采用热压成型的方法制备掺杂粉煤灰、以无机纤维为增强体的摩擦材料,并测试其磨损性能。选用BP神经网络建模,以摩擦材料配方、制备工艺、测试条件为输入变量,以材料的磨损率为输出变量,采用L-M算法对网络进行训练。结果表明,模型可以对材料磨损性能进行有效的预测,可用于配方及制备工艺的优化。
The friction materials filled with fly-ash and reinforced with inorganic fiber were prepared by heat-pressing molding method,and the wear properties of these materials were tested.The developed BP neural network in this paper was employed with the formulas of friction materials and manufacturing conditions in synergy with operating system as input and the wear volume as output,and the Levenberg-Marquardt algorithms were used for training the network.The results show that the model is valid to predict the wear properties,as well as that it is useful for optimizing the formation and manufacturing conditions.
参考文献
| [1] | 王科俊;王克成.神经网络建模预报与控制[M].哈尔滨:哈尔滨工程大学出版社,1996 |
| [2] | 吴耀庆,曾鸣,范力仁.基于BP神经网络的汽车摩擦材料性能预测与配方优化[J].材料导报,2010(10):74-78. |
| [3] | Guipu Xiao;Zikang Zhu .Friction materials development by using DOE/RSM and artificial neural network[J].Tribology International,2010(1/2):218-227. |
| [4] | Dragan Aleksendric .Neural network prediction of brake friction materials wear[J].Wear: an International Journal on the Science and Technology of Friction, Lubrication and Wear,2010(1/2):117-125. |
| [5] | Sudipto Ray;S. K. Roy Chowdhury .Prediction of contact temperature rise between rough sliding bodies: An artificial neural network approach[J].Wear: an International Journal on the Science and Technology of Friction, Lubrication and Wear,2009(9/10):1029-1038. |
| [6] | 李小雷.人工神经网络在高掺量粉煤灰混凝土配制中的应用研究[M].北京:煤炭工业出版社,2006 |
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%