Energy-absorption forecast of thin-walled structure by GA-BP hybrid algorithm

Abstract

In order to analyze the influence rule of experimental parameters on the energy-absorption characteristics and effectively forecast energy-absorption characteristic of thin-walled structure, the forecast model of GA-BP hybrid algorithm was presented by unifing respective applicability of back-propagation artificial neural network (BP-ANN) and genetic algorithm (GA). The detailed process was as follows. Firstly, the GA trained the best weights and thresholds as the initial values of BP-ANN to initialize the neural network. Then, the BP-ANN after initialization was trained until the errors converged to the required precision. Finally, the network model, which met the requirements after being examined by the test samples, was applied to energy-absorption forecast of thin-walled cylindrical structure impacting. After example analysis, the GA-BP network model was trained until getting the desired network error only by 46 steps, while the single BP-ANN model achieved the same network error by 992 steps, which obviously shows that the GA-BP hybrid algorithm has faster convergence rate. The average relative forecast error (A_RE) of the S_EA predictive results obtained by GA-BP hybrid algorithm is 1.543%, while the A_RE of the S_EA predictive results obtained by BP-ANN is 2.950%, which clearly indicates that the forecast precision of the GA-BP hybrid algorithm is higher than that of the BP-ANN.

Keywords

thin-walled structure / GA-BP hybrid algorithm / impact / energy-absorption characteristic / forecast

References

[1] MarkiewiceE, MarchandM, DucrocpP. Evaluation of different simplified crash model: Application to the under frame of a railway driver’s cab [J]. Int J of Vehicle Design, 2001, 26(26): 187-205

[2] ScholesA, LewisJ H. Development of crashworthiness for railway vehicle structures [J]. Journal of Rail and Rapid Transit, 1993, 207: 1-16

[3] AbramowiczW. Thin-walled structures as impact energy absorbers [J]. Thin-Walled Structures, 2003, 41(5): 91-107

[4] ChenW G, WierzbickiT. Relative merits of single-cell multi-cell, and foam-filled thin-walled structures in energy absorption [J]. Thin-Walled Structures, 2001, 39(4): 287-306

[5] WhiteM D, JonesN. Experimental study into the energy absorbing characteristics of top-hat and double-hat sections subjected to dynamic axial crushing [J]. Pro Instn Mech Engrs: Part D, 1999, 213: 259-278

[6] JonesN. Some recent developments in the dynamic inelastic behavior of structures [J]. Ships Offshore Struct, 2006, 1(1): 37-44

[7] ZareiH R, KrogerM. Multiobjective crashworthiness optimization of circular aluminum tubes [J]. Thin-Walled Structures, 2006, 44(3): 301-308

[8] HouS-j, LiQ, LongS-yao. Design optimization of regular hexagonal thin-walled columns with crashworthiness criterion [J]. Finite Element in Analysis and Design, 2007, 43(6/7): 555-565

[9] WisniewskiK, KolakowskiP. The effect of selected parameters on ship collision results by dynamic FE simulations [J]. Finite Elements in Analysis and Design, 2003, 39(2): 985-1006

[10] AvrahamS, RonenV. Estimating the cost of steel pipe bending, a comparison between neural networks and regression analysis [J]. Int J Production Economics, 1999, 62(3): 201-207

[11] JonesD M, WattonJ, BrownK J. Comparison of hot rolled steel mechanical property prediction models using linear multiple regression, non-linear multiple regression and non-linear artificial neural networks [J]. Ironmaking and Steelmaking, 2005, 32(5): 435-442

[12] JurgenB. Neural networks for cost estimation: Simulations and pilot application [J]. International Journal of Production Research, 2000, 38(6): 1231-1254

[13] PaoH T. A comparison of neural network and multiple regression analysis in modeling capital structure [J]. Expert Systems with Applications, 2008, 35(3): 720-727

[14] LiuG-s, YuJ-guo. Gray correlation analysis and prediction models of living refuse generation in Shanghai city [J]. Waste Management, 2007, 27(3): 345-351

[15] ChenC-S, LiuC-H, SuH-Chen. A nonlinear time series analysis using two-stage genetic algorithms for streamflow forecasting [J]. Hydrological Processes, 2008, 22(18): 3697-3711

[16] Navarro-Esbr’iJ, DiamadopoulosE, GinestarD. Time series analysis and forecasting techniques for municipal solid waste management [J]. Resources, Conservation and Recycling, 2002, 35(3): 201-214

[17] SivaS B, SekharM, RiotteJ, BraunJ J. Non-linear regression model for spatial variation in precipitation chemistry for South India [J]. Atmospheric Environment, 2009, 43(5): 1147-1152

[18] YangS-x, CaoY, LiuD, HuangC-feng. RS-SVM forecasting model and power supply-demand forecast [J]. Journal of Central South University of Technology, 2011, 18(6): 2074-2079

[19] ZhongZ-huaFinite element procedures for contact-impact problems [M], 1993New YorkOxford university press122-146

[20] WangX-c, ShaoMingBasic theory and numerical method of finite element method [M], 2002BeijingTsinghua University Press468-499

[21] KumarS, NareshR. Efficient real coded genetic algorithm to solve the non-convex hydrothermal scheduling problem [J]. International Journal of Electrical Power and Energy Systems, 2007, 29(10): 738-747

[22] LS-DYNA Keyword User’s Manual, Version 960 [M]. California: Livermore Software Technology Corporation, 2001: 563–612.