上海交通大学学报 ›› 2020, Vol. 54 ›› Issue (7): 674-681.doi: 10.16183/j.cnki.jsjtu.2020.99.010

• 学报(中文) • 上一篇    下一篇

基于支持向量机回归的曲面零件涡流测距标定方法

陶正瑞1,党嘉强1,徐锦泱1,安庆龙1,陈明1,王力2,任斐2   

  1. 1. 上海交通大学 机械与动力工程学院; 机械系统与振动国家重点实验室, 上海 200240; 2. 上海航天设备制造总厂有限公司, 上海 200245
  • 出版日期:2020-07-28 发布日期:2020-07-31
  • 通讯作者: 安庆龙,男,副研究员,博士生导师, 电话(Tel.): 021-34206317; E-mail: qlan@sjtu.edu.cn.
  • 作者简介:陶正瑞(1996-),男,安徽省天长市人,硕士生,研究方向为加工过程监控.
  • 基金资助:
    国家“高档数控机床与基础制造装备”科技重大专项课题(2017ZX04005001)

Eddy Current Distance Measurement Calibration Method for Curved Surface Parts Based on Support Vector Machine Regression

TAO Zhengrui 1,DANG Jiaqiang 1,XU Jinyang 1,AN Qinglong 1,CHEN Ming 1,WANG Li 2,REN Fei 2   

  1. 1. School of Mechanical Engineering; State Key Laboratory of Mechanical System and Vibration, Shanghai Jiao Tong University, Shanghai 200240, China; 2. Shanghai Aerospace Equipments Manufacturer Co., Ltd., Shanghai 200245, China
  • Online:2020-07-28 Published:2020-07-31

摘要: 以燃料贮箱筒段外表面聚氨酯泡沫层厚度测量精度要求为例,开展4种常见曲率筒段试件标定试验;提出基于支持向量机回归的涡流测距标定方法,建立提离距离预测模型(LDPM),用于曲面零件涡流测距;研究曲率对涡流测距测量误差的影响规律.根据测量误差产生的原因将测量误差分为两部分:表面曲率误差和其他误差,分析误差分量在传感器量程范围内相对大小的变化规律,为曲面测量误差补偿提供参考依据.此外,对比分析LDPM与常用标定方法获得的标定函数在测量精度、计算速率方面的优劣,为曲面零件涡流测距选用何种标定方法提供建议.研究结果表明:LDPM在传感器量程初始阶段,表面曲率误差对曲率不敏感;在量程中点附近,表面曲率误差绝对值随着曲率的增加,先减小后上升;在终点区域,表面曲率误差绝对值随着曲率的增加而上升.在整个量程范围内的不同测量区间,其他误差保持不变.此外,LDPM可以将测量误差控制在[-0.5, 0.5]mm,精度与5项多峰高斯拟合相当,高于4次多项式拟合,能够满足厚度测量精度要求,无损检测操作方便.

关键词: 涡流测距, 支持向量机回归, 误差补偿, 多峰高斯拟合

Abstract: Based on the accuracy requirements for the thickness of the polyurethane form layer on the outer surface of the fuel tank, the calibration testing of four common curvature section specimens is conducted. Besides, the eddy current ranging calibration method based on support vector machine regression analysis is proposed to establish the lifting distance prediction model (LDPM), which is used for curved surface eddy current distance measurement. The influence of curvature on the measurement errors of eddy current ranging is studied. The measurement errors are divided into the surface curvature errors and other errors according to the cause of the error. In the different ranges of the sensor range, the relative variation of the error components is analyzed, which provides a basis for surface measurement error compensation. In addition, the comparative analysis of LDPM and the calibration function obtained by using common calibration methods in terms of measurement accuracy and calculation rate are performed, which provides suggestions for the selection of the calibration method for eddy current ranging of curved parts. The results show that the surface curvature error is not sensitive to the curvature in the initial stage of the sensor range. Near the midpoint of the range, the absolute value of the surface curvature error decreases first and then increases with the increase of curvature. In the terminal region, the absolute value of the surface curvature error increases with the rise of curvature, while other errors remain constant within different measurement ranges over the entire range. Moreover, the LDPM can control the measurement error within [-0.5, 0.5]mm, and the accuracy is comparable to five multi-peak Gaussian fitting, higher than the fourth-order polynomial fitting, which can meet the thickness measurement accuracy requirements and is convenient for non-destructive testing.

Key words: eddy current distance measurement, support vector machine regression, error compensation, multi-peak Gaussian fitting

中图分类号: