复杂产品制造过程中轮廓控制方法研究
发布时间：2018-12-29 13:58
【摘要】：轮廓控制是利用控制图理论对呈现特定函数关系（此函数关系称为轮廓）的过程（或产品）质量特性进行控制，是质量工程领域统计过程控制中的研究热点。本文将针对复杂产品制造加工过程中的实际问题，主要就轮廓内测量点位置发生变化、轮廓内数据存在相关性、轮廓特定变异预先已知和轮廓中变量为角度变量 等情况下的轮廓控制方法分别进行研究。首先，针对不同轮廓内测量点位置发生变化的情况，构建了基于线轮廓度误差的非线性轮廓联合控制图，给出了联合控制图的设计方法。比较研究表明，联合控制图在检测参数误差时非常有效，且优于其他基于差异度量的控制方法；而且在同时检测过程中的参数误差和形状误差时对于小偏移非常敏感。另外，提出了联合控制图的实施步骤和过程调整原则，有利于控制方法的现场应用和对实际 生产过程的调整。其次，当轮廓内数据存在相关关系时，构建了基于高斯过程模型的内部相关性线性轮廓联合控制图,提出了两种休哈特类型多元控制图分别用于监测第二阶段中轮廓线性部分和内部相关性。所提方法在与其他方法比较研究中考虑了不同受控轮廓内部相关强度。仿真比较结果显示，当检测线性部分的变异时，所提方法在受控线性轮廓内部相关性较强时的控制性能较好；在监控线性轮廓内部相关性时，所提方法对于相关关系的较大偏移更为敏感。而且，应用分析研究表明， 所提联合控制图在实际应用中较为方便，，而且能有效对过程进行控制。再次，对于预先已知过程发生特定类型变异的情况，建立了三种用于快速检测过程中线性轮廓形状变化的定向控制图。比较研究显示，所提定向控制图均能快速检测出过程中特定变异，且具有稳健性；当失控轮廓在受控轮廓附近波动时，所提方法对线性轮廓的形状变化较为敏感；当失控轮廓偏离受控轮廓较多时，在 轮廓内测量点个数较多时所提方法也能快速有效检测线性轮廓的形状变化。最后，研究了角度变量线性轮廓控制方法，建立了角度变量线性轮廓的第二阶段控制图，并提出了第一阶段控制方法。模拟仿真分析表明，所提第二阶段控制图能快速有效地检测过程变异。
[Abstract]:Contour control is a control of the process (or product) quality characteristics that presents a specific functional relationship (or product profile) by using the control chart theory. It is a hot topic in statistical process control in the field of quality engineering. Aiming at the practical problems in the process of manufacturing and processing of complex products, this paper will mainly focus on the change of the position of the measuring points in the contour and the correlation of the data in the contour. The contour control methods are studied under the condition that the specific variation of contour is known in advance and the variables in the contour are angular variables. Firstly, the joint control chart of nonlinear contour based on line contour error is constructed for the change of measuring point position in different contours, and the design method of joint control chart is given. The comparative study shows that the joint control chart is very effective in detecting parameter errors and is superior to other control methods based on difference measurement, and is very sensitive to small offset when the parameter errors and shape errors are detected simultaneously. In addition, the implementation steps of the joint control chart and the principle of process adjustment are put forward, which is beneficial to the field application of the control method and the adjustment of the actual production process. Secondly, when the data in the contour has the correlation relation, the joint control chart of the internal correlation linear contour based on Gao Si process model is constructed. Two types of Heinhart multivariate control charts are proposed to monitor the linear and internal correlation of the contour in the second stage. In comparison with other methods, the proposed method takes into account the internal correlation strength of different controlled contours. The simulation results show that the proposed method has better control performance when the variation of the linear part is detected when the internal correlation of the controlled linear contour is strong. The proposed method is more sensitive to the large deviation of the correlation relationship when monitoring the internal correlation of the linear profile. Furthermore, the application analysis shows that the proposed joint control chart is more convenient in practical application and can effectively control the process. Thirdly, three directional control charts are established for rapid detection of changes in linear contour shape when a specific type of variation occurs in a process known in advance. The comparative study shows that the proposed directional control charts can quickly detect the specific variation in the process and have robustness, and when the runaway contour fluctuates near the controlled contour, the proposed method is more sensitive to the change of the shape of the linear contour. When the runaway contour deviates more from the controlled contour, the proposed method can detect the shape change of the linear contour quickly and effectively when the number of measured points in the contour is more. Finally, the linear contour control method of angle variable is studied, the second stage control diagram of linear contour of angle variable is established, and the first stage control method is proposed. Simulation results show that the proposed second stage control chart can detect process variation quickly and effectively.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TH16
本文编号：2394918
[Abstract]:Contour control is a control of the process (or product) quality characteristics that presents a specific functional relationship (or product profile) by using the control chart theory. It is a hot topic in statistical process control in the field of quality engineering. Aiming at the practical problems in the process of manufacturing and processing of complex products, this paper will mainly focus on the change of the position of the measuring points in the contour and the correlation of the data in the contour. The contour control methods are studied under the condition that the specific variation of contour is known in advance and the variables in the contour are angular variables. Firstly, the joint control chart of nonlinear contour based on line contour error is constructed for the change of measuring point position in different contours, and the design method of joint control chart is given. The comparative study shows that the joint control chart is very effective in detecting parameter errors and is superior to other control methods based on difference measurement, and is very sensitive to small offset when the parameter errors and shape errors are detected simultaneously. In addition, the implementation steps of the joint control chart and the principle of process adjustment are put forward, which is beneficial to the field application of the control method and the adjustment of the actual production process. Secondly, when the data in the contour has the correlation relation, the joint control chart of the internal correlation linear contour based on Gao Si process model is constructed. Two types of Heinhart multivariate control charts are proposed to monitor the linear and internal correlation of the contour in the second stage. In comparison with other methods, the proposed method takes into account the internal correlation strength of different controlled contours. The simulation results show that the proposed method has better control performance when the variation of the linear part is detected when the internal correlation of the controlled linear contour is strong. The proposed method is more sensitive to the large deviation of the correlation relationship when monitoring the internal correlation of the linear profile. Furthermore, the application analysis shows that the proposed joint control chart is more convenient in practical application and can effectively control the process. Thirdly, three directional control charts are established for rapid detection of changes in linear contour shape when a specific type of variation occurs in a process known in advance. The comparative study shows that the proposed directional control charts can quickly detect the specific variation in the process and have robustness, and when the runaway contour fluctuates near the controlled contour, the proposed method is more sensitive to the change of the shape of the linear contour. When the runaway contour deviates more from the controlled contour, the proposed method can detect the shape change of the linear contour quickly and effectively when the number of measured points in the contour is more. Finally, the linear contour control method of angle variable is studied, the second stage control diagram of linear contour of angle variable is established, and the first stage control method is proposed. Simulation results show that the proposed second stage control chart can detect process variation quickly and effectively.
【学位授予单位】：天津大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TH16
【参考文献】
相关期刊论文 前4条
1 何桢;王晶;李ng范;;基于Bootstrap方法的过程能力指数区间估计[J];工业工程;2008年06期
2 王晶;何桢;李ng范;;面向多品种小批量生产过程的Bootstrap控制图[J];工业工程;2009年02期
3 熊有伦;线轮廓度和面轮廓度的评定和判别[J];计量学报;1991年02期
4 李伯虎;;复杂产品制造信息化的重要技术——复杂产品集成制造系统[J];中国制造业信息化;2006年14期
本文编号：2394918
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2394918.html
