铸造铝硅合金力学性能自适应神经—模糊建模
|
摘要
铸造铝硅合金具有容重小、比强度高、铸造成形性和加工性能好等优点，已成为制造行业中最受重视的结构材料之一，该类合金中的ZL101A应用最为广泛。建立合金力学性能的模型预测，用于分析工艺参数、化学成分等对性能的影响，对提高其强度和韧性，稳定生产质量具有重要意义。
本文融合模糊推理和神经网络学习功能的自适应神经-模糊推理建模技术，不但可以逼近数据中存在的非线性关系，而且能够直接从中获得结构化知识描述，从而克服了传统反向传播等类型网络的“黑箱”问题。在预测和学习精度上，优于统计回归和单一的神经网络技术。
在此基础上，利用多年来积累的生产和实验数据，将自适应神经-模糊推理系统方法用于铸造铝硅合金性能建模，融合模糊推理系统与神经网络，把模糊系统处理不确定性方法与神经网络的连接结构和学习方法结合起来，从数据中自动提取模糊规则，实现了铸造铝硅合金强度、延伸率和硬度等的自适应模糊预报。
实验表明，本文所建立的自适应模糊预报模型，成功的实现了铸造铝硅合金力学性能以及其他材料物理性能的预报，可以用于分析工艺参数、化学成分等对性能的影响，为优化生产工艺、研制高性能合金提供依据。研究所形成的建模方法和技术具有一定的通用性和推广应用价值。
With high strength, low density, good castability and machineability, Al-Si series cast alloy has become one of the most important structural materials in manufacturing industry, and ZL101A is the most widely used. In order to improve the strength and ductility, it is important to develop model for predicting material properties and analysis the role of compositions and process parameters.
Besides the ability of mapping the complex non-linear relationships between compositions and properties, the proposed adaptive neuro-fuzzy inference system modeling method could extract "IF-THEN" rules directly from data, and get knowledge hidden in the data. This overcomes the "Black Box" shortcomings of BP artificial neural networks, which widely used in modeling and predicting. Further more, ANFIS has higher modeling and generation ability than regress analyses and artificial neural networks.
In this paper, based upon the accumulated product records and experimental data, the adaptive neuro-fuzzy inference system (ANFIS) method has been used to build mechanical properties models of Al-Si serises cast alloy. By the coupled use of adaptive fuzzy inference modeling and artificial neural network learning ability, a set of rules, which could help us better understanding the basic principles of alloying, have been generated directly from the experimental data. The developed fuzzy inference systems could mapping the relationships between compositions and mechanical properties accurately and could used to predict material properties
with the known compositions and/or process parameters. With the help of the developed models, one could analysis the influences of compositions and/or process on properties, and get information for the further improving of material property. By integrating in the genetic optimization procedure, the fuzzy inference systems could be used to determine the optimal compositions for the highest or desired ultimate properties.
The above ANFIS modeling method has successfully used in the property prediction of Al-Si series cast alloy. The methodology present here could also be used for other data based material property modeling and optimization practices.
本文融合模糊推理和神经网络学习功能的自适应神经-模糊推理建模技术，不但可以逼近数据中存在的非线性关系，而且能够直接从中获得结构化知识描述，从而克服了传统反向传播等类型网络的“黑箱”问题。在预测和学习精度上，优于统计回归和单一的神经网络技术。
在此基础上，利用多年来积累的生产和实验数据，将自适应神经-模糊推理系统方法用于铸造铝硅合金性能建模，融合模糊推理系统与神经网络，把模糊系统处理不确定性方法与神经网络的连接结构和学习方法结合起来，从数据中自动提取模糊规则，实现了铸造铝硅合金强度、延伸率和硬度等的自适应模糊预报。
实验表明，本文所建立的自适应模糊预报模型，成功的实现了铸造铝硅合金力学性能以及其他材料物理性能的预报，可以用于分析工艺参数、化学成分等对性能的影响，为优化生产工艺、研制高性能合金提供依据。研究所形成的建模方法和技术具有一定的通用性和推广应用价值。
With high strength, low density, good castability and machineability, Al-Si series cast alloy has become one of the most important structural materials in manufacturing industry, and ZL101A is the most widely used. In order to improve the strength and ductility, it is important to develop model for predicting material properties and analysis the role of compositions and process parameters.
Besides the ability of mapping the complex non-linear relationships between compositions and properties, the proposed adaptive neuro-fuzzy inference system modeling method could extract "IF-THEN" rules directly from data, and get knowledge hidden in the data. This overcomes the "Black Box" shortcomings of BP artificial neural networks, which widely used in modeling and predicting. Further more, ANFIS has higher modeling and generation ability than regress analyses and artificial neural networks.
In this paper, based upon the accumulated product records and experimental data, the adaptive neuro-fuzzy inference system (ANFIS) method has been used to build mechanical properties models of Al-Si serises cast alloy. By the coupled use of adaptive fuzzy inference modeling and artificial neural network learning ability, a set of rules, which could help us better understanding the basic principles of alloying, have been generated directly from the experimental data. The developed fuzzy inference systems could mapping the relationships between compositions and mechanical properties accurately and could used to predict material properties
with the known compositions and/or process parameters. With the help of the developed models, one could analysis the influences of compositions and/or process on properties, and get information for the further improving of material property. By integrating in the genetic optimization procedure, the fuzzy inference systems could be used to determine the optimal compositions for the highest or desired ultimate properties.
The above ANFIS modeling method has successfully used in the property prediction of Al-Si series cast alloy. The methodology present here could also be used for other data based material property modeling and optimization practices.
引文
1. J.R. Willianson D. L. Mellellan, Application of Aluminum Castings to Air Force System, AFS Trans., 1982, 90:557～564.
2．李成功 傅恒志 于翘．航空航天材料[M]．北京：国防工业出版社．2002
3．刘大利等．新型铝活塞[M]．北京：国防工业出版社．1999
4. J.T. Berry E. P. Coleman, Linking Solidification Conditions and Mechanical Behavior in Al Casting— — A Quarter Century of Evolving Evidence, AFS Trans. 1995,103:837～846.
5. Dobrzanski L A, Sitek W. Application of a neural network in modelling of hardenability of constructional steels [J]. J. Mater. Proc.Tech, 2000,78: 59-66.
6. Zadeh. L. A. Fuzzy Logic, Neural Networks, and Soft Computing [J]. Communications of the ACM, March 1994/Vol. 37, No. 3, 77～84
7．史忠植．知识发现．北京：清华大学出版社[M]，2002
8. J.S.R. Jang C.T.Sun E. Mizutani. Neruo-Fuzzy and Soft Computing [M]. Prentice Hall, 1997
9．刘启阳．Al—Si合金变质机理及共晶硅形态控制的研究与发展[J]．铸造，1988(6)：1～3
10. Apelian.D.Shivkumar S. Sigworth G. Fundamental Aspects of Heat Treatment of Cast Al-Si-Mg Alloys [J], AFS Trans. 1989, 97:727～741
11．刘启阳 李庆春 朱培钺．共晶合金中第二相粒状化学动力学[J]．金属科学与工艺，1987，6(1)：65～77
12. James G. Conley, Julie Huang, Jo Asada and Kenji Akiba, Modeling the effects of cooling rate, hydrogen content, grain refiner and modifier on microporosity formation in Al A356 alloys [J], Materials Science and Engineering A, Volume 285, Issues 1-2, 15 June 2000, Pages 49-55
13. Sang-Won Han, Koji Katsumata, Shinji Kumai and Akikazu Sato, Effects of solidification structure and aging condition on cyclic stress-strain response in Al-7% Si-0.4% Mg cast alloys[J], Materials Science and Engineering A, In Press, Uncorrected Proof, Available online 8 February 2002
14. Arun M. Gokhale and Shichen Yang, Application of Image Processing for Simulation of Mechanical Response of Multi-Length Scale Microstructures of Engineering Alloys[J], Metallurgical and Materials Transactions A, Volume 30A, Issue 9, September 1999, Pages 2369-2381
15．柳百成 荆涛等．铸造工程的模拟仿真与质量控制[M]．北京：机械工业出版社，2001
16．鲁薇华 罗金顺 倪峰．ZL108常规机械性能与化学成分关系及计算[J]．铸造
技术，1988(2)：26～28
17．蔡惠民．ZL108合金中RE、Mg、Cu最佳配比的试验[J]．特种铸造及有色合金，1991 (2)：6～9
18．范志康 王贻青 王家析．模糊回归在改善高碳当量灰铁组织中的应用[J]．铸造，1991(12)：9～12
19. Haykin, Simon., Neural Networks: A Comprehensive Foundation, Macmillan College Publishing, New York,NY, 1994
20. Dobrzanski L A, Sikek W. Gomparison of hardenability calculation methods of the heattreatable constructional steels [J]. Mater. Proc.Tech., 1999,64:117-126.
21. Wang L-X.. Fuzzy System are Universal Approximators[C]. Proc. of IEEE International Conference on Fuzzy Systems, San Diego, CA, 1992, 1163～1170
22. T. Yamakawa and M. Furukawa, A design algorithm of membership functions for a fuzzy neuron using example based learning[C], Proc. 1992 IEEE Int. Conf. on Fuzzy Systems, pp. 75-82.
23. Takagi T, Sugeno M. Fuzzy identification of systems and its application to modeling and control[J]. IEEE Trans on Systems, Man, and Cybernetics, 1985, 15(1): 116～132
24. Jang, J.-S. R., ANFIS: Adaptive-Network-based Fuzzy Inference Systems[J], IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, No. 3, pp. 665-685, May1993.
25．王永强，夏伯才，董杰．材枓性能的模糊神经网络预测[J]．重庆工学院学报．2001，15(5)：5～10．
26．谭志刚 王鸿正 杨荣全．利用散布图实现铸件化学成分控制与管理[J]．机械工程师，1999(4)：45
27. C. H. Caccres, Microstructure Design and Heat Treatment Selection for Casting Alloys Using the Quality Index[J], Journal of Materials Engineering and Performance, Volume 9, Issue 2, April 2000, Pages 215-221
28. Shivkamar S, Ricci S Jr., Apelian D. et al. An experimental study to optimize the heat treatment ofA356 alloy. AFS Tras. 1987:97:791～809
29. Shivkamar S, Ricci S Jr., Apelian D. Influence of solution parameters and simplified supersaturation treatments on tensile properties of A356 alloy. AFS Tarns., 1990:98:913～ 921
30. Tutile M M, Mclellan D L. Silicon particle characteristics in Al-Si-mg castings. AFS Trans.,1982;90:13～ 23
31．丁瑞鑫 张希顺 朱培钺 刘启阳．共晶Si粒状化过程定量金相研究[J]．金属科学与工艺，1984，3(3)：49～59
32．[32]阎承康 张亚克．高强度Al-Si合金的试验研究[J]．特种铸造及有色合金，1987(3)：15～19
33．[33] 黄良余 王玉棕 瞿春泉．Al-Si合金加Sb变质的研究[J]．金属学报 A，1986，22(4)：A310～A316
34. [34] Sang-Won Han, Koji Katsumata, Shinji Kumai and Akikazu Sato, Effects of solidification structure and aging condition on cyclic stress-strain response in Al-7% Si-0.4% Mg cast alloys[J], Materials Science and Engineering A, In Press, Uncorrected Proof, Available online 8 February 2002
35. [35] Shercliff H.R. and Ashby M.F. (1991), Modelling of thermal processing of aluminium alloys[J], Mat. Sci. Tech., Vol.7, January 1991, pp.85-88
36. [36] Shercliff H.R. and Ashby M.E A process model for age-hardening aluminium alloys: Part Ⅰ: The model[J], Acta metall, mater., (Overview) 1990, 38(10): 1789-1802
37. [37] Shercliff H.R. and Ashby M.F. A process model for age-hardening aluminium alloys: Part Ⅱ: Applications[J], Acta metall, mater., (Overview), 1990, 38(10): 1803-1812
38. [38] Shercliff H.R., Myhr O.R. and .Tjotta S. (1994), Modelling of processing of aluminium alloys[J], MRS Bulletin, January 1994, pp.25-28
39. [39] P. A. Rometsch and G. B. Schaffer, An age hardening model for Al-7Si-Mg casting alloys[J], Materials Science and Engineering A, 2002, 325(1-2:424-434
40．[40]宋新华 陈茁 俞汝勤．人工神经网络用于对位取代苯酚定量构效关系的研究．中国科学(B辑)，1993，23(3)：245～251
41. G. Drossel. Der Einflusse von Schmelzebehandlungen auf die Dichttheit von Gusskoerpern aus Al-Si-Gusslegierungen. Giessereitechnik, Vol. 27, No. 1,pp7-12(1981)
42. M. B. Djurdjievic, W. T. Kierkus, G. E. Byczynski, J. H. Sokolowski. Calculation of Liquidus Temperature for Aluminum 3XX Series of Alloys. AFS Trans. 1998, 106:143～147
43. R. Vijayaraghavan, N. Pelle, J. Boilean, J. Zindel, R. Beals. A Micromodel for Al-Si Alloys. Scripta Materialia, Vol. 35, No. 7, pp861-867(1998)
44. Special Issue on "Application of Neural Network Analysis in Materials Science". ISIJ International [J], Vol.39(1999), No. 10
45．刘洪霖 包宏．化工冶金过程人工智能优化[M]．北京：冶金工业出版社，1999
46．王国栋 刘相华．金属轧制过程人工智能优化[M]．北京：冶金工业出版社．2000
2．李成功 傅恒志 于翘．航空航天材料[M]．北京：国防工业出版社．2002
3．刘大利等．新型铝活塞[M]．北京：国防工业出版社．1999
4. J.T. Berry E. P. Coleman, Linking Solidification Conditions and Mechanical Behavior in Al Casting— — A Quarter Century of Evolving Evidence, AFS Trans. 1995,103:837～846.
5. Dobrzanski L A, Sitek W. Application of a neural network in modelling of hardenability of constructional steels [J]. J. Mater. Proc.Tech, 2000,78: 59-66.
6. Zadeh. L. A. Fuzzy Logic, Neural Networks, and Soft Computing [J]. Communications of the ACM, March 1994/Vol. 37, No. 3, 77～84
7．史忠植．知识发现．北京：清华大学出版社[M]，2002
8. J.S.R. Jang C.T.Sun E. Mizutani. Neruo-Fuzzy and Soft Computing [M]. Prentice Hall, 1997
9．刘启阳．Al—Si合金变质机理及共晶硅形态控制的研究与发展[J]．铸造，1988(6)：1～3
10. Apelian.D.Shivkumar S. Sigworth G. Fundamental Aspects of Heat Treatment of Cast Al-Si-Mg Alloys [J], AFS Trans. 1989, 97:727～741
11．刘启阳 李庆春 朱培钺．共晶合金中第二相粒状化学动力学[J]．金属科学与工艺，1987，6(1)：65～77
12. James G. Conley, Julie Huang, Jo Asada and Kenji Akiba, Modeling the effects of cooling rate, hydrogen content, grain refiner and modifier on microporosity formation in Al A356 alloys [J], Materials Science and Engineering A, Volume 285, Issues 1-2, 15 June 2000, Pages 49-55
13. Sang-Won Han, Koji Katsumata, Shinji Kumai and Akikazu Sato, Effects of solidification structure and aging condition on cyclic stress-strain response in Al-7% Si-0.4% Mg cast alloys[J], Materials Science and Engineering A, In Press, Uncorrected Proof, Available online 8 February 2002
14. Arun M. Gokhale and Shichen Yang, Application of Image Processing for Simulation of Mechanical Response of Multi-Length Scale Microstructures of Engineering Alloys[J], Metallurgical and Materials Transactions A, Volume 30A, Issue 9, September 1999, Pages 2369-2381
15．柳百成 荆涛等．铸造工程的模拟仿真与质量控制[M]．北京：机械工业出版社，2001
16．鲁薇华 罗金顺 倪峰．ZL108常规机械性能与化学成分关系及计算[J]．铸造
技术，1988(2)：26～28
17．蔡惠民．ZL108合金中RE、Mg、Cu最佳配比的试验[J]．特种铸造及有色合金，1991 (2)：6～9
18．范志康 王贻青 王家析．模糊回归在改善高碳当量灰铁组织中的应用[J]．铸造，1991(12)：9～12
19. Haykin, Simon., Neural Networks: A Comprehensive Foundation, Macmillan College Publishing, New York,NY, 1994
20. Dobrzanski L A, Sikek W. Gomparison of hardenability calculation methods of the heattreatable constructional steels [J]. Mater. Proc.Tech., 1999,64:117-126.
21. Wang L-X.. Fuzzy System are Universal Approximators[C]. Proc. of IEEE International Conference on Fuzzy Systems, San Diego, CA, 1992, 1163～1170
22. T. Yamakawa and M. Furukawa, A design algorithm of membership functions for a fuzzy neuron using example based learning[C], Proc. 1992 IEEE Int. Conf. on Fuzzy Systems, pp. 75-82.
23. Takagi T, Sugeno M. Fuzzy identification of systems and its application to modeling and control[J]. IEEE Trans on Systems, Man, and Cybernetics, 1985, 15(1): 116～132
24. Jang, J.-S. R., ANFIS: Adaptive-Network-based Fuzzy Inference Systems[J], IEEE Transactions on Systems, Man, and Cybernetics, Vol. 23, No. 3, pp. 665-685, May1993.
25．王永强，夏伯才，董杰．材枓性能的模糊神经网络预测[J]．重庆工学院学报．2001，15(5)：5～10．
26．谭志刚 王鸿正 杨荣全．利用散布图实现铸件化学成分控制与管理[J]．机械工程师，1999(4)：45
27. C. H. Caccres, Microstructure Design and Heat Treatment Selection for Casting Alloys Using the Quality Index[J], Journal of Materials Engineering and Performance, Volume 9, Issue 2, April 2000, Pages 215-221
28. Shivkamar S, Ricci S Jr., Apelian D. et al. An experimental study to optimize the heat treatment ofA356 alloy. AFS Tras. 1987:97:791～809
29. Shivkamar S, Ricci S Jr., Apelian D. Influence of solution parameters and simplified supersaturation treatments on tensile properties of A356 alloy. AFS Tarns., 1990:98:913～ 921
30. Tutile M M, Mclellan D L. Silicon particle characteristics in Al-Si-mg castings. AFS Trans.,1982;90:13～ 23
31．丁瑞鑫 张希顺 朱培钺 刘启阳．共晶Si粒状化过程定量金相研究[J]．金属科学与工艺，1984，3(3)：49～59
32．[32]阎承康 张亚克．高强度Al-Si合金的试验研究[J]．特种铸造及有色合金，1987(3)：15～19
33．[33] 黄良余 王玉棕 瞿春泉．Al-Si合金加Sb变质的研究[J]．金属学报 A，1986，22(4)：A310～A316
34. [34] Sang-Won Han, Koji Katsumata, Shinji Kumai and Akikazu Sato, Effects of solidification structure and aging condition on cyclic stress-strain response in Al-7% Si-0.4% Mg cast alloys[J], Materials Science and Engineering A, In Press, Uncorrected Proof, Available online 8 February 2002
35. [35] Shercliff H.R. and Ashby M.F. (1991), Modelling of thermal processing of aluminium alloys[J], Mat. Sci. Tech., Vol.7, January 1991, pp.85-88
36. [36] Shercliff H.R. and Ashby M.E A process model for age-hardening aluminium alloys: Part Ⅰ: The model[J], Acta metall, mater., (Overview) 1990, 38(10): 1789-1802
37. [37] Shercliff H.R. and Ashby M.F. A process model for age-hardening aluminium alloys: Part Ⅱ: Applications[J], Acta metall, mater., (Overview), 1990, 38(10): 1803-1812
38. [38] Shercliff H.R., Myhr O.R. and .Tjotta S. (1994), Modelling of processing of aluminium alloys[J], MRS Bulletin, January 1994, pp.25-28
39. [39] P. A. Rometsch and G. B. Schaffer, An age hardening model for Al-7Si-Mg casting alloys[J], Materials Science and Engineering A, 2002, 325(1-2:424-434
40．[40]宋新华 陈茁 俞汝勤．人工神经网络用于对位取代苯酚定量构效关系的研究．中国科学(B辑)，1993，23(3)：245～251
41. G. Drossel. Der Einflusse von Schmelzebehandlungen auf die Dichttheit von Gusskoerpern aus Al-Si-Gusslegierungen. Giessereitechnik, Vol. 27, No. 1,pp7-12(1981)
42. M. B. Djurdjievic, W. T. Kierkus, G. E. Byczynski, J. H. Sokolowski. Calculation of Liquidus Temperature for Aluminum 3XX Series of Alloys. AFS Trans. 1998, 106:143～147
43. R. Vijayaraghavan, N. Pelle, J. Boilean, J. Zindel, R. Beals. A Micromodel for Al-Si Alloys. Scripta Materialia, Vol. 35, No. 7, pp861-867(1998)
44. Special Issue on "Application of Neural Network Analysis in Materials Science". ISIJ International [J], Vol.39(1999), No. 10
45．刘洪霖 包宏．化工冶金过程人工智能优化[M]．北京：冶金工业出版社，1999
46．王国栋 刘相华．金属轧制过程人工智能优化[M]．北京：冶金工业出版社．2000