RECURRENT NEURAL NETWORK MODEL BASED ON PROJECTIVE OPERATOR AND ITS APPLICATION TO OPTIMIZATION PROBLEMS

The recurrent neural network (RNN) model based on projective operator was studied. Different from the former study, the value region of projective operator in the neural network in this paper is a general closed convex subset of n-dimensional Euclidean space and it is not a compact convex set in general, that is, the value region of projective operator is probably unbounded. It was proved that the network has a global solution and its solution trajectory converges to some equilibrium set whenever objective function satisfies some conditions. After that, the model was applied to continuously differentiable optimization and nonlinear or implicit complementarity problems. In addition, simulation experiments confirm the efficiency of the RNN.     

Crystalline Mean Curvature Flow of Convex Sets

We prove a local existence and uniqueness result of crystalline mean curvature flow starting from a compact convex admissible set in . This theorem can handle the facet breaking/bending phenomena, and can be generalized to any anisotropic mean curvature flow. The method provides also a generalized geometric evolution starting from any compact convex set, existing up to the extinction time, satisfying a comparison principle, and defining a continuous semigroup in time. We prove that, when the initial set is convex, our evolution coincides with the flat φ-curvature flow in the sense of Almgren-Taylor-Wang. As a by-product, it turns out that the flat φ-curvature flow starting from a compact convex set is unique.     

The Geometry of the Solution Set of Nonlinear Optimal Control Problems

In an optimal control problem one seeks a time-varying input to a dynamical systems in order to stabilize a given target trajectory, such that a particular cost function is minimized. That is, for any initial condition, one tries to find a control that drives the point to this target trajectory in the cheapest way. We consider the inverted pendulum on a moving cart as an ideal example to investigate the solution structure of a nonlinear optimal control problem. Since the dimension of the pendulum system is small, it is possible to use illustrations that enhance the understanding of the geometry of the solution set. We are interested in the value function, that is, the optimal cost associated with each initial condition, as well as the control input that achieves this optimum. We consider different representations of the value function by including both globally and locally optimal solutions. Via Pontryagin’s maximum principle, we can relate the optimal control inputs to trajectories on the smooth stable manifold of a Hamiltonian system. By combining the results we can make some firm statements regarding the existence and smoothness of the solution set.     

基于随机有限元的非线性结构稳健性优化设计

结合结构优化技术和摄动随机有限元方法研究了非线性结构稳健设计问题。将结构稳健性优化设计问题构造为双目标优化问题。优化目标包含结构性能函数的期望值和标准差。约束函数的变异也给予考虑,并采用基于函数梯度的算法进行求解。为对具有路径相关特征的非线性结构性能及结构响应的平均值及标准差进行分析。本文采用缩减的随机变量,提出了基于增量法的摄动随机有限元计算格式。在此框架下,进一步提出以一般泛函形式表达的结构性能的平均值和方差及其灵敏度的计算格式。为显示方法的有效性。文中给出几个数值算例。     

Iterative construction of solutions to nonlinear equations of lipschitzian and local strongly accretive operators

In this paper, we investigate the Ishikawa iteration process in a p-uniformly smooth Banach space X. Let T: X→X be a Lipschitzian and local strongly accretive operator and the set sol(T) of solutions of the equation Tx=f be nonempty. We show that sol(T) is a singleton and the Ishikawa sequence converges strongly to the unique solution of the equation Tx=f. In addition, whenever T is a Lipschitzian and local pseudocontractive mapping from a nonempty convex subset K of X and the set F(T) of fixed points of T is nonempty, we prove that F(T) is a singleton and the Ishikawa sequence converges strongly to the unique fixed point of T. Our results are the improvements and extension of the results of Deng and Ding⁽⁴⁾ and Tan and Xu⁽⁵⁾.     

多自由度振动系统的一类凸集响应

对激励用凸集合建模的线性时不变振动系统定义了响应集的概念，将多自由度振动系统的凸集响应问题转化成寻求响应集的问题，从而降低了求解的难度．并提出了求解一类凸集载荷的响应集的支撑超平面法．对有限时刻的响应集，导出了求任意方向边界点的解析解，而对无限时刻的响应集，建议了一种求任意精度的凸多胞形近似的方法．最后给出了一个实例，得到了不同时刻的响应集，并将凸集响应与谐波响应及非平稳随机响应作了对照     

Formulation and Solution of Dynamic Problems of Elastic Rod Systems Subjected to Boundary Conditions Described by Multivalued Relations

The dynamic behavior of rod systems under the action of external force factors described by multivalued (subdifferential) relations is studied. The mathematical formulation of the problem is given in the form of a dynamic quasivariational inequality. With the use of the Newmark difference scheme, successive approximations, and finite-element discretization, the problem is reduced to minimization of a convex nonsmooth finite-dimensional functional with respect to velocities at each time step. Introduction of auxiliary variables by the method of a modified Lagrangian reduces the problem of minimization of this functional to a sequence of smooth problems of nonlinear programming. The algorithm is verified using the numerical solution for a problem with one degree of freedom. The algorithm proposed is used to calculate the rods of deep-well pumps.     

接触分析的光滑模型及迭代算法

利用变分不等式和基于信息熵的凝聚函数把有摩擦接触问题模型化为一个标准的凸二次规划问题,极大地简化了这一复杂的问题,同时引入摩擦方向约束并构造了以无摩擦解为初值的迭代算法,在较摩擦系数时计算也能保证收敛,算例表明算法高效可靠。     

求解一类可分离凸规划的对偶显式模型DP-EM方法

推导对偶目标函数的精确显式表达式,可选用更多成熟高效的求解方法,从而进一步提高了非线性规划对偶理论求解结构拓扑优化问题的效率.研究工作来源于非线性凸规划同其对偶规划的间隙为零,可以等价转化为对偶问题求解,通常可以大大地缩小问题的规模,可是二者不具有显式关系却影响了对偶解法的应用.所幸的是,结构优化当中一大类问题包括连续体结构拓扑优化问题,不仅具有凸性,而且具有变量可分离性,于是原变量和对偶变量之间有了显式关系,因此,对偶解法成了38年来被应用的有效方法之一.然而长期以来,对偶问题的目标函数并不是显式,这缘于含参数的极小化问题导致目标函数为隐式表达,常见的显式化方法是进行二阶近似.本文突破了对偶问题难以显式化只能采用近似显式的定势,将我们提出的"对偶规划-显式模型"(DP-EM)方法应用于连续体结构拓扑优化,并与对偶序列二次规划(DSQP)算法及移动渐近线(MMA)算法为求解器的方法进行计算效率对比,结果显示:(1)MMA算法比DP-EM算法和DSQP算法的外部迭代次数均多;(2)DP-EM算法与DSQP算法外循环次数相同,而内循环数显著减少.说明了DP-EM算法具有显式对偶函数的优势.     

斜拉桥索力优化的强次可行序列二次规划法

提出了一种斜拉桥索力优化的实用方法-强次可行序列二次规划法. 该方法通过建立斜拉 桥索力优化的非线性规划模型，以主梁和索塔的弯曲应变能为目标函数，斜拉索的索力为设 计变量，结构应力和索力为约束条件，计入大跨度斜拉桥各种几何非线性因素的影响，采用 强次可行序列二次规划算法进行优化求解，确定斜拉桥成桥合理状态的索力. 运用该方法对 某斜拉桥进行索力优化，结果表明该方法简单、有效.     

The concave or convex peaked and smooth soliton solutions of Camassa-Holm equation

ＩｎｔｒｏｄｕｃｔｉｏｎＣａｍａｓｓａ ,Ｈｏｌｍ[1]ｏｂｔａｉｎｅｄａｃｌａｓｓｏｆｎｅｗｃｏｍｐｌｅｔｅｌｙｉｎｔｅｇｒａｂｌｅｓｈａｌｌｏｗｗａｔｅｒｅｑｕａｔｉｏｎ ,ｉ.ｅ.,Ｃａｍａｓｓａ＿Ｈｏｌｍｅｑｕａｔｉｏｎ2ｕｔ+ 2ｋｕｘ-12 ｕｘｘｔ+ 6ｕｕｘ =ｕｘｕｘｘｘ+ 12 ｕｕｘｘｘ. ( 1 )Ｆｏｒｅｖｅｒｙｋ,ｔｈｅＥｑ .( 1 )ｉｓａｃｌａｓｓｏｆｃｏｍｐｌｅｔｅｌｙｉｎｔｅｇｒａｂｌｅｓｙｓｔｅｍ .Ｔｈｉｓｃｌａｓｓｏｆｅｑｕａｔｉｏｎｉｓａｃｌａｓｓｏｆｎｏｔｏｎｌｙｓｔｒａｎｇｅｂｕｔａｌｓｏ…     

Solving inverse viscoelasticity problems via aggregate function approach

This paper presents a nonlinear programming model with multi-constraints of inequality to solve inverse viscoelasticity problems. By utilizing an aggregate function approach, multi-constraints are converted into a single smooth constraint. The optimization with a single constraint is realized by using a technique of multiplier penalty functions, and a standard BFGS algorithm is employed in the solution process. Results with time dependent and independent noise data are given.     

基于非概率可靠性的结构优化设计研究

基于不确定参量的凸集合描述,研究了考虑非概率可靠性约束时,结构优化设计模型的求解问题。由于非概率可靠性指标是用一个极小极大模型来定义的,故以该指标作为设计约束,将得到一个嵌套的二级优化模型。为了求解该模型,提出了一种序列线性化的计算方法。利用非概率可靠性分析的拉格朗日乘子,逐步构造可靠性指标的一阶近似,通过序列线性规划法求解二级优化问题。该算法可用于区间变量和超椭球凸集模型并存的情形,具有较好的适用性。论文给出了主要的敏度计算公式,并通过简单算例对所提算法进行了验证。     

Linear control systems with fuzzy parameter

We consider linear control systems with undetermined parameters that are elements of a fuzzy set. For this system, we introduce the notion of pencil of trajectories and establish some of its properties. We also introduce the notion of fuzzy reachable set and prove that it is convex and compact. __________ Translated from Neliniini Kolyvannya, Vol. 9, No. 1, pp. 61–67, January–March, 2006.     

Ergodic theorem for infinite iterated function systems

1IntroductionandProblem IteratedFunctionSystems(IFS)theorycanbesaidtobethecontinuationanddevelopment ofdynamicalsystemtheory.DynamicalsystemstheorydealswithiterationofonemapbutIFS theorydealswithiterationofmanymaps. IFStheory’srootwasveryearlybutthebeginningofactivedevelopmentwasHutchinson’s paper(1981).Heresearchedselfsimilarityoffractalsetsusingsystemoffinitenumberofsimilar contractionmapsofRn.Barnsleycalledafinitesetofcontractionmappsasaniteratedfunction systemsandsystemizedIFStheo…     

结构非概率鲁棒可靠性准则

通过反例指出Ben-Haim的鲁棒可靠性准则是错误的同时．利用集合数学理论中的凸集合间的偏序关系，给出了新的非概率凸模型理论的鲁棒可靠性准则。并用数值算例说明了所提出的结构非概率凸模型鲁棒可靠性理论的正确性和实用性。     

斜拉桥合理成桥状态确定的序列二次规划法

本文首先介绍斜拉桥合理成桥状态的概念和现有的斜拉桥索力优化方法。然后基于序列二次规划(SQP)算法,提出了一种用于确定斜拉桥成桥合理状态的实用方法,序列二次规划法。该方法通过建立斜拉桥索力优化的非线性规划模型,以斜拉桥主梁和索塔的弯曲应变能为目标函数,以各斜拉索的索力为设计变量,结构的应力和索力为约束条件,并计人大跨度斜拉桥各种几何非线性因素的影响,采用强次可行序列二次规划法进行优化求解,确定斜拉桥成桥合理状态的索力。运用该方法和空间非线性有限元分析程序分析了某斜拉桥的合理成桥状态,计算结果表明该方法简单、有效。     

Two non-probabilistic set-theoretical models for dynamic response and buckling failure measures of bars with unknown-but-bounded initial imperfections

This paper is concerned with the problem of comparison of two non-probabilistic set-theoretical models for dynamic response and buckling failure measures of bars with unknown-but-bounded initial imperfections. Two kinds of non-probabilistic set-theoretical models are convex models and interval analysis models. In convex models and interval analysis models, the uncertain quantities are considered to be unknown except that they belong to given sets in an appropriate vector space. In this case, all information about the dynamic response and buckling failure measures of bars is provided by the set of dynamic responses and buckling failure measures consistent with the constraints on the uncertain quantities. The dynamic response estimate is actually a set in appropriate response space rather than a single vector. The set estimate is the smallest calculable set which contains the uncertain dynamic response, but it is usually impractical to calculate this set. Two set estimate methods are developed which can calculate the time varying box or hyperrectangle, i.e. interval vector in the response space that contains the true dynamic response. Comparison between convex models and interval analysis models in mathematical proofs and numerical calculations shows that, under the condition of the outer enclosed ellipsoid from a hyperrectangle or an interval vector, the set dynamic response predicted by interval analysis models is smaller than that yielded by convex models; under the condition of the outer enclosed hyperrectangle or an interval vector from an ellipsoid, the dynamic response set calculated by convex models is smaller than that obtained by interval analysis models.     

Minimization of semicoercive functions: a generalization of Fichera’s existence theorem for the Signorini problem

The existence theorem of Fichera for the minimum problem of semicoercive quadratic functions in a Hilbert space is extended to a more general class of convex and lower semicontinuous functions. For unbounded domains, the behavior at infinity is controlled by a lemma which states that every unbounded sequence with bounded energy has a subsequence whose directions converge to a direction of recession of the function. Thanks to this result, semicoerciveness plus the assumption that the effective domain is boundedly generated, that is, admits a Motzkin decomposition, become sufficient conditions for existence. In particular, for functions with a smooth quadratic part, a generalization of the existence condition given by Fichera’s theorem is proved.