基于PSO⁃BiLSTM神经网络的机身筒段应力预测

doi:10.7527/S1000-6893.2022.26991

Abstract

Abstract:

Before aircraft fuselage components are assembled and docked， shape control tools are usually used to adjust the shape of the fuselage tube section to meet the docking standard. In order to avoid damage to the fuselage due to excessive local stress during the shape control process， it is necessary to monitor changes in stress. This research proposes a stress prediction method， which combines Particle Swarm Optimization and Bidirectional Long Short-Term Memory （PSO-BiLSTM） neural network. In this paper， the model input data is converted into sliding window sequential data with extremely high data correlation， and the stress data set is trained and tested. The experimental results show that the PSO-BiLSTM neural network has obvious advantages in processing sequential data. This is because the PSO-BiLSTM network has long memory cells and high model capacity. Stress prediction loss converges within 0.3% Root Mean Square Error （RMSE） error range. Compared with the competitive model RNN network， the standard Long Short-Term Memory （LSTM） network and the Bidirectional Long Short-Term Memory （BiLSTM） neural network， the PSO-BiLSTM model not only predicts more accurately， but also significantly improves training efficiency.

Key words: assembly, fuselage tube section, shape control, real-time monitoring, stress prediction, particle swarm optimization

CLC Number:

V214.4⁺1

Chao YANG, Kaifu ZHANG. Stress prediction of fuselage tube section based on PSO⁃BiLSTM neural network[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 426991-426991.

Figures/Tables 14

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Table 1

Table 2

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Table 3

Comparison of evaluation indicators of four kinds of model at monitoring point 2

模型	RMSE	MAE	$R 2$
RNN	0.011 8	0.027 8	0.763 5
LSTM	0.008 8	0.020 3	0.862 5
BiLSTM	0.004 5	0.012 5	0.956 6
PSO-BiLSTM	0.003 2	0.010 2	0.978 4

Table 3

Table 4

Comparison of evaluation indicators of four kinds of model at monitoring point 4

模型	RMSE	MAE	$R 2$
RNN	0.010 6	0.024 7	0.8542
LSTM	0.008 3	0.016 4	0.938 9
BiLSTM	0.004 5	0.009 6	0.975 1
PSO-BiLSTM	0.003 6	0.007 4	0.986 9

Table 4

References 22

1	荆涛，田锡天. 基于蒙特卡洛-自适应差分进化算法的飞机容差分配多目标优化方法［J］. 航空学报， 2022， 43（3）： 425278.
	JING T， TIAN X T. Multi-objective optimization method for aircraft tolerance allocation based on Monte Carlo-adaptive differential evolution algorithm［J］. Acta Aeronautica et Astronautica Sinica， 2022， 43（3）： 425278 （in Chinese）.
2	唐成顺，孙丹，唐威，等．基于LSTM循环神经网络的汽轮机转子表面应力预测模型［J］．中国电机工程学报，2021， 41（2）： 451-460.
	TANG C S， SUN D， TANG W， et al. A turbine rotor surface stress prediction model based on LSTM recurrent neural network［J］. Proceedings of the CSEE， 2021， 41（2）： 451-460 （in Chinese）.
3	袁泓磊，李尚平，李向辉，等．基于深度学习模型的甘蔗转运车节点应力预测［J］．装备制造技术，2020（4）： 1-4.
	YUAN H L， LI S P， LI X H， et al. Stress prediction of key nodes of sugarcane transport vehicle based on deep learning［J］．Equipment Manufacturing Technology， 2020（4）： 1-4 （in Chinese）.
4	GULGEC N， TAKAC M， PAKZAD S. Innovative sensing by using deep learning framework［C］∥Dynamics of Civil Structures， Volume 2 Proceedings of the 35th IMAC， A Conference and Exposition on Structural Dynamics 2017. 2017： 293-300.
5	HUANG R J， WEI C J， WANG B H， et al. Well performance prediction based on Long Short-Term Memory （LSTM） neural network［J］. Journal of Petroleum Science and Engineering， 2022， 208（Part D）： 109686.
6	MA Y H， DU X， SUN X M. Adaptive modification of the turbofan engine nonlinear model based on LSTM neural networks and hybrid optimization method［J］. Chinese Journal of Aeronautics， 2021， 1-23.Avilable from：.
7	HAJIAGHAYI M， VAHEDI E. Code failure prediction and pattern extraction using LSTM networks［C］∥ 2019 IEEE Fifth International Conference on Big Data Computing Service and Applications （Big Data Service）. Newark： Institute for Electrical and Electronic Engineers， 2019： 55-62.
8	ZHANG Y. Aeroengine fault prediction based on bidirectional LSTM neural network［C］∥ 2020 International Conference on Big Data， Artificial Intelligence and Internet of Things Engineering （ICBAIE）. Piscataway， NJ： IEEE Press， 2020： 317-320.
9	PAN D W， SONG Z， NIE L Q， et al. Satellite telemetry data anomaly detection using Bi-LSTM prediction based model［C］∥ 2020 IEEE International Instrumentation and Measurement Technology Conference （I2MTC）. Piscataway， NJ： IEEE Press， 2020： 1-6.
10	车畅畅，王华伟，倪晓梅，等. 基于1D-CNN和Bi-LSTM的航空发动机剩余寿命预测［J］. 机械工程学报， 2021， 57（14）： 304-312.
	CHE C C， WANG H W， NI X M， et al. Residual life prediction of aeroengine based on 1D-CNN and Bi-LSTM［J］. Journal of Mechanical Engineering， 2021， 57（14）： 304-312 （in Chinese）.
11	BIAN C， HE H L， YANG S K. Stacked bidirectional long short-term memory networks for state-of-charge estimation of lithium-ion batteries［J］. Energy， 2020， 191： 116538.
12	LIU J W， LI Q， CHEN W R， et al. Remaining useful life prediction of PEMFC based on long short-term memory recurrent neural networks［J］. International Journal of Hydrogen Energy， 2019， 44（11）： 5470-5480.
13	QUE Z Q， LIU Y Y， GUO C， et al. Real-time anomaly detection for flight testing using AutoEncoder and LSTM［C］∥ 2019 International Conference on Field-Programmable Technology （ICFPT）. Piscataway， NJ： IEEE Press， 2019： 379-382.
14	LI D Z， YAN M J， MIAO Z W， et al. LSTM neural network based tensile stress prediction of rubber streching［C］∥ 2020 IEEE International Conference on Networking， Sensing and Control （ICNSC）. Piscataway， NJ： IEEE Press， 2020： 1-5.
15	SUN Q， TANG Z， GAO J P， et al. Short-term ship motion attitude prediction based on LSTM and GPR［J］. Applied Ocean Research， 2022， 118： 102927.
16	YAO W Y， HUANG P， JIA Z X. Multidimensional LSTM networks to predict wind speed ［C］∥ 2018 37th Chinese Control Conference （CCC）. Piscataway， NJ： IEEE Press， 2018： 7493-7497.
17	DU Y， CUI N X， LI H X， et al. The vehicle’s velocity prediction methods based on RNN and LSTM neural network［C］∥ 2020 Chinese Control and Decision Conference （CCDC）. Piscataway， NJ： IEEE Press， 2020： 99-102.
18	MA J Y， LUO D C， LIAO X P， et al. Tool wear mechanism and prediction in milling TC18 titanium alloy using deep learning［J］. Measurement， 2020， 173（1）： 108554.
19	CUI Z Y， KE R M， PU Z Y， et al. Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values ［J］. Transportation Research Part C： Emerging Technologies， 2020， 118： 102674.
20	SIAMI-NAMINI S， TAVAKOLI N， NAMIN A S. The performance of LSTM and BiLSTM in forecasting time series［C］∥ 2019 IEEE International Conference on Big Data （Big Data）. Piscataway， NJ： IEEE Press， 2019： 3285-3292.
21	KENNEDY J， EBERHART R. Particle swarm optimization ［C］∥ Proceedings of IEEE International Conference on Neural Networks. Piscataway， NJ： IEEE Press， 1995： 1942-1948.
22	GREFF K， SRIVASTAVA R K， KOUTNIK J， et al. LSTM： a search space odyssey［J］. IEEE Transactions on Neural Networks and Learning Systems， 2017， 28（10）： 2222 -2232.

超参数	RNN	LSTM	BiLSTM	PSO-BiLSTM
隐藏层神经元数量	80	80	70	100
最大时代	200	150	120	110
学习率	0.001 8	0.001 5	0.001 6	0.002 4
辍学率	0.5	0.5	0.3	0.2
学习率下降因子	0.3	0.5	0.3	0.3

[1]	Zhenyu WANG, Jizhen WANG, Jingyi YANG, Pengyuan HE, Chenghao PAN, Lingbo ZHOU, Cheng HE, Huan HE. Dynamics model updating of structures at high temperature based on novel particle swarm optimization algorithm [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 226559-226559.
[2]	Heye XIAO, Jianfeng YANG, Junqiang BAI, Xudong ZHANG, Lirong WU. Modular UAVs configuration method responded to task requirements [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 327100-327100.
[3]	Xiaoxu HE, Pei LEI, Deng PAN, Yang YANG, Xiankun LI, Zhenbo DENG. Posture and position computing method for aircraft component based on PSO and WSVD [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(7): 427162-427162.
[4]	Qilei GUO, Weimin SANG, Junjie NIU, Ye YUAN. UAV flight strategy considering icing risk under complex meteorological conditions [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2023, 44(1): 627518-627518.
[5]	ZHANG Sheng, YANG Yu, WANG Zhigang, SHI Xintong. Design and validation of eccentric beam for variable camber trailing edge [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(6): 525892-525892.
[6]	WANG Hao, CHEN Genliang. Research progress and perspective of robotic equipment applied in aviation assembly [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(5): 626128-626128.
[7]	JING Tao, TIAN Xitian. Multi-objective optimization method for aircraft tolerance allocation based on Monte Carlo-adaptive differential evolution algorithm [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(3): 425278-425278.
[8]	ZHANG Ruipeng, FENG Yanxiang, YANG Yikang. Hybrid particle swarm algorithm for multi-UAV cooperative task allocation [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(12): 326011-326011.
[9]	YAN Chuliang. Development and prospect of aircraft structural life reliability assessment technology in China [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(10): 527869-527869.
[10]	ZHAO Gang, LI Jinyue, XU Maocheng, ZHANG Pengfei. Research status and prospect of key aero-engine assembly technology [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022, 43(10): 527484-527484.
[11]	YE Jiesong, GONG Baichun, LI Shuang, DU Yanli, HAO Mingrui. Multi-agent formation cooperative control using relative bearing information and single-spacing measurement [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(7): 324610-324610.
[12]	GUO Yongheng, JIANG Xiong, XIAO Zhongyun, WANG Ziwei, LU Fengshun. An automatic parallel and implicit assembly method for overset grid [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(6): 124369-124369.
[13]	JIA Qingxuan, DUAN Jiaqi, CHEN Gang. Uncalibrated visual servo of space robots performing on-orbit assembly alignment task [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(6): 424063-424063.
[14]	SUI Dong, XING Yaping, TU Shichen. Repair optimization strategy for air route networks under severe weather conditions [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(2): 324300-324300.
[15]	TU Jianbo, LI Zhen, GE Haotian, LIU Liang, LIU Honghui. Multi-objective optimization of rotor-stack assembly based on geometric algebra theory [J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2021, 42(10): 524197-524197.

Stress prediction of fuselage tube section based on PSO⁃BiLSTM neural network

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 14

References 22

Related Articles 15

Recommended Articles

Metrics

Comments