非线性目标函数问题的求解策略

线性规划问题是不等式的一项重要应用之一,其考查目的是利用不等式的几何意义求与不等式相关的最值问题.根据目标函数的不同可以分为线性目标函数及非线性目标函数,以下介绍常见的非线性目标函数问题的求解策略.     

一道伯克利数学问题的再讨论

对伯克利数学问题集中关于余弦函数的一个不等式建立了双边不等式.同时给出了关于反正弦函数、反余弦函数、反正切函数、反双曲正弦函数、反双曲正切函数的双边不等式.     

变分不等式的极大熵函数方法

利用极大熵函数方法将不等式组及变分不等式的求解问题转化为近似可微优化问题,给出了不等式组及变分不等式问题近似解的可微优化方法,得到了不等式组和变分不等式问题的解集合的示性函数.     

时标上的推广的Pachpatte型不等式及其在边值问题中的应用

该文利用单调化技巧研究了时标上的推广的Pachpatte型不等式, 该不等式右端有一个非常数项和三个包含未知函数与没有假设单调性的非线性函数的复合函数的积分项, 不等式左端是未知函数与非线性函数的复合函数. 所得不等式不仅把Pachpatte型不等式的离散形式和连续形式统一起来, 而且推广了已有的时标上的相应不等式. 最后, 用得到的结果研究时标上边值问题解的估计.     

例谈不等式恒成立问题

<正>我们知道,不等式恒成立的证明可以转化成对函数最值问题的研究,进而可以借助导数工具研究函数最值.而利用不等式的性质,可以对不等式进行等价转化,这就会使研究的函数模型发生变化.同时,我们可以从函数角度认识不等式,两个等价的不等式因为形式的不同,所对应的函数图象的关系也会有不同.今天我们通过一道题来体会此类问题解决的策略和值得关注的地方.     

函数不等式的解题技巧(连堂讲稿)

[复习说明 ]函数不等式既具有函数的抽象性又具有不等式的灵活性 ,对能力和潜能的考查具有独特作用 ,因此倍受命题者青睐 ,近几年高考试题就是很好的例证 .本专题复习的重点是 :函数与不等式的合理转化 ;难点是 :函数性质与不等式性质的综合运用 .[内容提要 ]1 .求解函数型不等式的基本方法 :构造函数、变更主元、特值转化、系数代换、判别式法、单调性及数形结合等 .2 .常用化归方式 :利用条件结合函数性质、不等式性质 ,将问题化归为函数问题或不等式问题 .[范例精选 ]例 1　设 f ( x) =ax2 bx　且1≤ f ( - 1 )≤ 2 ,　 2≤ f ( 1 )≤ …     

一道伯克利数学问题的探讨

对伯克利数学问题集中关于余弦函数的一个不等式建立了双边不等式,同时给出了关于正弦函数、正切函数及双曲正弦、双曲余弦、双曲正切的双边不等式.     

函数与方程思想在三角问题中的应用

函数思想,是指用函数的概念和性质去分析问题、转化问题和解决问题.方程思想,是从问题的数量关系入手,将问题中的条件转化为数学模型:方程、不等式或方程与不等式的混合组,然后通过解方程(组)或不等式(组)来使问题获解.函数与方程犹如亲兄弟,彼此身上存在对方的影子,两者互相转化接轨,形成了函数与方程思想.本文将用函数与方程思想来解决三角函数的证明求值问题.     

不等式情境，函数性视角——一道不等式存在性问题的探究

<正>含参不等式的存在性问题是高考数学试卷中比较常见的一类综合应用问题,经常交汇融合函数与不等式的相关知识,场景变化多端,形式创新多变,是知识综合与创新应用的一个重要载体.此类问题经常借助含参不等式的合理恒等变形与等价转化,综合利用不等式的基本性质转化为函数问题,从函数的视角来分析,借助函数的基本性质、图象等来处理与应用,实现问题的巧妙解决.     

泰勒公式的应用与技巧

<正> 在高等数学中,有关极值的判定、函数不等式和定积分不等式等问题的证明,往往技巧性很高.通常被人们认为这是数学中的难点,这是因为每个不同的数学问题都具有本身独特的处理方法.由于定积分不等式依赖干函数不等式,而函数不等式的证明方法通常用:拉格朗日中值定理,单调性、函数的极值和凸函数性质等.如何在众多的习题中找到其较好解法.就解题实践而论.对于某些结构特殊的题目,用一般方法求解,求证,常     

关于p-凸函数的Hadamard型不等式

对于P-凸函数在给定连续区间上的算术平均问题,通过P-凸函数理论将其转化为定积分问题,利用定积分的定义计算和定积分运算,建立了P-凸函数的Hadamard 型不等式,给出了证明和发现不等式的实例.     

Extension of Talenti’s Inequality and Maximum Values Relative to Rearrangement Classes

The article starts by revisiting and extending the Talenti’s inequality where the sharpness of the extended inequality is also addressed. The process leading to the extension comprises two steps. First, an observation that the Talenti’s inequality indeed can be formulated in terms of a rearrangement class. Second, proving that the inequality holds even when the rearrangement class is replaced by a much bigger (modulo trivial cases) set namely an appropriate closure of the class. The article then continues to introduce and explore a related maximization problem, associated to the classical Poisson equation, where the admissible set is the class of rearrangements of a given function. The article briefly explains the physical interest in this optimization problem. The existence of optimal solutions is proved and the optimality conditions they satisfy are explicitly derived. The particular case where the rearrangement class is built out of a characteristic function is also discussed.     

Inequalities for distances between points and distance preserving mappings

In this paper, we generalize the short diagonals lemma by proving a new inequality for distances between six points. Moreover, we apply this inequality to a partial solution to the Aleksandrov–Rassias problem.     

An optimal control problem for two-dimensional Schrödinger equation

In this paper we consider an optimal control problem controlled by three functions which are in the coefficients of a two-dimensional Schrödinger equation. After proving the existence and uniqueness of the optimal solution, we get the Frechet differentiability of the cost functional using Hamilton-Pontryagin function. Then we state a necessary condition to an optimal solution in the variational inequality form using the gradient.     

协方差不等式的证明和推论

利用构造二次函数的方法,证明了改进的协方差不等式,并在证明过程中得到3个推论,进一步探讨了随机变量具有线性关系时的若干结论.     

Error bound results for convex inequality systems via conjugate duality

The aim of this paper is to implement some new techniques, based on conjugate duality in convex optimization, for proving the existence of global error bounds for convex inequality systems. First of all, we deal with systems described via one convex inequality and extend the achieved results, by making use of a celebrated scalarization function, to convex inequality systems expressed by means of a general vector function. We also propose a second approach for guaranteeing the existence of global error bounds of the latter, which meanwhile sharpens the classical result of Robinson.     

一道积分不等式的多种证法

通过对一道积分不等式提供多种不同的证法,希望能对学生创造性思维及发散性思维的培养、开阔解题思路、提高综合应用数学知识的能力等有所帮助,并使学生对证明不等式的常用方法有所了解.     

A variational technique for optimal boundary control in a hyperbolic problem

We investigate the problem of controlling the boundary functions in a one dimensional hyperbolic problem by minimizing the functional including the final state. After proving the existence and uniqueness of the solution to the given optimal control problem, we get the Frechet differential of the functional and give the necessary condition to the optimal solution in the form of the variational inequality via the solution of the adjoint problem. We constitute a minimizing sequence by the method of projection of the gradient and prove its convergence to the optimal solution.     

Almost Sure Convergence and Square Functions of Averages Of Riemann Sums

Under a very moderate assumption on the Fourier coefficients of a periodic function, we show the convergence almost everywhere of the sequence of averages of its associated Riemann sums. The structure of the set of averages is analyzed by proving a spectral regularization type inequality, which allows to control the corresponding Littlewood-Paley square function.     

A generic approach to proving NP-hardness of partition type problems

This note presents a generic approach to proving NP-hardness of unconstrained partition type problems, namely partitioning a given set of entities into several subsets such that a certain objective function of the partition is optimized. The idea is to represent the objective function of the problem as a function of aggregate variables, whose optimum is achieved only at the points where problem Partition (if proving ordinary NP-hardness), or problem 3-Partition or Product Partition (if proving strong NP-hardness) has a solution. The approach is demonstrated on a number of discrete optimization and scheduling problems.