基于人工神经网络和遗传算法,结合177组不锈钢复合板实测数据,构建不锈钢复合板剪切强度模型.研究确定不锈钢铬当量、铬镍当量比、复合板覆层厚度以及基材厚度为网络输入量,复合板剪切强度为输出量,隐含层节点数由试探寻优法确定,优化网络结构为4-7-1；比较Levenberg-Marquardt、Quick-Propagation、Standard Back-Propagation算法的训练误差、测试误差及计算迭代步数,确定以误差最小、计算速度最快的LM算法训练网络；另外,利用提前终止法避免ANN模型产生的过拟合的问题；在此基础上,引入遗传算法进一步优化ANN网络的权值和阈值,使得复合板剪切强度预测值与实测值相关系数达到0.997;将所构建模型用于实际不锈钢复合板剪切强度的预测,与实测值相近,进一步验证预测模型的有效性和可靠性.
参考文献
| [1] | 焦少阳,董建新,张麦仓,郑磊.双金属热轧复合的界面结合影响因素及结合机理[J].材料导报,2009(01):59-62. |
| [2] | 吴维,温彤,蒲思洪.层状复合板层间剪切强度的研究现状[J].机械制造,2009(01):34-37. |
| [3] | 肖纪美.不锈钢的金属学问题[M].北京:冶金工业出版社,2006:49-53. |
| [4] | Mehmet Sirac Ozerdem;Sedat Kolukisa .Artificial Neural Network Approach To Predict The Mechanical Properties Of Cu-sn-pb-zn-ni Cast Alloys[J].Materials & design,2009(3):764-769. |
| [5] | 乌日根,董俊慧,王玉荣.基于神经网络的RE-Ni-Cu合金铸铁腐蚀性能预测[J].兵器材料科学与工程,2009(01):28-31. |
| [6] | S.H. Mousavi Anijdan;A. Bahrami;H.R. Madaah Hosseini .Using genetic algorithm and artificial neural network analyses to design an Al-Si casting alloy of minimum porosity[J].Materials & Design,2006(7):605-609. |
| [7] | 甘忠,成武冬,许旭东,潘栋杰.BP神经网络预测铝合金应力松弛量研究[J].兵器材料科学与工程,2009(02):35-38. |
| [8] | 邹德宁,李娇,李云华,张子义.神经网络的BP算法研究高速钢轧辊的热处理工艺[J].铸造技术,2007(11):1518-1521. |
| [9] | Zhiyu Chen;Dening Zou;Junhui Yu;Ying Han.Artificial Neural Network Approach to Predict Mechanical Properties of 301 Austenitic Stainless Steel[J].Materials Science Forum,2010:145-148. |
| [10] | M. Zakeri;A. Bahrami;S.H. Mousavi Anijdan .Using genetic algorithm in heat treatment optimization of 17-4PH stainless steel[J].Materials & Design,2007(7):2034-2039. |
| [11] | H. Mirzadeh;A. Najafizadeh .ANN modeling of strain-induced martensite and its applications in metastable austenitic stainless steels[J].Journal of Alloys and Compounds: An Interdisciplinary Journal of Materials Science and Solid-state Chemistry and Physics,2009(1/2):352-355. |
| [12] | Blanco A;Delgado M;Pegalajar MC .A real-coded genetic algorithm for training recurrent neural networks.[J].Neural Networks: The Official Journal of the International Neural Network Society,2001(1):93-105. |
上一张
下一张
上一张
下一张
计量
- 下载量()
- 访问量()
文章评分
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%