基于粒子群优化算法⁃长短时记忆模型的刀具磨损预测方法

doi:10.13229/j.cnki.jdxbgxb.20210778

摘要/Abstract

摘要：

为确保车削加工的表面质量和加工稳定性，实现对车刀磨损状态的实时准确监控，提出了基于小波阈值去噪、长短时记忆（LSTM）网络和粒子群优化算法（PSO）的刀具磨损状态预测模型。采用改进多项式阈值函数对刀具加速度振动信号进行去噪，构建了优质的信号输入样本。训练长短时记忆网络对刀具后刀面磨损值进行预测和磨损状态分类。利用粒子群优化算法对网络进行参数寻优，结果表明，提出的PSO-LSTM模型在预测和分类精度方面均优于未优化的LSTM网络。

关键词: 机械制造及其自动化, 车削, 刀具磨损, 状态监测, 深度学习, 长短时记忆网络

Abstract:

In order to ensure the surface quality and machining stability of turning， the real-time and accurate monitoring of turning tool wear state is realized. A tool wear state prediction model based on wavelet threshold denoising， Long and Short Time Memory（LSTM） network and Particle Swarm Optimization（PSO） was proposed. The improved polynomial threshold function was used to denoise the tool acceleration vibration signal， and the high quality signal input sample was constructed. The wear values of the tool rear face were predicted and the wear states were classified by training the LSTM network. The proposed PSO-LSTM model is superior to the unoptimized LSTM network in terms of prediction and classification accuracy by using PSO.

Key words: mechanical manufacturing and automation, turning, tool wear, condition monitoring, deep learning, long and short time memory network

中图分类号:

TP183

吴飞,农皓业,马晨浩. 基于粒子群优化算法⁃长短时记忆模型的刀具磨损预测方法[J]. 吉林大学学报(工学版), 2023, 53(4): 989-997.

Fei WU,Hao-ye NONG,Chen-hao MA. Tool wear prediction method based on particle swarm optimizationlong and short time memory model[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(4): 989-997.

图/表 14

图1

图2

表1

表2

图3

图4

图5

图6

表3

初始化PSO-LSTM模型参数"

网络参数	初始值
初始化权重	［-0.5， 0.5］
遗忘门偏置量	1或2
输入门偏置量	［0， 0.8］
输出门偏置量	［0， 0.8］
迭代次数	100
种群规模	50
学习因子 $c 1$	1.5
学习因子 $c 2$	1.7
隐藏层单元数 $m$	100
学习率 $r$	0.01

表3

图7

图8

图9

表4

表5

参考文献 21

1	Kurada S, Bradley C. A review of machine vision sensors for tool condition monitoring[J]. Computers in Industry, 1997, 34(1): 55-72.
2	Mobley R K. An Introduction to Predictive Maintenance [M]. New York: Elsevier, 2002.
3	Gao D, Liao Z R, Lv Z, et al. Multi-scale statistical signal processing of cutting force in cutting tool condition monitoring[J]. The International Journal of Advanced Manufacturing Technology, 2015, 80(9): 1843-1853.
4	Antić A, Šimunović G, Šarić T, et al. A model of tool wear monitoring system for turning[J]. Tehnicki vjesnik/Technical Gazette, 2013, 20(2): 247-254.
5	Moia D F G, Thomazella I H, Aguiar P R, et al. Tool condition monitoring of aluminum oxide grinding wheel in dressing operation using acoustic emission and neural networks[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2015, 37(2): 627-640.
6	吴雪峰, 刘亚辉, 毕淞泽. 基于卷积神经网络刀具磨损类型的智能识别[J]. 计算机集成制造系统, 2020, 26(10): 2762-2771.
	Wu Xue-feng, Liu Ya-hui, Bi Song-ze. Intelligent recognition of tool wear type based on convolutional neural networks[J]. Computer Integrated Manufacturing Systems, 2020, 26(10): 2762-2771.
7	Mohanraj T, Shankar S, Rajasekar R, et al. Tool condition monitoring techniques in milling process — a review[J]. Journal of Materials Research and Technology, 2020, 9(1): 1032-1042.
8	Cheng M H, Jiao L, Shi X C, et al. An intelligent prediction model of the tool wear based on machine learning in turning high strength steel[J]. Journal of Engineering Manufacture, 2020, 234(13): 1580-1597.
9	Kong D D, Chen Y J, Li N. Gaussian process regression for tool wear prediction[J]. Mechanical Systems and Signal Processing, 2018, 104: 556-574.
10	Du M H, Wang P X, Wang J H, et al. Intelligent turning tool monitoring with neural network adaptive learning[J]. Complexity, 2019(5A): 1-21.
11	Kong D D, Chen Y J, Li N, et al. Tool wear estimation in end milling of titanium alloy using NPE and a novel WOA-SVM model[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(7): 5219-5232.
12	Yen C L, Lu M C, Chen J. Applying the self-organizing feature map (SOM) algorithm to AE-based tool wear monitoring in micro-cutting[J]. Mechanical Systems and Signal Processing, 2013, 34(1/2): 353-366.
13	Ertunc H M, Loparo K A, Ocak H. Tool wear condition monitoring in drilling operations using hidden Markov models (HMMs)[J]. International Journal of Machine Tools and Manufacture, 2001, 41(9): 1363-1384.
14	郭亮, 高宏力, 张一文, 等. 基于深度学习理论的轴承状态识别研究[J]. 振动与冲击, 2016, 35(12): 166-170, 195.
	Guo Liang, Gao Hong-li, Zhang Yi-wen, et al. Research on bearing condition monitoring based on deep learning[J]. Journal of Vibration and Shock, 2016, 35(12): 166-170, 195.
15	Huang Z, Zhu J, Lei J, et al. Tool wear predicting based on multi-domain feature fusion by deep convolutional neural network in milling operations[J]. Journal of Intelligent Manufacturing, 2020, 31(4): 953-966.
16	Ma J, Luo D, Liao X, et al. Tool wear mechanism and prediction in milling TC18 titanium alloy using deep learning[J]. Measurement, 2021, 173: No. 108554.
17	Kong D, Chen Y, Li N. Monitoring tool wear using wavelet package decomposition and a novel gravitational search algorithm⁃least square support vector machine model[J]. Journal of Mechanical Engineering Science, 2019, 234(3): No. 095440621988731.
18	Xie Y, Zhang C, Liu Q. Tool wear status recognition and prediction model of milling cutter based on deep learning[J]. IEEE Access, 2020(9): 1616-1625.
19	Cheng M H, Jiao L, Shi X C, et al. An intelligent prediction model of the tool wear based on machine learning in turning high strength steel[J]. Proceedings of the Institution of Mechanical Engineers Part B Journal of Engineering Manufacture, 2020, 234(13): 1580-1597.
20	EN 10084-2008. Case hardening steels⁃technical delivery conditions [S].
21	. Tool-life testing with single-point turning tools [S].

相关文章 15

[1]	刘嫣然,孟庆瑜,郭洪艳,李嘉霖. 图注意力模式下融合高精地图的周车轨迹预测[J]. 吉林大学学报(工学版), 2023, 53(3): 792-801.
[2]	何科,丁海涛,赖宣淇,许男,郭孔辉. 基于Transformer的轮式里程计误差预测模型[J]. 吉林大学学报(工学版), 2023, 53(3): 653-662.
[3]	高金武,贾志桓,王向阳,邢浩. 基于PSO-LSTM的质子交换膜燃料电池退化趋势预测[J]. 吉林大学学报(工学版), 2022, 52(9): 2192-2202.
[4]	李晓英,杨名,全睿,谭保华. 基于深度学习的不均衡文本分类方法[J]. 吉林大学学报(工学版), 2022, 52(8): 1889-1895.
[5]	申铉京,张雪峰,王玉,金玉波. 像素级卷积神经网络多聚焦图像融合算法[J]. 吉林大学学报(工学版), 2022, 52(8): 1857-1864.
[6]	胡丹,孟新. 基于时变网格的对地观测卫星搜索海上船舶方法[J]. 吉林大学学报(工学版), 2022, 52(8): 1896-1903.
[7]	高明华,杨璨. 基于改进卷积神经网络的交通目标检测方法[J]. 吉林大学学报(工学版), 2022, 52(6): 1353-1361.
[8]	欧阳继红,郭泽琪,刘思光. 糖尿病视网膜病变分期双分支混合注意力决策网络[J]. 吉林大学学报(工学版), 2022, 52(3): 648-656.
[9]	宋林,王立平,吴军,关立文,刘知贵. 基于信息物理融合和数字孪生的可靠性分析[J]. 吉林大学学报(工学版), 2022, 52(2): 439-449.
[10]	曹洁,马佳林,黄黛麟,余萍. 一种基于多通道马尔可夫变迁场的故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(2): 491-496.
[11]	蒋仁言,熊彬彬. 基于等效加工时间模型的机床退化过程建模[J]. 吉林大学学报(工学版), 2022, 52(2): 483-490.
[12]	燕杨,尤紫如,姚远,黄文博. 基于注意力U-Net的视网膜血管分类识别[J]. 吉林大学学报(工学版), 2022, 52(12): 2933-2940.
[13]	杨军,高志明. 基于无监督变形网络的三维模型稠密对应关系计算[J]. 吉林大学学报(工学版), 2022, 52(12): 2971-2983.
[14]	侯春萍,杨庆元,黄美艳,王致芃. 基于语义耦合和身份一致性的跨模态行人重识别方法[J]. 吉林大学学报(工学版), 2022, 52(12): 2954-2963.
[15]	刘元宁,朱琳,朱晓冬,刘震,吴浩萌. 基于生成模型提升训练的深度学习虹膜识别方法[J]. 吉林大学学报(工学版), 2022, 52(12): 2924-2932.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

主轴转速/（r·min^-1）	进给量/（mm·min^-1）	背吃刀量/mm
850	100	0.4
850	150	0.5
850	200	0.6
950	100	0.5
950	150	0.6
950	200	0.4
1150	100	0.6
1150	150	0.4
1150	200	0.5

状态	后刀面磨损/mm	定制后刀面磨损值/mm
初期磨损	0.00~0.10	0， 0.05
正常磨损	0.10~0.25	0.10， 0.15， 0.20
剧烈磨损	0.25~0.30	0.25
磨钝失效	≥0.30	0.30

识别方法	MSE	RMSE	MAPE
LSTM	2.8122	1.6769	0.0111
PSO?LSTM	0.8993	0.9483	0.0059

模型	初期	正常	剧烈	磨钝	总体
LSTM	90.0	94.0	93.3	87.5	92.1
PSO?LSTM	96.7	96.0	96.7	93.8	96.0
提升	6.7	2.0	3.4	6.3	3.9