基于进化神经网络的激光熔覆层质量预测

  本文关键词：基于神经网络和遗传算法的激光多层熔覆厚纳米陶瓷涂层工艺优化，由笔耕文化传播整理发布。

摘要

为了有效地控制激光熔覆层质量，采用反向传播(BP)算法建立了激光熔覆层质量(熔覆层宽度、熔覆层深度和稀释率)随激光功率、光斑直径和扫描速度变化的进化神经网络预测模型。针对BP算法存在收敛速度慢、容易陷入局部极小值及全局搜索能力弱等缺陷，采用遗传算法训练BP神经网络，取代了一些传统的学习算法，设计了基于进化神经网络的学习算法。经过理论分析和实验验证，在神经网络的输出端得到期望的线性输出，并在训练样本之外，选取了5组工艺参数检验神经网络模型的可靠性，预测结果与相应的实验结果的最大相对误差为2．14%。结果表明，运用该模型可以方便、准确地选择激光工艺参数，提高激光熔覆层的加工质量。

关键词

Abstract

Artificial neural networks were introduced in the area of laser cladding forming. The prediction model of surface quality in laser cladding parts,including the width,depth of cladding layer and dilution,was proposed based on the improved learned arithmetic. The model combined the global optimization searching performance of the genetic algorithm and the localization of the back propagation(BP)neural networks. Five technical parameters were selected to test the reliability of the model. The simulation and experimental results show that the evolutionary neural network based on genetic algorithm can effectively overcome the problem of falling into local minimum point. This method can get higher accuracy of prediction. It improves the measurement precision with the maximum relative error 2.14% between the predicted content and the real value.

补充资料

中图分类号：TG156．9;TP183

所属栏目：激光与光电子技术应用

收稿日期：2006-07-26

修改稿日期：2006-09-06

网络出版日期：0001-01-01

作者单位    点击查看

徐大鹏：江苏大学 机械工程学院，镇江 212013
周建忠：江苏大学 机械工程学院，镇江 212013
郭华锋：江苏大学 机械工程学院，镇江 212013
季霞：江苏大学 机械工程学院，，镇江 212013

联系人作者：周建忠(zhoujz@ujs.edu.cn)

备注：徐大鹏|男|硕士研究生|主要从事基于激光熔覆的金属零件快速制造技术的研究|(1979－)

【1】LIU X M,GUAN Z Z. The relationship between the process parameter of laser cladding by powder feeding method and the laser layer parameters [J].Transactions of Metal Heat Treatment,1998,A26(5):29～34.

【2】van ACKER K,VANHOYWEGHEN D,PERSOONS R et al. Influence of tungsten carbide particle size and distribution on the wear resistance of laser clad WC/Ni coatings [J].Wear,2005,258(1～4):194～202.

【3】SONG J L,DENG Q L,CHEN Ch Y et al. Rebuilding of metal components with laser cladding forming [J].Applied Surface Science,2006,252(22):7934～7940.

【4】WU X W,ZHU B D,ZENG X Y et al. Critical state of laser cladding with power auto-feeding [J].Surface and Coatings Technology,1996,79(l～3):200～204.

【5】JI S Q,LI P,ZENG X Y. Microstructure and mechanical property analyses of the metal parts direct fabricated by laser cladding [J].Laser Technology,2006,30(2):130～133(in Chinese).

【6】YAN P F,ZHANG C S. Artificial neural network and simulated evolutionary computing [M].Beijing: Tsinghua University Press,2000.15～22(in Chinese).

【7】LEI Y J,ZHANG S W,LI X W et al. MATLAB genetic algorithm toolbox and application [M].Xi''an : University of Electronic Science and Technology of China Press,2005.15～33(in Chinese).

【8】STEEN W M,WEERASINGHE V M,MONSON P. Some aspects of the formation of laser clad tracks [J].SPIE,1996,650:226～234.

【9】LIU J C,LIU L J. Experimental study on fabrication of thin-wall metallic features by laser cladding [J].Chinese Journal of Mechanical Engineering,2004,40(10):185～188(in Chinese).

【10】ZHU B D,ZENG X Y,TAO C Y et al. Effect of laser processing parameters on dilution of the cladded coating [J].Chinese Journal of Materials Research,1994,8(4):315～319(in Chinese).

引用该论文

XU Da-peng,ZHOU Jian-zhong,GUO Hua-feng,JI Xia. Quality prediction of laser cladding layer based on improved neural network[J]. Laser Technology, 2007, 31(5): 0511

徐大鹏,周建忠,郭华锋,季霞. 基于进化神经网络的激光熔覆层质量预测[J]. 激光技术, 2007, 31(5): 0511

被引情况

【1】邵珺,华文深,周中亮,高鸿启. 神经网络和遗传算法在相关峰判读中的应用. 激光技术, 2009, 33(4): 422-425

【2】张毅,姚建华,胡晓冬,陈智君. 激光再制造粉末输送流量检测系统设计. 激光技术, 2009, 33(6): 568-570

【3】王东生,杨友文,田宗军,沈理达,黄因慧. 基于神经网络和遗传算法的激光多层熔覆厚纳米陶瓷涂层工艺优化. 中国激光, 2013, 40(9): 903001--1

  本文关键词：基于神经网络和遗传算法的激光多层熔覆厚纳米陶瓷涂层工艺优化，由笔耕文化传播整理发布。