摘要
针对函数式复杂且观测点位置固定的非线性profile监控问题,提出一种基于提升小波重构与支持向量数据描述(Support Vector Data Description,SVDD)的方法来监控非线性轮廓图的异常波动。首先采用提升小波对原始数据进行去噪重构处理,还原初始数据的有用信息。其次对去噪后的重构数据进行SVDD模型训练,通过Bootstrap重采样的方法确定控制限,提高模型寻参效率。最后对训练的模型进行生产过程的异常轮廓监控性能研究。通过计算机仿真实验表明,该方法在质量监控过程中的平均链长较短,能及时发现异常轮廓。
In order to solve the problem of nonlinear profile monitoring with complex functions and fixed observation points,a method based on lifting wavelet reconstruction and support vector data description(Support Vector Data Description,SVDD)is proposed to monitor the abnormal fluctuations of nonlinear contour images.First,the lifting wavelet is used to reconstruct the original data and restore the useful information of the original data.Secondly,the reconstructed data after denoising is trained by SVDD model,and the control limit is determined by the method of Bootstrap to improve the efficiency of finding model parameters.Finally,the abnormal profile monitoring performance of the training model is studied.The results of computer simulation show that the average chain length of this method is short in the process of quality control and the abnormal profile can be found in time.
作者
李虹灿
田光杰
LI Hong-can;TIAN Guang-jie(Business School,Zhengzhou University,Zhengzhou 450001,China)
出处
《组合机床与自动化加工技术》
北大核心
2019年第9期152-156,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金项目(71672182,71711540309,71272207)
国家自然科学基金重点项目(U1604262)
