【作者】 庞继红；

【导师】 张根保；

【作者基本信息】 重庆大学 ， 机械工程， 2011， 博士

【摘要】 “设计是产品质量的源头”,这一观点已被越来越多的专家和制造企业所认同,企业对产品设计质量越来越重视,同时投入大量的资源努力提高产品设计质量。本论文以数控机床为例,对复杂机电产品设计质量展开研究,针对设计质量特性的多尺度性、多领域性、多学科性、演化性和耦合性等特点,将关键设计质量特性(KDQCs)的概念引入到产品设计过程中,研究了关键设计质量特性的定义、关键质量特性的提取、基于反向映射的机床精度设计、基于优化决策模型的可靠性设计、设计质量的综合评价等技术。本论文在国家自然科学基金重点项目“复杂机电产品质量特性多尺度耦合理论与预防性控制技术”(项目批准号50835008 )、国家“高档数控机床与基础制造装备”科技重大专项(2009ZX04014-016)及数字制造装备与技术国家重点实验室(华中科技大学)开放基金的资助下,进行了复杂机电产品设计质量相关理论研究,提出提高复杂机电产品设计质量的解决方案,具体内容包括以下几个方面:(1)首先,讨论了产品设计质量特性基础理论。根据对已有文献的研究,分析了设计质量特性和关键设计质量特性的基本概念,论述了设计质量的发展历程及国内外的研究现状,说明了本论文的课题来源、主要研究内容及总体结构。然后,研究了产品关键设计质量特性的基础理论,分析了产品关键设计质量特性的形成机理,得出了产品关键设计质量特性的演变规律,并分类研究了关键设计质量特性对产品质量的影响,结合设计质量控制的技术与方法,提出了基于关键质量特性的产品设计质量控制策略。(2)研究了复杂机电产品关键设计质量特性提取与排序。综合分析了关键设计质量特性在不同子系统、不同层次、不同功能的多元性,建立了复杂机电产品多元质量特性的功能分解树。针对关键设计质量特性提取的复杂性和不确定性,综合运用模糊理论和FAHP方法建立关键设计质量特性的提取模型,用三角模糊数表示模糊比较判断的方法计算主观权重,并使用信息熵(IE)确定质量特性指标层的客观权重。然后应用模糊Borda方法组合FAHP法和信息熵法进行综合提取,根据模糊Borda数值大小最终决定多元设计质量特性的重要度,从中提取关键设计质量特性。该综合方法可实现程序化的求解,减少人工干预并提高了提取结果的准确性。最后用数控机床实例验证了该综合提取方法的合理性和可行性。(3)研究了机电产品精度这一关键设计质量特性的反向映射技术。针对数控机床设计过程的特殊性和复杂性,在考虑传统的质量功能配置(QFD)多源、多阶段复杂处理过程的基础上,本论文增加了产品运行阶段的反向映射,通过识别影响加工质量的关键要素,并结合用户需求,提出基于反向映射的设计关键质量特性的修正方法,综合映射成数控机床的精度设计质量特性。该方法使得产品设计、制造和使用过程的质量控制形成一个闭环系统,使产品质量信息能够及时、全面和准确地反映到产品各阶段,提高产品设计质量和用户满意度。其次,通过对加工质量要求(QRs)和关键设计质量特性的转换关系进行研究,提出了基于粗糙集(RS)和质量功能配置中质量屋(HoQ)的反向映射模型。接着,运用粗糙集理论简化和提取满足加工质量要求的要素,并提出了应用近似精确粗糙数确定加工质量要求的权重。然后,通过构建关键设计质量特性的反向映射模型,实现主要加工质量要求和关键设计质量特性之间的转换,帮助产品设计人员完成设计目标。最后,以卧式加工中心为例,验证了该模型的实用性和有效性。(4)研究了产品可靠性这一关键设计质量特性的优化决策技术。通过分析机电产品设计方案中多元设计质量特性的特定要求,给出了可靠性的优化方案决策流程,建立了可靠性的多属性指标体系,保证产品设计质量符合预定设计质量特性要求。提出了基于改进“逼近理想值的排序方法”(TOPSIS)和ELECTRE I的综合方法进行产品设计方案的决策。建立了产品可靠性这一关键设计质量特性最优设计方案的多目标决策模型,应用改进的TOPSIS方法计算各设计质量特性的决策指标的重要性,通过综合应用ELECTRE I方法准确地确定产品可靠性设计方案的优劣程度。最后,以高档数控机床的可靠性设计方案为例,说明该方法的有效性和可行性。(5)研究了机电产品设计质量的模糊物元综合评价技术。分析了产品设计质量评价的应用现状,给出了设计质量评价的原则及多元质量特性评价指标体系模型。针对产品质量多层次指标评价问题,提出了一种基于模糊物元(FME)的综合评价方法。该方法利用模糊物元理论分析多元质量特性及其属性特征,并建立了产品设计质量综合评价模型。针对传统决策法中权值确定的主客观评价的片面性,采用网络层次分析法(ANP)和IE技术方法综合确定评价指标权重,建立基于最小二乘法的优化组合权重模型计算组合权重,并运用模糊物元法给出评价结果。最后以数控机床的设计质量评价为例,讨论了该方法的可行性和合理性。更多还原

【Abstract】 Design quality is a key source of product quality, and this view has been shared now by more and more experts and manufacturing enterprises. Recently, more and more enterprises pay attention to the product design quality, set the best resource configuration and give the effective way to improve product quality. In view of the Quality Characteristics (QCs) of multi-scale, multi-field, multidisciplinary, evolvability and coupling, the theory of Key Design Quality Characteristics (KDQCs) is applied to the design process of computer numerical control (CNC) machine, then, the techniques and methods of extracting, mapping, optimizing and evaluating of KDQCs are proposed in this study. This paper is supported by“Complex mechanical and electrical products quality characteristic’s multi-scale coupling and preventive control”(National Natural Science Foundation, No. 50835008), the National Major Scientific and Technological Special Project for“High-grade CNC and Basic Manufacturing Equipment”(No. 2009ZX04014-016), and supported by Open Research Foundation of State Key Lab. of Digital Manufacturing Equipment & Technology in Huazhong University of Science & Technology. By the theories on KDQCs are studied under the support of these grants, we put forward the solutions for improving the design quality of mechanical and electrical products. Topics covered in this dissertation are as follows:(1) Firstly, this paper introduced the basic theory of key design quality characteristics (QCs) of mechanical and electrical products. Secondly, the key concepts, development and technology of key QCs of mechanical and electrical products are analyzed based on the existing literature. Then, the source, the main works and frame of this treatise are illustrated here. Furthermore, the basic theory of key design QCs for mechanical and electrical products is studied.Finally, based on the research on theory about the formation and its effect on quality of key design QCs, the strategy of quality control of key design QCs are proposed.(2) This paper introduced the multi-scale DQCs of mechanical and electrical products in different subsystem, levels and function, and built the multi-scale QCs model of mechanical and electrical products. Aiming at the complexity of ranking multiple QCs of mechanical and electrical products, the mathematical method of Fuzzy Analytical Hierarchy Process (FAHP) and Information Entropy (IE) is presented, which can effectively solve the uncertain problems such as subjectivity and ambiguity existed in the computational process of some conventional methods. Fuzzy theory and FAHP are introduced to establish the ordering model of multiple QCs. Then, the subjective weights are determined by the method of fuzzy comparison which is expressed by the triangular fuzzy number, while the objective weights of the QCs are obtained by Information Entropy. In addition, the importance of multiple QCs is sequenced by Fuzzy Borda method combining with FAHP and Information Entropy, and finally determined according to the value of fuzzy Borda. The importance sequencing problem is effectively solved by this method based on considering the subjective and the objective together. Finally, a practical example is illustrated to show the rationality and feasibility of the integrated method.(3) Based on the particularity and complexity in the design process of mechanical and electrical products, integrated with the multi-source and multi-stage process of Quality Function Deployment (QFD), this paper increased mapping machining process to identify the elements of machining quality. Then, the user needs and quality requirements (QRs) are mapped into the QCs of mechanical and electrical products, and the product design, manufacturing and using stages are applied to form a closed loop system, so that product quality information is timely, roundly and accurately reflected in the product stages, so as to improve design quality and customer satisfaction. In order to improve the machining accuracy, the accuracy design QCs for mechanical and electrical products should be analyzed and evaluated. This paper studies the transfer relationships between QRs and accuracy design QCs, and proposes a reverse mapping model combines rough set (RS) with house of quality (HoQ) in QFD. RS theory is first applied to simplify and extract the QRs of the machining quality requirements. Then, a novel concept known as approximate accuracy rough number is proposed to determine the final importance weights of the QRs. Furthermore, a reverse mapping model is designed to translate the main QRs into accuracy design QCs, which would help product designers achieve the design goals in product development. The applicability of the proposed reverse mapping model is demonstrated by an illustrative example of CNC machine.(4) By analyzing the specific requirements of multiple DQCs of design scheme for mechanical and electrical products, the design optimization program decision-making process are proposed to ensure the design quality based on multi-attribute index system. In this paper, we propose a novel method combined the improved Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) with Elimination and Choice Translating Reality (ELECTRE) I for reliability design scheme decision of mechanical and electrical products. Based on the improved TOPSIS method and ELECTRE I, the decision model is built to select the optimal design scheme. The improved TOPSIS method is applied to determinate the weights of reliability design factors through the decision model. ELECTRE I method is then designed to rank reliability design scheme in order of decision maker’s preference. To evaluate performance of the developed algorithm, an illustrative example of CNC machine is given. The computational results show that the proposed approach is reliable and performs well.(5) The paper analyzed the present application and proposed the evaluation index system model for multivariate QCs of mechanical and electrical products. A comprehensive evaluation method based on fuzzy matter element (FME) analysis is proposed to address problems of multiple performance quality evaluation for mechanical and electrical products. The FME theory is utilized to analyze multi-scale QCs and attributes. Then, the comprehensive quality evaluation model for mechanical and electrical products is founded by using FME analysis method. Analytical network process (ANP) and IE are applied to obtain the objective index weights. Furthermore, the combinational weight is calculated with the optimal mathematics model based on the least square method. Finally, the FME method is used to determine the assessment result. The case study indicated that the method had a certain rationality and feasibility.更多还原