机理⁃数据混合驱动的叶片加工变形预测方法

doi:10.7527/S1000-6893.2023.29037

Abstract

Abstract:

As a typical thin-walled part of aviation， blade is prone to elastic deformation and machining residual stress deformation in the process of milling， which affects the machining accuracy and quality of blade. To address the deformation mechanism modeling problem of complex machining process of blade and the data-driven model prediction problem relying on limited collected samples， a mechanism-guided Sparrow optimized Extreme Learning Machine （SSA-ELM） thin-walled blade machining deformation prediction model is established. Firstly， the machining deformation mechanism model of thin-walled blade considering elastic deformation and machining residual stress deformation is established. Secondly， the Extreme Learning Machine Neural Network（ELMNN）is designed， and the Sparrow Search Algorithm （SSA） is used to optimize the parameters of the ELMNN hidden layer network. Then， the dataset of the data-driven model is constructed， and the Monte Carlo simulation is used to enhance the processed deformation data samples to establish the data-driven model. Finally， the mechanism-guided SSA-ELM is used to predict the thin-walled blade machining deformation， and the root mean square error and correlation coefficient of the machining deformation are used as the model accuracy evaluation index. The results show that the root mean square error of the mechanism guided and SSA optimized machining deformation prediction model is reduced by 77.25% and 30.5% respectively， which proves that the model has good prediction performance.

Key words: thin-walled blade, elastic deformation, machining residual stress deformation, machining deformation prediction, mechanism model, data-driven model

CLC Number:

V261.2

Lizhuo DONG, Siqi ZHANG, Zhao ZHANG, Baohai WU. Prediction method of blade machining deformation driven by mechanism⁃data hybrid[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(13): 629037-629037.

Figures/Tables 20

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Table 1

Fig.9

Table 2

Input data of influence factors for data driven model

序号	柔度K $x - 1$ /（mm∙N^-1）	柔度K $y - 1$ /（mm∙N^-1）	坐标y/mm	坐标z/mm	法矢 $i$	法矢 $j$
1	3.69×10^-5	7.47×10^-6	-105.023 0	1 688.156 1	-0.844 2	-0.526 5
2	3.66×10^-5	7.27×10^-6	-98.255 0	1 688.095 1	-0.869 2	-0.486 7
3	3.64×10^-5	7.12×10^-6	-91.309 4	1 688.047 7	-0.892 5	-0.445 1
4	3.62×10^-5	7.01×10^-6	-84.191 8	1 687.990 6	-0.915 9	-0.397 1
⋮	⋮	⋮	⋮	⋮	⋮	⋮
265	5.69×10^-6	4.30×10^-6	105.919 5	319.540 9	-0.818 4	-0.567 5
266	6.00×10^-6	4.62×10^-6	116.593 3	319.540 1	-0.818 3	-0.567 7

Table 2

Table 3

Table 4

Fig.10

Fig.11

Fig.12

Table 5

Fig.13

Fig.14

Fig.15

References 31

1	万敏. 薄壁件周铣加工过程中表面静态误差预测关键技术研究［D］. 西安：西北工业大学， 2005： 21-46.
	WAN M. Numerical prediction of static form errors in the peripheral milling of thin-walled workpiece［D］.Xi’an： Northwestern Polytechnical University， 2005： 21-46 （in Chinese）.
2	LI P F， LIU Y， GONG Y D， et al. New deformation prediction of micro thin-walled structures by iterative FEM［J］. The International Journal of Advanced Manufacturing Technology， 2018， 95（5）： 2027-2040.
3	YAO C F， ZHANG J Y， CUI M C， et al. Machining deformation prediction of large fan blades based on loading uneven residual stress［J］. The International Journal of Advanced Manufacturing Technology， 2020， 107（9）： 4345-4356.
4	王成龙，贺永海，孙杰，等. 大型薄壁筒件数字减薄中的变形自适应补偿方法［J］. 航天制造技术， 2021（3）： 23-27.
	WANG C L， HE Y H， SUN J， et al. A deformation adaptive compensation method for digital thinning of large thin-walled cylinder parts［J］. Aerospace Manufacturing Technology， 2021（3）： 23-27 （in Chinese）.
5	LI W T， WANG L P， YU G. Chatter prediction in flank milling of thin-walled parts considering force-induced deformation［J］. Mechanical Systems and Signal Processing， 2022， 165： 108314.
6	GE G Y， XIAO Y K， FENG X B， et al. An efficient prediction method for the dynamic deformation of thin-walled parts in flank milling［J］. Computer-Aided Design， 2022， 152： 103401.
7	JAYANTI S， REN D， ERICKSON E， et al. Predictive modeling for tool deflection and part distortion of large machined components［J］. Procedia CIRP， 2013， 12： 37-42.
8	廖凯，张萧笛，车兴飞，等. 铝合金薄壁件加工变形的力学模型构建与分析［J］. 哈尔滨工业大学学报， 2018， 50（5）： 166-172.
	LIAO K， ZHANG X D， CHE X F， et al. Construction and analysis of mechanic model of deformation for Al alloy thin-walled component［J］. Journal of Harbin Institute of Technology， 2018， 50（5）： 166-172 （in Chinese）.
9	黄晓明，孙杰，李剑峰. 基于刚度与应力演变机制的航空整体结构件加工变形预测理论建模［J］. 机械工程学报， 2017， 53（9）： 201-208.
	HUANG X M， SUN J， LI J F. Mathematical modeling of aeronautical monolithic component machining distortion based on stiffness and residual stress evolvement［J］. Journal of Mechanical Engineering， 2017， 53（9）： 201-208 （in Chinese）.
10	LI X Y， LI L， YANG Y F， et al. Variance-based sensitivity analysis for the influence of residual stress on machining deformation［J］. Journal of Manufacturing Processes， 2021， 68： 1072-1085.
11	ZHAO Z W， LI Y G， LIU C Q， et al. A subsequent-machining-deformation prediction method based on the latent field estimation using deformation force［J］. Journal of Manufacturing Systems， 2022， 63： 224-237.
12	LIU F， ZHANG N S， WANG A M， et al. Deformation prediction of thin-walled parts based on BP neural network［C］∥ 2021 2nd International Symposium on Computer Engineering and Intelligent Communications （ISCEIC）. Piscataway： IEEE Press， 2021： 169-172.
13	郭建烨，郑若池. 基于改进烟花算法的薄壁件铣削加工参数优化［J］. 制造技术与机床， 2021（6）： 70-74， 80.
	GUO J Y， ZHENG R C. Optimization of milling parameters of thin-walled parts based on improved firework algorithm［J］. Manufacturing Technology & Machine Tool， 2021（6）： 70-74， 80 （in Chinese）.
14	王峰，徐雷，贺云翔，等. 基于MPSO-BP对5A06铝合金薄壁件变形预测［J］. 组合机床与自动化加工技术， 2019（5）： 84-89.
	WANG F， XU L， HE Y X， et al. The deformation prediction of 5A06 aluminum alloy thin-wall parts based on MPSO-BP［J］. Modular Machine Tool & Automatic Manufacturing Technique， 2019（5）： 84-89 （in Chinese）.
15	张俊涛. 基于数字孪生的薄壁件铣削加工变形控制研究［D］. 哈尔滨：哈尔滨理工大学， 2022： 40-50.
	ZHANG J T. Research on milling deformation control of thin-walled parts based on digital twin［D］. Harbin： Harbin University of Science and Technology， 2022： 40-50 （in Chinese）.
16	GUO J， WANG B， HE Z X， et al. A novel method for workpiece deformation prediction by amending initial residual stress based on SVR-GA［J］. Advances in Manufacturing， 2021， 9（4）： 483-495.
17	翟小飞，马仕洪，魏伟，等. 基于神经网络和贝叶斯优化的核电站机组功率参数自动寻优方法［J］. 自动化应用， 2021（3）： 51-53， 57.
	ZHAI X F， MA S H， WEI W， et al. Automatic optimization method for power parameters of nuclear power plant units based on neural network and Bayesian optimization［J］. Automation Application， 2021（3）： 51-53， 57 （in Chinese）.
18	HAO X Z， LI Y G， LI M Q， et al. A part deformation control method via active pre-deformation based on online monitoring data［J］. The International Journal of Advanced Manufacturing Technology， 2019， 104（5）： 2681-2692.
19	ZHAO Z W， LI Y G， LIU C Q， et al. On-line part deformation prediction based on deep learning［J］. Journal of Intelligent Manufacturing， 2020， 31（3）： 561-574.
20	ZHAO Z W， LI Y G， LIU C Q， et al. Predicting part deformation based on deformation force data using Physics-informed Latent Variable Model［J］. Robotics and Computer-Integrated Manufacturing， 2021， 72： 102204.
21	CAO L， ZHANG X M， HUANG T， et al. Online monitoring machining errors of thin-walled workpiece： A knowledge embedded sparse Bayesian regression approach［J］. IEEE/ASME Transactions on Mechatronics， 2019， 24（3）： 1259-1270.
22	YAN Q H， LUO M， TANG K. Multi-axis variable depth-of-cut machining of thin-walled workpieces based on the workpiece deflection constraint［J］. Computer-Aided Design， 2018， 100： 14-29.
23	丛靖梅，莫蓉，吴宝海，等. 面向性能的压气机叶片铣削加工误差分析及统计［J］. 航空制造技术， 2017， 60（15）： 38-44.
	CONG J M， MO R， WU B H， et al. Performance oriented machining error analysis and statistic of compressor blade［J］. Aeronautical Manufacturing Technology， 2017， 60（15）： 38-44 （in Chinese）.
24	邓宇锋. 透平叶片变切削力加工参数研究［J］. 组合机床与自动化加工技术， 2015（2）： 135-137， 141.
	DENG Y F. Research on parameters of the alterable cutting forces in machining turbine blade［J］. Modular Machine Tool & Automatic Manufacturing Technique， 2015（2）： 135-137， 141 （in Chinese）.
25	孙祺. 基于有限元方法的旋转叶片-机匣碰摩动力学研究［D］. 沈阳：东北大学， 2018： 9-23.
	SUN Q. Research on rotating blade-casing rubbing induced vibration response based on finite element method［D］. Shenyang： Northeastern University， 2018： 9-23 （in Chinese）.
26	潘和林. 钛合金薄壁件铣削变形的预测与控制［D］. 济南：山东大学， 2016： 51-66.
	PAN H L. Deflection prediction and control in milling of thin-wall titanium alloy components［D］. Jinan： Shandong University， 2016： 51-66 （in Chinese）.
27	XUE J K， SHEN B. A novel swarm intelligence optimization approach： Sparrow search algorithm［J］. Systems Science & Control Engineering， 2020， 8（1）： 22-34.
28	李雅丽，王淑琴，陈倩茹，等. 若干新型群智能优化算法的对比研究［J］. 计算机工程与应用， 2020， 56（22）： 1-12.
	LI Y L， WANG S Q， CHEN Q R， et al. Comparative study of several new swarm intelligence optimization algorithms［J］. Computer Engineering and Applications， 2020， 56（22）： 1-12 （in Chinese）.
29	赵渊，张夏菲，周家启. 电网可靠性评估的非参数多变量核密度估计负荷模型研究［J］. 中国电机工程学报， 2009， 29（31）： 27-33.
	ZHAO Y， ZHANG X F， ZHOU J Q. Load modeling utilizing nonparametric and multivariate kernel density estimation in bulk power system reliability evaluation［J］. Proceedings of the CSEE， 2009， 29（31）： 27-33 （in Chinese）.
30	袁修开，吕震宙，池巧君. 基于核密度估计的自适应重要抽样可靠性灵敏度分析［J］. 西北工业大学学报， 2008， 26（3）： 297-302.
	YUAN X K， LÜ Z Z， CHI Q J. Achieving efficient estimation of reliability sensitivity of a multi-mode system without requiring knowledge of design point［J］. Journal of Northwestern Polytechnical University， 2008， 26（3）： 297-302 （in Chinese）.
31	岳彩旭，张俊涛，刘献礼，等. 薄壁件铣削过程加工变形研究进展［J］. 航空学报， 2022， 43（4）： 525164.
	YUE C X， ZHANG J T， LIU X L， et al. Research progress on machining deformation of thin-walled parts in milling process［J］. Acta Aeronautica et Astronautica Sinica， 2022， 43（4）： 525164 （in Chinese）.

试验序号	铣削力	试验值/N	计算值/N	误差/%
1	F_x	1 291	1 286.29	0.36
1	F_y	689	697.33	1.21
2	F_x	1 168	1 286.29	10.13
2	F_y	700	697.33	0.38
3	F_x	1 400	1 286.29	8.12
3	F_y	715	697.33	2.47

序号	g（x，y）/mm	h（y）/mm²	h（z）/mm²
1	-0.042 8	11 029.788 5	2 849 871.018 0
2	-0.043 4	9 654.045 0	2 849 665.066 6
3	-0.044 0	8 337.406 5	2 849 505.037 5
4	-0.044 6	7 088.259 2	2 849 312.265 7
⋮	⋮	⋮	⋮
265	-0.007 7	11 218.940 5	102 106.386 8
266	-0.008 1	13 593.997 6	102 105.875 5

序号	加工变形量/mm
1	0.755 6
2	0.639 3
3	0.532 5
4	0.413 9
⋮	⋮
265	0.533 6
266	0.555 1

评价指标	纯数据驱动模型	机理引导的数据驱动模型
评价指标	纯数据驱动模型	优化前	优化后
RMSE	0.133 2	0.043 6	0.030 3
R²	0.762 1	0.976 3	0.988 6
训练时间/s	28	27	61

[1]	Gang LI, Yu ZHANG, Si LI, Kunpeng ZHU. Modeling of high speed and high precision milling forces based on kinematics: Comprehensive modeling and experimental [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(8): 427261-427261.
[2]	Zhiyang ZHENG, Yang ZHANG, Zhao ZHANG, Baohai WU, Ying ZHANG. Layout optimization of auxiliary support for thin-walled blade based on GA-SVR [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(4): 426805-426805.
[3]	Gongye YU, Weidong CAI, Minghui HU, Wencai LIU, Bo MA. Intelligent migration diagnosis of mechanical faults driven by hybrid fault mechanism and domain adaptation [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(2): 426800-426800.
[4]	CAI Jin, YAN Xue, LI Wei, MENG Qingxun. Numerical analysis of ultrasonic shot peening based on DEM-FEM coupling [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(4): 525925-525925.
[5]	GUO Qiuting, SUN Yan, GUO Zheng, LIU Guangyuan. Separation method for Reynolds number/static aeroelastic coupling effect in wind tunnel test [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(11): 526312-526312.
[6]	ZHANG Yizhi, CHENG Cheng, FAN Yitong, LI Gaohua, LI Weipeng. Data-driven correction of turbulence model with physics knowledge constrains in channel flow [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2020, 41(3): 123282-123282.
[7]	WANG Yuntao, SUN Yan, MENG Dehong, ZHANG Shujun, YANG Xiaochuan. Numerical simulation of aerodynamic characteristics of CRM-WB configuration with support system and wing deformation [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2017, 38(10): 121202-121202.
[8]	Zhu Xi-xiong;Zhu Guo-rui;Huang Xu-sheng;Cheng Jiang-ying. THE NON-LINEAR THERMOVISCOELASTIC-PLASTIC CONSTITUTIVE RELATIONS FOR AN AERONAUTICAL PMMA [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 1992, 13(11): 594-601.

Prediction method of blade machining deformation driven by mechanism⁃data hybrid

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 20

References 31

Related Articles 8

Recommended Articles

Metrics

Comments