF A NEW TWO-LEVEL METHOD FOR STEADY NAVIER-STOKES EQUATIONS

Residual-based a posteriori error estimate for conforming finite element solutions of incom-pressible Navier-Stokes equations,which is computed with a new two-level method that is differentfrom Volker John,is derived.A posteriori error estimate contains additional terms in comparison tothe estimate for the solution obtained by the standard finite element method.The importance of theadditional terms in the error estimates is investigated by studying their asymptotic behavior.For opti-mal scaled meshes,these bounds are not of higher order than the convergence of discrete solution.Thetwo-level method aims to solve the nonlinear problem on a coarse grid with less computational work,then to solve the linear problem on a fine grid,which is superior to the usual finite element methodsolving a similar nonlinear problem on the fine grid.     

一种改进的神经网络控制方法及其仿真研究

分析了一般神经网络控制系统中学习误差选择的问题,指出系统误差不等于理论上神经网络用于学习的误差,因而网络的性能会受到影响。进而针对局部逼近神经网,提出了一种改进的控制器结构,并讨论了其学习算法。仿真实验研究表明该方法收敛速度快,学习能力强,证明其在系统控制中的合理性和有效性。     

处理机具有不同开始加工时间的排序问题$Q,a_i‖C_{max}$

讨论任务的加工是不可中断、处理机是恒速机的排序问题$Q,a_i‖C_{max}$,证明了用ＬＰＴ算法求解该问题的误差界是２．     

由单变量受扰观测序列估计混沌系统敏感参数

针对高维非双曲线型混沌系统敏感参数难以估计的问题,在充分挖掘非线性系统及其观测序列本身特性的基础上,提出了一种由单状态变量受扰观测序列估计其系统敏感参数的新方法,有效的解决了由正的李雅谱诺夫指数所引起的误差扩散问题。该方法不同于常规的重影轨迹估计算法,故不受非线性系统非双曲线特性的限制,不必考虑系统对单步误差的要求。也不同于常规的序列去噪算法,因为去噪并不能得到系统的一条真实轨迹,因而无法精确估计系统的敏感参数。作为该方法的潜在重要应用,它为压缩非双曲线型混沌系统的受扰观测序列提供了一种新途径。     

A MATHEMATICAL PROGRAMMING MODEL FOR THE COEXISTENCE OF COMPETITIONS AND COOPERATIONS PROBLEMS

We study in this paper a mathematical programming model for the coexistence of competitions and cooperations problems. We introduce a new solution concept, s-optimal solution for the problem, which always exists under compact and continuous conditions. It is shown that an s-optimal solution can be obtained by solving a nonlinear programming problem. Some examples are given to explain how to compute an s-optimal solution.