锻压技术

基于RBF神经网络的强力旋压连杆衬套力学性能预测研究

英文标题：Study on mechanical property prediction of power spinning connecting rod bushing based on RBF neural network

作者：佘勇樊文欣陈东宝曹存存

单位：中北大学

分类号：TG376

出版年,卷(期):页码：2016,41(6):128-132

摘要：

针对难以运用公式来表达强力旋压连杆衬套工艺参数与力学性能之间的复杂关系问题，建立了旋压工艺参数（减薄率、热处理温度、进给比）与力学性能（布氏硬度、伸长率、屈服强度、抗拉强度）之间的径向基函数（RBF）神经网络模型。用实验所得的数据对RBF神经网络进行训练，再用训练好的RBF神经网络对成形件的力学性能进行预测，通过与实测值对比分析，并与用BP神经网络所建模型的预测结果进行比较，发现RBF神经网络模型具有较BP神经网络更优的预测性能。RBF神经网络模型预测能力强、建模时间短、能有效提高连杆衬套工艺的设计效率和降低实际实验的所需成本。

For the difficulty of expressing the complex relationships between process parameters and mechanical property of connecting rod bushing formed in power spinning using a formula, it was established the model of RBF neural network between the spinning process parameters (thinning ratio, heat treatment temperature and feed ratio) and mechanical property (brinell hardness, elongation, yield strength and tensile strength). Then, the RBF neural network was trained by experimental data, and the mechanical property of formed part was predicted by the above trained RBF neural network. Comparing the experimental data with the predicted results of BP neural network, it is found that the RBF neural network has better prediction performance than BP neural network. Therefore, the RBF neural network model is of high prediction performance and short modeling time. Thus, it can effectively improve the design efficiency of connecting rod bushing process and reduce the cost in actual experiment.

基金项目：

基金项目:山西省自然科学基金资助项目(2012011023-2)；山西省高校高新技术产业化项目(20120021)

作者简介：佘勇（1990-），男，硕士研究生 E-mail：sy09020641@163.com 通讯作者:樊文欣（1964-），男，博士，教授 E-mail：fanwx@nuc.edu.cn

参考文献：

［1］原霞，王铁，樊文欣，等.基于simufact 的连杆衬套旋压工艺参数的模拟研究[J]. 热加工工艺，2013，42(1)：63-66. Yuan X, Wan T, Fan W X, et al. Simulation research on spinning technological parameters of connecting rod bushing based on simufact[J]. Hot Working Technology, 2013, 42(1): 63-66.

［2］张利鹏，刘智冲，周宏宇. 筒形件强力旋压发展过程及其现状分析[J].塑性工程学报，2006，13(1)：43-46，57.Zhang L P, Liu Z C, Zhou H Y. Development process and current situation analysis of power spinning for cylindrical parts[J]. Journal of Plasticity Engineering, 2006, 13(1): 43-46，57.

［3］樊文欣，张涛，宋河金，等. 强力旋压加工的高速柴油机连杆衬套[J].车用发动机，1997,(2)：32-35.Fan W X, Zhang T, Song H J, et al. A connecting rod bushing of high speed diesel engines by the powerful swivel press technique[J]. Vehicle Engine, 1997,(2): 32-35.

［4］雷萍. 小波神经网络技术在齿轮箱轴承故障诊断中的应用[D].兰州：兰州理工大学，2009.Lei P. Application of Wavelet Neural Network to Gearbox Bearings Fault Diagnosis[D]. Lanzhou: Lanzhou University of Technology, 2009.

［5］冯志刚，樊文欣，赵俊生，等. 基于BP神经网络的强力旋压成形连杆衬套壁厚预测[J].热加工工艺，2014，43(3)：129-130，134.Feng Z G, Fan W X, Zhao J S, et al. Wall thickness prediction of connecting rod bushing of power spinning forming based on BP neural network[J]. Hot Working Technology, 2014, 43(3): 129-130, 134.

［6］李涛，樊文欣，赵俊生，等. 基于BP神经网络的强力旋压成形本构关系模型[J]. 锻压技术，2014，39(2)：150-153.Li T, Fan W X, Zhao J S, et al. Research on constitutive relation of tube power spinning forming based on BP neural network[J]. Forging & Stamping Technology, 2014, 39(2): 150-153.

［7］盛仲飙. BP神经网络原理及MATLAB仿真[J].渭南师范学院学报，2008，23(5)：65-67.Sheng Z B. Principle of BP neural network and MATLAB simulation[J]. Journal of Weinan Teachers University, 2008, 23(5): 65-67.

［8］宋献锋，张克辉. 基于模糊RBF神经网络的板带横向厚度和纵向厚度综合控制[J].热加工工艺，2012，41(13)：132-137.Song X F, Zhang K H. Strip horizontal thickness and vertical thickness complex cxontrol based on fuzzy RBF neural-network[J]. Hot Working Technology, 2012, 41(13): 132-137.

［9］田银，谢延敏，孙新强，等. 基于鱼群RBF神经网络和改进蚁群算法的拉深成形工艺参数优化[J]. 锻压技术，2014，39(12)：129-136.Tian Y, Xie Y M, Sun X Q, et al. Process parameters optimization of deep drawing based on fish RBF neural network and improved ant colony algorithm[J]. Forging & Stamping Technology, 2014,39(12): 129-136.

［10］刘维群，李为华. 基于自组织选取中心的广义RBF神经网络学习算法[J]. 信阳师范学院学报:自然科学版，2007，20(4)： 515-517. Liu W Q, Li W H. An algorithm for generalized RBF network based on self-organizing selection center[J]. Journal of Xinyang Normal University:Natural Science Edition, 2007, 20(4): 515-517.

［11］尤文坚，叶雪英，唐仕云. 基于径向基神经网络农机数量预测的研究[J].中国农机化学报，2013，34(2)：38-41.You W J, Ye X Y, Tang S Y. Research on forecast of the number of agricultural machinery based on RBF neural network[J]. Journal of Chinese Agricultural Mechanization, 2013, 34(2): 38-41.

服务与反馈：

【文章下载】【加入收藏】