金刚石滚轮修整时的径向圆跳动状态在线判别

付庭斌; 朱振伟; 张瑞; 赵华东

doi:10.13394/j.cnki.jgszz.2021.0119

金刚石滚轮修整时的径向圆跳动状态在线判别

doi: 10.13394/j.cnki.jgszz.2021.0119

郑州大学机械与动力工程学院, 郑州 450001

基金项目: 郑州市协同创新项目（18XTZX12006）。

详细信息

通讯作者:
赵华东，男，1978年生，博士生导师。主要研究方向：智能制造。E-mail: 82662906@qq.com

中图分类号: TQ164
计量
- 文章访问数: 502
- HTML全文浏览量: 249
- PDF下载量: 35
- 被引次数: 0
出版历程
- 收稿日期: 2021-06-21
- 修回日期: 2021-09-24
- 刊出日期: 2022-05-27

On-line discrimination of radial runout state during diamond roller trimming

College of Mechanical and Power Engineering, Zhengzhou University, Zhengzhou 450001, China

摘要

摘要: 金刚石滚轮修整砂轮时的性能受其径向圆跳动的影响，而其径向圆跳动状态判别的智能化程度较低。为此，对金刚石滚轮修整状态下的径向圆跳动磨削声发射信号，提出一种基于小波分解和SVM的在线检测方法。将磨削声发射信号通过小波变换并分解，提取小波分解系数的有效值、方差及能谱系数3种特征参数。结果表明：将3种特征参数彼此组合输入到SVM中进行状态识别时的准确率都在96.0%以上；3种特征参数同时输入时的准确率最高，达到了98.3%。该检测方法具有实际应用价值。
- 金刚石滚轮 /
- 声发射 /
- 小波变换 /
- 修整状态识别 /
- 支持向量机
Abstract: The performance of diamond roller when dressing grinding wheel was affected by its radial runout, but the intelligent degree of judging its radial runout state was low. Therefore, an on-line detection method based on wavelet decomposition and SVM was proposed for the grinding acoustic emission signal of radial runout under the trimming state of diamond roller. The grinding acoustic emission signal was transformed and decomposed by wavelet transform, and the three characteristic parameters of wavelet decomposition coefficients were extracted, which were effective value, variance value and energy spectrum coefficient. The results show that the accuracies of combining the three feature parameters into SVM for state recognition are more than 96.0%. When the three characteristic parameters are input at the same time, the accuracy is the highest, reaching 98.3%. The detection method has practical application value.
- diamond roller /
- acoustic emission /
- wavelet transform /
- trimming state recognition /
- support vector machine(SVM)

HTML全文

图 1 金刚石滚轮轮廓

Figure 1. Diamond roller profile

下载: 全尺寸图片幻灯片

图 2 金刚石滚轮实物

Figure 2. Real diamond roller

下载: 全尺寸图片幻灯片

图 3 CVD金刚石模型

Figure 3. CVD diamond model

下载: 全尺寸图片幻灯片

图 4 修整试验平台

Figure 4. Trimming test platform

下载: 全尺寸图片幻灯片

图 5 粗修状态下的AE信号

Figure 5. AE signal in rough trimming state

下载: 全尺寸图片幻灯片

图 6 精修状态下的AE信号

Figure 6. AE signal in finishing state

下载: 全尺寸图片幻灯片

图 7 修整完成状态下的AE信号

Figure 7. AE signal in trimming completed state

下载: 全尺寸图片幻灯片

图 8 粗修状态下的AE信号频谱

Figure 8. AE signal spectrum in rough trimming state

下载: 全尺寸图片幻灯片

图 9 精修状态下的AE信号频谱

Figure 9. AE signal spectrum in finishing state

下载: 全尺寸图片幻灯片

图 10 修整完成时的AE信号频谱

Figure 10. AE signal spectrum in trimming completed state

下载: 全尺寸图片幻灯片

图 13 不同修整状态下的AE信号小波能谱系数

Figure 13. Wavelet energy spectrum coefficients of AE signals under different trimming states

下载: 全尺寸图片幻灯片

图 11 不同修整状态下的AE信号小波系数有效值

Figure 11. Effective values of wavelet coefficients of AE signals under different trimming states

下载: 全尺寸图片幻灯片

图 12 不同修整状态下的AE信号小波系数方差

Figure 12. Variances of AE signal wavelet coefficients under different trimming states

下载: 全尺寸图片幻灯片

表 1 修整试验参数

Table 1. Trimming test parameters

参数	类型或取值
滚轮直径 D₁ / mm	130
滚轮宽度 W₁ / mm	16
砂轮直径 D₂ / mm	200
砂轮中金刚石粒度代号	120/140
砂轮宽度 W₂ / mm	3
砂轮中金刚石浓度 C / %	120
砂轮结合剂	V
砂轮转速 n₁ / (r·min⁻¹)	4 000
滚轮转速 n₂ / (r·min⁻¹)	70
纵向走刀速度 n₃ / (mm·min⁻¹)	2.4
粗修时的进给量 s₁ / μm	8
精修时的进给量 s₂ / μm	2
精修完成时的进给量 s₃ / μm	2

下载: 导出CSV

表 2 小波系数有效值

Table 2. Effective values of wavelet coefficients

序号	a₅	d₅	d₄	d₃	d₂	d₁
1	0.177	0.210	0.051	0.026	0.009	0.005
2	0.255	0.201	0.047	0.024	0.008	0.004
3	0.215	0.136	0.037	0.021	0.008	0.004
4	0.023	0.122	0.029	0.016	0.006	0.004
$ \vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $
597	0.197	0.195	0.046	0.027	0.009	0.004
598	0.207	0.160	0.041	0.021	0.071	0.004
599	0.238	0.124	0.032	0.018	0.061	0.004
600	0.207	0.160	0.041	0.021	0.071	0.004

下载: 导出CSV

表 3 小波系数方差

Table 3. Variances of wavelet coefficients

序号	a₅	d₅	d₄	d₃	d₂	d₁
1	3.1×10⁻²	4.4×10⁻²	0.3×10⁻²	6.8×10⁻⁴	8.1×10⁻⁵	2.0×10⁻⁵
2	3.5×10⁻²	3.1×10⁻²	0.1×10⁻²	5.0×10⁻⁴	5.9×10⁻⁵	1.8×10⁻⁵
3	3.6×10⁻²	2.4×10⁻²	0.2×10⁻²	3.6×10⁻⁴	6.2×10⁻⁵	2.3×10⁻⁵
4	3.5×10⁻²	1.6×10⁻²	0.1×10⁻²	3.9×10⁻⁴	4.4×10⁻⁵	1.7×10⁻⁵
$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $
597	3.6×10⁻²	3.6×10⁻²	2.2×10⁻²	6.1×10⁻⁴	5.6×10⁻⁵	1.9×10⁻⁵
598	2.8×10⁻²	1.8×10⁻²	1.3×10⁻²	5.5×10⁻⁴	6.5×10⁻⁵	1.8×10⁻⁵
599	3.4×10⁻²	3.2×10⁻²	1.9×10⁻²	5.2×10⁻⁴	5.8×10⁻⁵	2.0×10⁻⁵
600	3.7×10⁻²	2.3×10⁻²	1.2×10⁻²	3.4×10⁻⁴	3.8×10⁻⁵	1.9×10⁻⁵

下载: 导出CSV

表 4 小波能谱系数

Table 4. Wavelet energy spectrum coefficients

序号	a₅	d₅	d₄	d₃	d₂	d₁
1	67.23	0.258	0.394	1.749	2.552	27.82
2	81.04	0.312	0.297	1.283	1.804	15.26
3	71.53	0.296	0.266	1.082	2.492	24.33
4	67.07	0.632	0.556	2.219	3.257	26.26
$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $	$\vdots $
597	52.97	0.748	1.031	3.956	4.895	36.61
598	61.74	0.256	0.419	1.875	3.101	32.61
599	53.27	0.239	0.242	0.983	1.323	43.94
600	53.94	0.329	0.357	1.620	2.447	41.31

下载: 导出CSV

表 5 修整状态分类测试的准确率

Table 5. Accuracy of trimming state classification test

输入的AE信号特征参数	输入特征个数m	SVM的准确率 A_c / %
$ {X}_{rms}^{m} $和$ {V}_{var}^{m} $	12	96.8
$ {X}_{rms}^{m} $和$ {\eta }^{k} $	12	96.5
$ {V}_{var}^{m} $和$ {\eta }^{k} $	12	97.2
$ {X}_{rms}^{m} $、$ {V}_{var}^{m} $和$ {\eta }^{k} $	18	98.3

下载: 导出CSV

参考文献(13)

[1]	龚子维. 磨削砂轮钝化识别与声发射特征参数研究 [D]. 南京: 南京大学, 2020. GONG Ziwei. Identification of blunting state of grinding wheel and study on characteristic parameters of acoustic emission [D]. Nanjing: Nanjing University, 2020.
[2]	吕长飞, 吴小玉, 王茵, 等. 外圆磨削颤振监测方法设计 [J]. 机床与液压,2019,47(8):166-168, 66. doi: 10.3969/j.issn.1001-3881.2019.08.036 LYU Changfei, WU Xiaoyu, WANG Yin, et al. Design of chatter detection in external cylindrical grinding [J]. Machine Tool & Hydraulics,2019,47(8):166-168, 66. doi: 10.3969/j.issn.1001-3881.2019.08.036
[3]	LIU G J, WANG Q, KANG R K. Study on the wavelet transform based monitor signal processing method for grinding wheel dull [J]. Key Engineering Materials,2008,375/376:598-602. doi: 10.4028/www.scientific.net/KEM.375-376.598
[4]	徐水竹, 杨京, 张仲宁, 等. 基于小波变换的磨削噪声在线监测方法 [J]. 声学技术,2011,30(3):275-279. doi: 10.3969/j.issn.1000-3630.2011.03.014 XU Shuizhu, YANG Jing, ZHANG Zhongning, et al. On-line grinding state monitoring with wavelet transform of noise signal [J]. Acoustical Technology,2011,30(3):275-279. doi: 10.3969/j.issn.1000-3630.2011.03.014
[5]	巩亚东, 吕洋, 王宛山, 等. 基于多传感器融合的磨削砂轮钝化的智能监测 [J]. 东北大学学报,2003(3):248-251. GONG Yadong, LYU Yang, WANG Wanshan, et al. Intelligent monitoring for grinding wheel passivation based on multi-sensor fusion [J]. Journal of Northeastern University,2003(3):248-251.
[6]	陈俊奇. 基于磨削声音信号特征的砂带磨损状态监测方法研究 [D]. 上海: 上海交通大学, 2018. CHEN Junqi. Acoustic signal based grinding belt wear condition monitoring [D]. Shanghai: Shanghai Jiao Tong University, 2018.
[7]	王起硕. 基于多传感器融合的微晶刚玉砂轮磨削性能在线检测系统的研究 [D]. 济南: 山东大学, 2017. WANG Qishuo. Online detection system of grinding performance of microcrystalline corundum grinding wheel based on multi-sensor fusion [D]. Jinan: Shandong University, 2017.
[8]	丁宁, 段景淞, 石建, 等. 基于声发射砂轮磨损监测系统的研究 [J]. 南京航空航天大学学报,2020,52(1):48-52. DING Ning, DUAN Jingsong, SHI Jian, et al. Research on grinding wheel wear monitoring system based on acoustic emission [J]. Journal of Nanjing University of Aeronautics & Astronautics,2020,52(1):48-52.
[9]	朱欢欢, 迟玉伦, 闻章, 等. 断续磨削表面烧伤机理与在线监测方法研究 [J]. 表面技术,2021(9):379-389. ZHU Huanhuan, CHI Yulun, WEN Zhang, et al. Research on burn mechanism of intermittent grinding surface and online monitoring method [J]. Surface Technology,2021(9):379-389.
[10]	CHEN X, OPOZ T T. Effect of different parameters on grinding efficiency and its monitoring by acoustic emission [J]. Production & Manufacturing Research,2016,4(1):190-208.
[11]	郭力, 郭君涛, 霍可可. 金刚石砂轮与氧化锆磨削接触的声发射监测 [J]. 制造技术与机床,2019(5):94-98. GUO Li, GUO Juntao, HUO Keke. Acoustic emission monitoring on grinding contact between diamond grinding wheel and zirconia [J]. Manufacturing Technology and Machine Tool,2019(5):94-98.
[12]	郑敏敏, 高小榕, 谢海鹤. 心电信号小波去噪的改进算法研究 [J]. 中国生物医学工程学报,2017,36(1):114-118. doi: 10.3969/j.issn.0258-8021.2017.01.015 ZHENG Minmin, GAO Xiaorong, XIE Haihe. Research on an improved algorithm for wavelet denoising of ECG [J]. Chinese Journal of Biomedical Engineering,2017,36(1):114-118. doi: 10.3969/j.issn.0258-8021.2017.01.015
[13]	李鹏瑞. 基于贝叶斯分类器的煤层底板破坏程度预测 [D]. 济南: 山东科技大学, 2016 LI Pengrui. Prediction of failure degree of coal seam floor based on bayesian classifier [D]. Jinan: Shandong University of Science and Technology, 2016.