摘要
在精密机械热动态过程的离散化模型基础上,提出一种识别精密机械热动态特性参数的新方法——奇异值分解算法.在分析精密机械零部件的非定常导热问题时,由有限元法获得热动态过程空间离散化模型.采用热模态分析方法实现离散化模型解耦变换,模态坐标下的特性参数为热特征值(广义时间常数的倒数).辨识热动态特性参数的方法是通过构造热脉冲响应矩阵,采用矩阵奇异值分解的方法,以最少的参数和最小的阶次来描述精密机械热动态过程,进而求得热特征值.实测数据表明,该方法能有效地识别热特征值和快速估算出热平衡时间.
A singular value decomposition algorithm for identifying these characteristics was proposed based on the discrete model of the thermal dynamics process of precision machine. The non-stationary thermal conduction and the thermal characteristics of a machine tool structure were modeled by FEM. Thermal modal analysis was introduced to diagonalize system equations and the characteristics parameters on the modal coordinate are thermal eigenvalues. A approach was presented deriving thermal dynamics characteristics with minimum parameter number and order. The singular value decomposition technique was used to identify the thermal modal parameter with thermal pulse response and obtain thermal eigenvalues. An example shows that the proposed method is effective for identifying thermal eigenvalues and calculating thermal balance time.
出处
《浙江大学学报(工学版)》
EI
CAS
CSCD
北大核心
2004年第4期474-477,共4页
Journal of Zhejiang University:Engineering Science
基金
国家自然科学基金资助项目(5985018).