试验设计

The thesis also draw several conclusions as follows:(1) Orthogonal experiment design method can be viewed as a special case of genetic algorithm, i. e. a genetic algorithm with a fixed initial population, an oriented-mutation operator and one evolution epoch.(2) In terms of running steps, genetic algorithm is more complex than orthogonal experiment design method.

论文首次对正交试验设计法和遗传算法这两种独立发展起来的方法进行比较研究，提出(1)正交试验设计法是遗传算法的一种特例，即是一种初始种群固定的、只使用定向变异算子的、只进化一代的遗传算法；(2)遗传算法产生的优化解优于正交试验设计法产生的优化解，且遗传算法处理交互作用项的效率高于正交试验设计法；(3)遗传算法的步骤比正交试验设计法复杂，所需的试验次数也要多。

The main contents includes single factorial experiment design, multi-factorial experiment design, parameter design, incomplete block designs, regression design and response surface analysis, optimal design and etc.

主要内容有：单因子试验设计、多因子试验设计、参数设计、不完全区组设计、回归设计和响应曲面分析、最优设计等。

It focuses extensively on multiple and partial regression and correlation, some uses of Chi squares, and some advanced experimental designs and data analysis, including: two- and three- factor factorial, Latin square, nested, split-plot, confounding and fractional replication experiments.

本课程进一步介绍多元回归与偏回归，复相关与偏相关，χ2的应用，和部分高级试验设计与结果分析，包括：两因素和三因素因子式试验设计与分析，拉丁方试验设计与分析，巢式试验设计与分析，裂区试验设计与分析，混杂和部分重复试验设计与分析等内容。

At the same time, 9 groups data are conversely choosed according to uniform orthogonal array from 71 groups observed data of suspended grid plunge pool, the mold is set up by the method of PPR, the surplus 62 groups data are used to proceed examination, the empirical type optimum design of experiments are proceeded inversely, all of these confirm that a lot of model experimental results can be replaced by a few of test times under nonstrict uniform experimental condition. The result can provide the feasibility reference basis for same kind of model design of experiments in future.

同时从71组悬栅消力池观测数据中，按均匀正交表反选9组数据，用PPR方法建模，剩余的62组数据进行检验，逆向进行了实证性优化试验设计，证实了即使在非严格的均匀试验条件下，仍然可以用少量的试验次数替代大量的模型试验结果，为今后同类模型试验设计提供可行性的参考依据。

This paper formed a new method to get the wanted uniform design by computer sampling a number of experiment plan, counting their uniformities and choosing the best one of them as the last experiment plan.

这个试验设计方法，比使用均匀设计表来安排试验更简单直接，而且可以证明，其设计的试验点分布的均匀性，在抽样方案数足够多的情况下高于按均匀设计表来安排的试验。

Based on the ETA/VPG and LS-DYNA software platform and "vehicle-barrier-curve-road" integrated system dynamics model, in the research, the orthogonal experimental design method and the dynamic explicit finite element method were applied jointedly. The rear wheel elevation was considered as criterion for evaluation and six cross-sectional shape design variables were created. The application of L25 (5^6) orthogonal table for simulation and the use of SPSS statistical package help the variance analysis of simulation results, and the influence levels of various factors were studied to determine the optimal parameters of concrete barrier.

为此，以NJ型混凝土护栏尺寸参数为参考基础，以防撞等级为SS级的护栏为研究目标，以大型客车总质量为18t、碰撞速度为80km/h、碰撞角度为20°的碰撞条件为研究工况，基于eta/VPG及Ls-dyna软件平台及&汽车－护栏－弯道&一体化车－路系统动力学模型，联合应用正交试验设计方法及动态显式有限元方法，以客车后轮抬高量为考核指标，对控制弯道混凝土护栏截面形状的6个设计变量，应用L25(5^6)正交试验表安排仿真试验，借助SPSS统计软件包，对仿真试验结果进行了方差分析，找出了各因素对结果的影响程度，并据此确定了弯道混凝土护栏的最佳尺寸参数。

The proposed method contains two main parts: one part is determination of the noncontinuous factors by the orthogonal design method and the other is determination of the continuous factors by the uniform design method.

联合试验设计法主要分为：用正交试验设计法确定非连续性因素及其在备选水平中的适宜水平，以及用均匀试验设计确定连续性因素及其备选水平中适宜的水平范围两部分。

With the help of orthorhombic experiments, the main research has been determined which includes optimization experiments of the key parameters optimization and multi- parameters comprehensive matching. The prototype was designed based on content and aim of the experiment.The experiment results of key parameters optimization show that Phase angle of adverse circulation should be larger than mixing arms phase angle on the single shaft.

本文在分析强制式搅拌机特点的基础上，探讨了与搅拌质量有关的参数，并给出了各参数优化的目标，利用正交试验设计方法，确定了试验研究的主要内容：包括关键参数的优化试验和多参数综合匹配的优化试验，根据试验内容和试验目的，设计了试验样机。

Basic topics and ideas in the design of experiments: randomization and randomization tests; the validity and analysis of randomized experiments; randomized blocks; Latin and Greco-Latin squares; plot techniques; factorial experiments; the use of confounding and response surface methodology; weighing designs, lattice and incomplete block and partially balanced incomplete block designs.

实验设计：实验设计：试验设计的基本思想和论题：随机化和随机化检验；随机试验的效度与分析；随机区组设计；拉丁方和希腊拉丁方；制图技术；析因试验，混杂法和反应曲面法的运用；加权设计，方格设计和不完全区组设计，部分平衡的不完全区组设计。

Basic topics and ideas in the design of experiments: randomization and randomization tests; the validity and analysis of randomized experiments; randomized blocks; Latin and Greco-Latin squares; plot techniques; factorial experiments; the use of confounding and response surface methodology; weighing

实验设计：试验设计的基本思想和论题：随机化和随机化检验；随机试验的效度与分析；随机区组设计；拉丁方和希腊拉丁方；制图技术；析因试验，混杂法和反应曲面法的运用；加权设计，方格设计和不完全区组设计，部分平衡的不完全区组设计。

This one mode pays close attention to network credence foundation of the businessman very much.

这一模式非常关注商人的网络信用基础。

Cell morphology of bacterial ghost of Pasteurella multocida was observed by scanning electron microscopy and inactivation ratio was estimated by CFU analysi.

扫描电镜观察多杀性巴氏杆菌细菌幽灵和菌落形成单位评价遗传灭活率。