考虑数值噪声的电子机箱设备多学科优化研究

发布时间：2018-03-09 01:36

本文选题：电子机箱　切入点：多学科优化　出处：《西安电子科技大学》2012年硕士论文　论文类型：学位论文

【摘要】：电子机箱设备设计涉及机械振动学、传热学和电磁学等多个学科。随着电子设备向高密度小型化发展，各学科之间的耦合问题变得愈加突出。必须从系统集成的角度，运用数值优化方法对其进行多学科综合优化设计。本文在某预研基金的支持下，围绕电子设备的多学科综合优化问题展开研究，主要工作如下： 1）阐述了电子机箱设备机-电-热多学科仿真分析的调用流程，，以及将结构位移场信息传递给热分析和电磁场分析模块的技术路线。分析了电子机箱设备多学科优化的主要特点。 2）由于数值噪声的影响，电子机箱设备优化的设计函数常常是不光滑或不连续的，为学科之间的解耦和优化计算带来较大困难。为此，借鉴全局优化的相关理论，提出了考虑数值噪声的电子机箱设备多学科全局优化方法。首先，在解耦方法上，根据同时分析和设计思想，将解耦问题转化为一个优化问题，在学科层引入Kriging替代模型以便过滤数值噪声，并采用极大似然估计法确定新增样本点的位置，较大程度上减少了解耦分析所需的重分析次数；其次，采用罚函法建立了耦合系统的多学科优化模型。通过一个典型的热-电耦合算例验证了模型和方法的有效性。 3）开发了电子机箱设备多学科综合优化软件平台的优化模块，在考虑多场耦合的情况下进行电子机箱设备结构综合分析优化。编写了优化模块与CAD建模以及各学科分析模块之间的接口程序，解决了优化计算中流程定制、算法集成、变量归并等关键技术。并进行了案例验证。
[Abstract]:The design of electronic crate equipment involves mechanical vibration, heat transfer, electromagnetism and so on. With the development of electronic equipment towards high density miniaturization, the problem of coupling between various disciplines becomes more and more prominent. With the support of a pre-research fund, this paper focuses on the multidisciplinary integrated optimization of electronic equipment. The main work is as follows:. 1) the transfer flow of the simulation analysis of the electronic crate equipment, the electric-thermal multidisciplinary simulation, is described. The main characteristics of multi-disciplinary optimization of electronic crate equipment are analyzed, and the technical route of transferring the information of structural displacement field to the modules of thermal analysis and electromagnetic field analysis is presented. 2) due to the influence of numerical noise, the design function of electronic chassis equipment optimization is often non-smooth or discontinuous, which makes it difficult to decouple and optimize calculation between disciplines. Therefore, the relevant theories of global optimization are used for reference. A multi-disciplinary global optimization method for electronic crate equipment considering numerical noise is proposed. Firstly, the decoupling problem is transformed into an optimization problem according to the simultaneous analysis and design idea. In order to filter the numerical noise by introducing the Kriging substitute model at the subject level, and using the maximum likelihood estimation method to determine the location of the new sample points, the reanalysis times required for decoupling analysis are reduced to a great extent. Secondly, the maximum likelihood estimation method is used to determine the position of the new sample points. A multi-disciplinary optimization model of coupled system is established by using penalty function method, and the validity of the model and method is verified by a typical thermal-electric coupling example. 3) the optimization module of multi-disciplinary integrated optimization software platform for electronic chassis equipment is developed. The integrated analysis and optimization of electronic chassis equipment structure are carried out under the consideration of multi-field coupling. The interface program between optimization module, CAD modeling and each subject analysis module is compiled, which solves the process customization and algorithm integration in optimization calculation. Variable merging and other key techniques.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH122

【参考文献】