基于Kriging模型的叶盘系统多场耦合动力学多学科设计优化理论与试验研究

项目名称： 基于Kriging模型的叶盘系统多场耦合动力学多学科设计优化理论与试验研究

项目编号： No.51275081

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 机械、仪表工业

项目作者： 袁惠群

作者单位： 东北大学

项目金额： 78万元

中文摘要： 针对压气机叶盘系统设计分析优化的瓶颈问题，提出了多场耦合作用叶盘系统动力学多学科设计优化框架，以某压气机叶盘系统为对象开展研究。研究内容为：叶盘系统多场耦合作用动力学分析与模块集成；叶片失谐参数识别；压气机单级叶盘系统动测试验与多场耦合动力学分析模块可靠性验证；基于近似技术/智能优化算法的叶盘系统多场耦合动力学多学科设计优化。关键问题为：静叶尾迹影响的压气机单级可压缩旋转流场CFD建模与仿真；耦合界面载荷信息传递原理；流场网格更新策略；叶盘系统动力学多学科设计优化准则。创新点为：提出综合考虑气动、传热、叶盘结构耦合作用、接触非线性、离心刚化、旋转软化的叶盘系统动力学模型；提出基于Kriging模型与假设弹性体法相结合的耦合界面载荷信息传递及流场网格更新方法；基于Isight平台的叶盘系统多场耦合动力学分析模块集成；基于Kriging模型近似技术/改进遗传算法的叶盘系统动力学多学科设计优化。

中文关键词： 叶盘系统；多场耦合动力学；滑移网格；Kriging模型；优化设计

英文摘要： It takes the bladed disk system of gas compressor as the object, and presents a multidisciplinary design optimization to dynamic characteristics of blade-disk under multi-physical coupling field. Research content: With the multi-physical coupling field acting on bladed disk system, the dynamic characteristics is analyzed, then all the modules are integrated together. The mistuned parameters of blades are identified. The single-stage bladed disk of compressor is tested dynamically and the reliability performance of module of dynamic characteristics analysis is verified.Based on approximation method and intelligent optimization arithmetic, multidisciplinary design optimization of dynamics of bladed disk is developed.Key issues: CFD modeling and simulation on compressible and rotational flow field of single blade disk with the effect of the wake of stationary blades , and the principle of load transfer between coupling interfaces, and the strategy of flow field grid updating, and the multidisciplinary design optimization rule of dynamics of bladed disk system .Main innovations: Considering the influence of aerodynamic load, heat transference, coupling of bladed disk, contact nonlinear, rigidization caused by centrifugal effect and spin softening, systemic dynamics model is established. Based on Kriging model and el

英文关键词： blade-disk structure；multi-field coupling dynamics；sliding mesh；Kriging model；optimization design