查询词典 optimization

In this thesis,the AM1,MNDO,MINDO/3(mainly AM1)and INDO/S-CI semiempirical MO methods were used toinvestigate the excited-state intramolecular protontransfer reactions of salicylic acid derivatives—salicylic acid,methyl salicylate,salicylaldehyde,o-hydroxyaceto-phenone,salicylamide and 3-hydroxy-picolinamide (6 conformers and 2-3 anion species);2-(2'-hydroxy-5' methylphenyl) benzotriazole(4 conformers),2-(2' hydroxyphenyl) benzimidazole (3 conformers and 3anion species),Bis-2,5-(2-benzoxazolyl)hydroquinone(3 conformers),2-(2'-hydroxyphenyl)benzothiazole(2conformers) and 7-azaindole dimer (2 conformers).Theinvestigations were described as follows.Geometry optimization,relative stability andhydrogen bonding energy First,for sylicylic acid derivative molecules,the AM1,MNDO and MINDO/3 methods were used toinvestigate ground-state geometry optimization,energies,relative stabilities and hydrogen-bondingenergies on the five kinds of the molecules(designing 6 conformers and 2-3 anion species).Comparing with experimental data,the optimizedgeometry,the order of stability,the hydrogen-bonding energies and the distances between O-O in O-H..O hydrogen bonds by AM1 method were in agreementwith the experimental data,however,the C-C bondlengths optimized by MNDO and MINDO/3 were longer,C-O and O-H bond lengths were shorter;for C-N bondlengths,the results opitimized by MNDO method werethe same as those by AM1 method,nevertheless the C-Nbond lengths given by MINDO/3 method were muchshorter.For some sylicylic acid derivatives(e.g.methyl salicylate,salicylamide),the order ofstabilities on the conformers given by MNDO andMINDO/3 methods were not in agreement with theexisting conformers deduced by experimental methods,and the hydrogen bonding energies calculated by MNDO.and MINDO/3 methods were smaller.Second,the studyon the other systems found that the optimizedgeometry of the proton-transfered product with INDOmethod could not be obtained,only could theoptimized geometry of reactant be obtained,and thecalculated hydrogen bonding energies were greater.Many results of calculation indicated that the studyon the excited-state intramolecular proton transferreaction system using AM1 method was suitable andreliable.

本论文用AM1、MNDO、MINDO/3（主要是AM1）和INDO/S-CI半经验分子轨道方法对水杨酸衍生物系列——水杨酸、水杨酸甲酯、水杨醛、O-羟基乙酰苯酮、水杨酰胺和3-羟基吡啶酰胺（6种异构体和2-3种阴离子）；2-（2'-羟基-5'-甲基苯基）苯并三〓唑（4种异构体）；2-（2'-羟基苯基）苯并咪唑（3种异构体和3种阴离子）；2，5-二间氮杂氧茚氢醌（3种异构体）；2-（2'-羟基苯基）间〓杂硫茚（2种异构体）和7-〓吲哚二体（2种异构体）的激发态分子内质子转移反应在以下几个方面进行了较系统的理论研究：几何构型优化和相对稳定性及氢键能首先以水杨酸衍生物系列分子为例，用AM1、MNDO和MINDO/3方法考察了5种分子（每种分子设计6种异构体和2-3种阴离子）的基态几何构型优化，能量、相对稳定性和氢键能计算，通过和实验数据进行比较，AM1方法给出的优化几何构型、稳定性次序、氢键能和O—H。。。O氢键的0—0距离与实验数据吻合最好，MNDO和MINDO/3方法优化的C-C键长偏长，C-O键和O-H键长偏短；对于C-N键长，MNDO和AM1优化结果差别不大，而MINDO/3给出了过短的C-N键长，MNDO和MINDO/3方法给出的有些水杨酸衍生物分子（如水杨酸甲酯和水杨酰胺）异构体的稳定性次序和实验上推测的可存在异构体结果不一致，MNDO和MINDO/3方法给出的氢键能偏低，对其他体系的研究发现INDO方法常常不能得到质子转移产物的优化几何构型，只能得到反应物的优化构型，并且估算的氢键能偏高，大量的计算结果表明AM1方法对本论文研究的激发态分子内质子转移反应体系是适宜和可靠的。

Accordingly, we know the connection of the universal duality and boundness of a primal-dual pair feasible set.We apply these theories to practical applications in chapter four.In chapter five we make many conclusions and bring forward research expectation for the future.conic optimization; dual conic optimization; duality gap; universal duality

研究主要有：利用Nesterov和Todd的齐次模型判断锥规划与其对偶锥规划解的存在性；类似于线性规划推导出锥规划的KKT-条件；从锥规划的泛对偶性得到锥规划与对偶锥规划的零对偶间隙存在的条件，从而了解泛对偶性与原—对偶

Unfortunately, just the lack of design model to guide software design process of EC brings many issues in practices of EC. It is difficult to design reusable EC software based on the basic structure of EC. For example, the EC software aimed to solve function optimization can't be used to solve combinatory optimization without an overall modification.

由于缺少一个软件设计模型来指导演化算法的软件设计过程，仅从演化计算的基本结构出发设计出的软件很难用来求解不同种类的问题，比如针对函数优化而设计的演化算法不经大幅度的修改就不能够用来求解组合优化问题。

From the view of the optimization methods, these methods confined to the optimization under singular-goal condition, there are no models which compressively reflect mathematics model of these goals in all-round way, so the research in this respect is feasible .necessary that having great significance too.

虽然也出现了很多综合优化模型和多目标优化，但这些模型往往只考虑工期与成本、工期与资源，质量与成本的单一关系，或者只考虑工期、成本、资源的关系，往往忽略了工程质量。

This paper discusses the characteristics of the portal frame building constructer and studies optimization method to design them.The systematic analysis of the portal frames steel structural characteristics is done in section 2. And in section 3 and section 4, the structural optimization design is introduced based on the principles of finite element analysis and the technology of Object Oriented Programming in the software ANSYS.

本文系统地分析研究了门式刚架钢结构的特点，在此阶段，以有限元为基础，借助大型有限元ANSYS面向对象程序设计技术进行优化设计，研究基于功能强大的ANSYS作结构优化设计的问题：提出有效的设计变量、优化目标函数(以降低结构重量、充分发挥材料性能为目标)，编写APDL优化程序控制文件，在ANSYS程序上实现门式刚架结构优化设计。

The contents include the layout optimization models of the difform graph elements, the optimality conditions , the optimality functions, the optimization algorithms for the subproblems, the algorithms for the non-overlap constraints and the unproved genetic algorithm.

主要包括不同图元的布局优化模型、子问题的最优性条件、最优性函数和优化算法、判断不干涉性算法及改进的遗传算法。

To avoid the prematurity phenomenon of particle swarm optimization, a novel hybrid genetic particle swarm optimization algorithm is proposed. An eliminative mechanism is introduced into the iteration of PSO.

针对粒子群算法容易出现&早熟&的缺点，提出了一种改进的混合遗传粒子群(hybrid genetic particle swarm optimization， HGPSO)算法。

Some methods aiming at the two problems have been researched. First, the optimal number of ATE site can be calculated based on a cost model of DAT-ATE. Second, the problem of test schedule which is equitant with two-dimensional Bin-Packing problem is presented. Then a TCG denotation satisfied with T-admissible rule is more convenient for combinational optimization. With Simulated annealing algorithm, better test scheduling results can be achieved; but there is no great optimization space because the efficiency of scheduling is greater than 90%. Third, A novel test compression method——Multi-capture testing is proposed to compress the stimuli. And the ATPG results show that MC achieves high compression ratio which is greater than 90% in some large-scale circuits. MISR, widely used in LBIST, is employed to compress the responses. The aliasing analysis demonstrates that the fault coverage reduction is little due to the existence of aliasing. Fourth, an improved TIC (called S-TIC) aiming at structured test is proposed based on ARM's TIC. A MC scheduling algorithm is proposed to compute the SoC test time and combine MC vectors to SoC test vectors. The scheduling results shows that test time is greatly reduced when using the proper scheduling priority.

首先，给出了低成本ATE的成本模型，根据此模型得到最低测试成本时系统级测试调度的基本参数——测试Site数；其次，本文将DAT-Scan方式的SoC测试调度等价为两维BP问题，为有效地求解该问题，将该BP问题表示为TCG图并通过模拟退火算法解决其求解过于复杂的问题，调度的结果表明该方法在测试时间上有一定改善，并指出DAT-Scan测试调度效率已经大于90％，进一步优化的空间较小；第三，本文尝试通过测试压缩来解决激励所需的测试带宽，提出了Multi-capture结构并解决了MC测试过程中的&矢量&退化问题，MC测试的ATPG结果表明其测试压缩率很高（大电路接近90％）；第四，在激励获得很高压缩比的情况下，测试响应也需压缩，本文采用LBIST中常用的MISR作为MC测试的响应压缩电路，理论分析和实验结果都证明了MC测试的别名对故障覆盖率影响较小（小于2％），并对两类别名的成因做了具体分析；第五，在MC测试及其响应压缩的基础上，本文改进了ARM公司的测试控制器TIC使之适合于MC测试，为了解决MC测试矢量合成问题，本文抽取了MC测试模型，通过固定优先级的MC测试调度模拟将MC矢量合成为ATE矢量，并模拟出了总的MC测试时间。

A typical wheel optimization design was exploringly studied, and the results show that through the wheel structural optimization design, the weight of the wheel may be reduced further. The benefits are material saving and improving of the vehicles'driving quality. It has vital significance to reduce the cost of the wheel.

对某一类型车轮的优化设计进行了探索性的研究，研究结果表明，通过对车轮结构的优化设计，可以进一步降低车轮的重量，在保证车轮性能的前提下充分地节省材料并改善车辆行驶性能，对降低车轮制造的成本具有重要的意义。

First, an attempt is made to improve the generability of neural network and the interpretability of fuzzy system. Then, combining empirical models with improved optimization algorithms, a systematic data-driven soft computing modeling and optimization framework is to be developed.

主要对以下内容开展研究：改进神经网络学习算法，简化模糊推理系统；改进遗传算法和粒群优化算法；开展软计算在工程材料中的应用研究。

Nowadays, most of research are to build a transmutative Petri Nets through adding controlling place sets, controlling arc sets and controlling policy to the basic Petri Nets, while the Controlled Petri Nets could be used to argue many controlling theory problems conveniently and to induce many logically and physically supervisory and solve the Event Feedback Controlling Problems and State Feedback Controlling Problem in DEDS supervisory theory.

目前大多数的研究表现为在变形后的受控Petri网基础上，利用各种方法求得各种逻辑型、结构型控制器，解决DEDS监控理论中的事件反馈控制问题与状态反馈控制问题。

On one hand, there are discussions with the works council and union about extension of short time working up to the end of September.

一方面，有讨论，工程理事会和联盟关于延长工作时间短至9月底。