Neural network in predicting the mechanical strength of Al6061/SiC composites used in aerospace industries
محورهای موضوعی : Computer Engineering
1 - دانشکده فنی و مهندسی، واحد دامغان، دانشگاه آزاد اسلامی، دامغان، ایران
کلید واژه: Fiber Type, Heat Treatment, Yield Strength, Ultimate Tensile Strength, Elastic Modulus ,
چکیده مقاله :
In this research, the effect of three parameters, the percentage and type of SiC reinforcing material and the type of thermal cycle of composite manufacturing on the elastic modulus, yield strength and ultimate strength of Al/SiC composite have been investigated. For this purpose, the information related to composite with Al6061alloy reinforced with SiC particles was extracted. After normalizing the input and output information, the perceptron recurrent network was designed and it was tried to extract the optimal parameters by changing different parameters of the network such as the frequency of network training, changing the learning coefficient of the network and weight and bias coefficients and comparing the sum of squared errors in different conditions. Finally, by defining the adaptive learning coefficient in the network, it was tried to improve the speed and accuracy of the network. The results showed that by repeating the network training 10000000 times, the sum of squares of error was reduced to the range of 7-10. Also, the lowest error sum of squares is related to the learning coefficient α=0.002. The results related to the definition of the adaptive learning coefficient showed that if the network is trained 100,000 times, the training speed decreases to 60,000 times with the same error for the adaptive learning coefficient and the constant learning coefficient. In other words, if the adaptive learning coefficient is used, the training speed and error reduction will be higher than using the constant learning coefficient.
1. Jung, H.K., et al., Analysis of anisotropy in elastic constants of SiCp/2124 Al metal matrix composites. 1999. 41(12): p. 1261-1267.
2. Goujon, C. and P. Goeuriot, Solid state sintering and high temperature compression properties of Al-alloy5000/AlN nanocomposites. Materials Science and Engineering: A, 2001. 315(1-2): p. 180-188.
3. Ahmadifard, S., et al., Fabrication of A5083/SiC surface composite by friction stir processing and its characterization. In Persian, Journal of Science and Technology of Composites, 2015. 2(4): p. 36-31.
4. Ahmadian, H., et al., Mechanical properties of Al-Mg/MWCNT nanocomposite powder produced under different parameters of ball milling process. Diamond and Related Materials, 2022. 121: p. 108755.
5. Dehestani, M., E. Adolfsson, and L.A. Stanciu, Mechanical properties and corrosion behavior of powder metallurgy iron-hydroxyapatite composites for biodegradable implant applications. Materials & Design, 2016. 109: p. 556-569.
6. Liu, Z.Y., et al., Fabrication of CNT/Al composites with low damage to CNTs by a novel solution-assisted wet mixing combined with powder metallurgy processing. Materials & Design, 2016. 97: p. 424-430.
7. Wang, X.J., et al., Processing, microstructure and mechanical properties of micro-SiC particles reinforced magnesium matrix composites fabricated by stir casting assisted by ultrasonic treatment processing. Materials & Design, 2014. 57: p. 638-645.
8. Sarvari, M. and M. Divandari, Melt-Solid-Gas behavior in the interface of Al/Mg composite produced by centrifugal casting. In Persian, Journal of Science and Technology of Composites, 2015. 2(3): p. 82-75.
9. Borzouyi Kutenayi, S., S. Kiahosseini, and M. Talebpour, The effect of caspian sea water on corrosion resistance and compressive strength of reinforced concrete containing different SiO2 pozzolan. International Journal of Engineering, 2017. 30(10): p. 1464-1470.
10. Rahnama, H. and M.M. Shokrieh, Exact micromechanical stress analysis of long fiber composites under uniform tensile loading. In Persian, Journal of Science and Technology of Composites, 2015. 2(3): p. 10-1.
11. Walker, E.E., The properties of powders. Part VI. The compressibility of powders. Transactions of the Faraday Society, 1923. 19: p. 73-82
12. JIANG, L., et al., Thermal analysis for brake disks of SiC/6061 Al alloy co-continuous composite for CRH3 during emergency braking considering airflow cooling. Transactions of Nonferrous Metals Society of China, 2012. 22(11): p. 2783-2791.
13. Knowles, A.J., et al., Microstructure and mechanical properties of 6061 Al alloy based composites with SiC nanoparticles. Journal of Alloys and Compounds, Volume 615, Supplement 1, 5 December 2014, Pages S401-S405, 2014. 615: p. S401-S405.
14. Alhajeri, S.N., et al., Microstructure and microhardness of an Al-6061 metal matrix composite processed by high-pressure torsion. Materials Characterization, Volume 118, August 2016, Pages 270-278, 2016. 118: p. 270-278.
15. Lange, F.F., et al., Deformation consolidation of metal powders containing steel inclusions. Acta Metallurgica et Materialia, 1991. 39(2): p. 209-219.
16. Sridhar, I. and N.A. Fleck, Yield behaviour of cold compacted composite powders. Acta Materialia, 2000. 48(13): p. 3341-3352.
17. Kim, K.T. and J.H. Cho, A densification model for mixed metal powder under cold compaction. International Journal of Mechanical Sciences, 2001. 43(12): p. 2929-2946.
18. Genel, K., S.C. Kurnaz, and M. Durman, Modeling of tribological properties of alumina fiber reinforced zinc–aluminum composites using artificial neural network. Materials Science and Engineering: A, Volume 363, Issues 1–2, 20 December 2003, Pages 203-210, 2003. 363(1-2): p. 203-210.
19. Muthukrishnan, N. and J.P. Davim, Optimization of machining parameters of Al/SiC-MMC with ANOVA and ANN analysis. Journal of Materials Processing Technology, 2009. 209(1): p. 225-232.
20. Vassilopoulos, A.P., E.F. Georgopoulos, and V. Dionysopoulos, Artificial neural networks in spectrum fatigue life prediction of composite materials. International Journal of Fatigue, 2007. 29(1): p. 20-29.
21. Li, H., et al., Numerical simulation of chemical vapor infiltration of propylene into C/C composites with reduced multi-step kinetic models. Carbon, 2005. 43(14): p. 2937-2950.
22. Kumar, C.S., et al., Failure strength prediction of glass/epoxy composite laminates from acoustic emission parameters using artificial neural network. Applied Acoustics 2017. 115: p. 32-41.
23. Danels, D.L.M. and C.A. Hoffman, Microstructure and Orientation effects on Properties of Discontinuos Silicon Carbide/Aluminum Composites. 1984, NASA: USA. p. 1-30.
24. Menhaj, M.B., Computational Intelligence Fundamentals of Neural Networks. Vol. 5. 2009, Tehran: In Persian Amirkabir University of Technology. 716.
25. Safaei Kuchaksaraei, J., Artificial intelligence in cancer diagnosis and treatment methods. Future Generation of Communication and Internet of Things, 2023. 2(3): p. 35-42.
