본문 바로가기
HOME> 논문 > 논문 검색상세

학위논문 상세정보

실험계획법을 이용한 CNC 스핀들형상 최적화에 관한 연구 원문보기

  • 저자

    배경태

  • 학위수여기관

    경상대학교 대학원

  • 학위구분

    국내석사

  • 학과

    정밀기계공학전공 재료 파괴 및 설계 전공

  • 지도교수

  • 발행년도

    2014

  • 총페이지

    ?, 72 p.

  • 키워드

    실험계획법 반응표면법 스핀들 유한요소해석;

  • 언어

    kor

  • 원문 URL

    http://www.riss.kr/link?id=T13534309&outLink=K  

  • 초록

    To improve the accuracy of a machine, research needs to be conducted on the relationship between output variables and design variables of spindle shaped part from the viewpoint of thermal and static. Therefore, research (by examining the correlation of each variable to find optimum conditions) has been carried out. Moreover, DOE(design of experiments) is used in extensive field. The purpose of DOE is finding the effect of design variable based on the length and diameter of the spindle. In addition, aiming to improve the accuracy, the relation output variables and optimization of output variables on thermal deformation and deformation are investigated. Model of this study is the grinding spindle in which a hydrostatic bearing is applied. Initial model is designed by using ANSYS Design modeler. This model performed preliminary analysis based on the experimental results of the previous studies. Preliminary analysis analyzes the thermal deformation and static deformation of the spindle. Futhermore, to optimize the spindle shape, sample is extracted by the rotational center composite design method on surface reaction analysis. The influence of output variables and design variables is compared through main effect analysis. In the consideration of the influence, response surfaces are created through ANSYS and Minitab. Generated response surface were applied to Kriging model. The Kriging model is applied in order to interpolate the extreme design area of the reaction surface. Additionally, in order to verify the reliability of the response surface, sensitivity analysis and the coefficient of determination are observed. To optimize model, genetic algorithm method and screening method are selected. In this way, optimized model is compared with the initial model. In comparison with the initial model, the deformation of the optimized model designed by DOE is decreased to 4.1 ㎛, while the thermal deformation is decreased to 1.2 ㎛. Therefore, it is efficient to design spindle shaped part through DOE to improve the accuracy of the machine.


 활용도 분석

  • 상세보기

    amChart 영역
  • 원문보기

    amChart 영역