【作者】 程贤福；

【导师】 肖人彬；

【作者基本信息】 华中科技大学 ， 机械设计及理论， 2007， 博士

【摘要】 由于设计理论与方法在产品开发中起着重要作用,激发了人们对设计的重视,涌现出了多种设计理论与方法。其中Suh提出的公理设计是具有代表性的设计理论,该理论力图为设计建立一个科学基础,通过为设计者提供一个基于逻辑和理性思维过程及工具的理论基础来改进设计活动。利用公理设计理论可增强设计的创造性、减少设计的复杂性以及设计的反复和迭代、设计方案的稳健性,提升设计学的科学性,并使设计活动系统化、逻辑化和理性化。另外,不同背景下发展起来的设计理论与方法各有其适合的应用领域,需要将不同设计理论与方法中的设计工具集成以适应产品设计过程中的各阶段,实现它们之间的优势互补。而揭示不同设计理论与方法的内在关系,是实现优势互补的基础。本文借鉴国内外相关研究成果,以公理设计理论与稳健设计方法为重点研究对象,将设计评价方法、优化技术引入其中,研究公理设计理论及其在其它设计领域应用的普适性,探讨公理设计与稳健优化设计之间的关系,为稳健优化设计提供一种新的研究途径。详细分析了公理设计域中元素的分解过程和相邻域之间的之字型映射过程,阐述了具体利用独立公理进行设计的过程。将公理设计应用于优化设计中,提出了基于公理设计的多目标优化设计方法。使用正交试验和方差分析技术确定设计变量对各设计目标的影响程度,按设计目标的多少对设计参数进行分组,使之类似于无耦合的优化设计,避免了多个设计目标之间的反复权衡。分析了信息公理评价方法的特点,提出了适合于设计评价的信息量计算方法,建立了基于公理设计和模糊数学的设计方案评价模型,给出了具体的评价过程。该方法可根据用户需求来修改各评价指标的设计范围,从而使整个评价过程更加灵活。公理设计与稳健设计具有很大的相似性,在分析公理设计与稳健设计各自优劣的基础上,深入探讨了独立公理和信息公理与稳健设计的关系,揭示了公理设计与稳健设计之间的内在联系。并针对质量特性常用的几种统计分布类型,分别建立了公理设计中的信息量与稳健设计方法的质量损失函数/信噪比之间的数学联系。从而延伸了公理设计改善产品设计质量的功效和扩展稳健设计的通用性。此外,本文还对稳健优化设计问题进行了研究,针对设计函数为隐函数或高度非线性函数的稳健优化问题,提出了一种基于容差模型和正交试验的稳健优化方法。随后针对多目标稳健优化问题,建立了多目标优化的损失函数,根据信息公理与损失函数的一致性关系,建立以最小化各目标的总损失函数为目标函数,提出了基于公理设计和相容决策支持问题法的多目标稳健优化设计方法。基于上述理论与方法研究成果,通过门座起重机设计实例验证了其可行性、有效性和实用性。更多还原

【Abstract】 Product design is an important stage to determine the product performance, and effective product design can improve quality, reduce cost, and reduce time to market, whereas the design theories and methods play a major role in product design. Up to now, many design theories and methods have been presented. Among which the Axiomatic Design (AD) is a representative design theory. Axiomatic design offers a scientific base for design and improves design activities by providing the designer with a theoretical foundation based on logical and rational thought processes and tools. It is helpful to provide a systematic way of designing products and large systems, reduce the random search process, minimize the iterative trial-and-error process, make designers more creative and develop better designs and select the best design among good alternatives.Any design theory and method has its own characteristics and emphases. Furthermore, modern products tend to be more complicate, whereas the time to market and life cycle of products are reduced. As a result, it is difficult for the sole design method to satisfy the design requirements. Therefore, it is necessary to integrate different design tools of design theory and methods, so that the mutually combination of them has complementary advantages. Whereas revealing the inherent connection among the different design theory and methods is the base of achieving complementary advantagesUsing the relevant theoretical research achievements for reference, the axiomatic design theory and robust design method are deeply researched in this dissertation, in which the methods of evaluation and optimization design are introduced. The relation between axiomatic design and robust design is discussed and application of axiomatic design in other design fields, and a new approach to the theoretical research on robust optimal design is presented.The process of decomposition of domain elements and zigzagging mapping between neighborhood domains in axiomatic design are analyzed in detail, and the designing process utilizing Independence Axiom is expounded. By introducing axiomatic design into optimization design, a multi-objective optimization method based on axiomatic design is proposed. The contribution of design variable to design objectives can be calculated using orthogonal experiment and Analysis of Variance (ANOVA), and design variables are classified into groups and the number of groups is the same as that of the design objectives in order to make the design similar to the uncoupled or decoupled optimization design, which avoid of trade-off between design objectives.Then a new way of computing the information content is presented, which transforms numerical value from the fuzzy information by combining the theory of fuzzy mathematic and information axiom for the design is evaluated in quantity. For decision-maker, there is no need to provide the detail weighting factors of evaluation index and only through specifying their design and system range, the optimal design scenario can be selected according to the information contents. This approach allows the decision-maker to express preference flexibly.There exist some common concepts between axiomatic design and robust design. The relation between two axioms and robust design has been discussed thoroughly, and the inherent connection is revealed which indicates that the design satisfying the above two axioms is more robust. Furthermore, for several typical statistical distribution forms of product’s quality characteristic, the mathematical relationship between the information content and quality loss as well as the relationship between information content and signal-to-noise ratio has been established. Due to the unification of axiomatic design and robust design, the virtue of axiomatic design for improving product quality can be promoted and the general application of robust design can be extended.In addition, robust optimal design is also studied in this dissertation. To solve implicit and high non-linear design function in engineering design optimization, a general approach for robust optimal design was proposed, which was based on tolerance model and orthogonal experimental method. Then for multi-objective robust optimal design, the loss function of multi-objective optimization is defined. According to the accordant relation between Information Axiom and quality loss function in the robust design process of product quality, that the total loss function of each sub-objective function is considered as objective function, and the approach of multi-objective robust optimal design based on axiomatic design and comprise decision support problem is proposed.Using the above methods, the design of portal crane is successfully accomplished. The case study indicates that the proposed methods are feasible and effective.更多还原