基于声发射的KDP晶体表面波纹度智能控制研究  

Research on Intelligent Control for KDP Crystal Waviness Based on Acoustic Emission

作　　者：关佳亮[1] 任勇[1] 陈玲[1] 赵显辉[1] 朱生根 GUAN Jia-liang REN Yong CHEN Ling ZHAO Xian-hui ZHU Sheng-gen(College of Mechanical Electrical, Beijing University of Technology, Beijing 100124, China Beijing Precision Machinery ＆ Engineering Research Co. , Ltd. , Beijing 101312, China)

机构地区：[1]北京工业大学机电学院,北京100124 [2]北京工研精机股份有限公司,北京101312

出　　处：《组合机床与自动化加工技术》2016年第10期91-94,98,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家科技重大专项:特殊晶体材料单点金刚石飞刀切削超精密机床(2011ZX04004-042)

摘　　要：针对KDP晶体超精密加工过程中切削效率较低、难以捕捉表面状态与追求高效率、高表面质量加工存在矛盾的问题,研究基于声发射的KDP晶体表面波纹度智能监测,建立反馈机制,智能在线控制KDP晶体表面波纹度。首先,利用二次通用回归旋转组合法设计工艺参数组合,采用单点金刚石飞刀切削技术(SPDT)对KDP晶体进行不同工艺参数下波纹度采样实验,同时采集了不同表面波纹度状态下的声发射信号;然后,分析可有效监测波纹度状态的时域和频域特征,利用MATLAB软件对声发射信号进行多辨小波分解和重构,提取各子频段的能量百分比,筛选出与波纹度密切相关的特征小波频段;最后,利用RBF神经网络技术建立表面波纹度状态与信号特征值的映射系统,训练并测试了网络系统的可靠性,实现了表面波纹度状态的智能监测。In the process of ultra-precision machining of KDP crystals,it is difficult to discriminate the surface quality and ensure cutting efficiency,but actually pursue the high efficiency and high precision machining,so it is necessary to do research of KDP crystal surface waviness intelligent monitoring based on the acoustic emission,final develop the feedback mechanism and intelligent online control system. First of all,the KDP crystal waviness process optimization parameters combination was designed by using Quadratic regression rotation and waviness sampling experiments was carried out by single point diamond turning（ SPDT） technology,meanwhile collected AE signal under different surface waviness state; Then,analyzed the time domain and frequency domain characteristics which can effectively monitor of waviness state,using the MATLAB software for acoustic emission signal wavelet decomposition and reconstruction,extracted the percentage of energy spectrum,filtrated the characteristics of wavelet spectrum was closely related with waviness; Finally,the contoral system in which signal characteristic value mapping surface waviness state was established based on RBF neural network technology,trained and tested the reliability of the network system,monitoring the state of the surface waviness was implemented.

关 键 词：KDP晶体 表面波纹度 声发射 小波分析 神经网络 

分 类 号：TH161[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]