BP和RBF神经网络对复杂型面零件点云漏洞的修补应用  

Application of BP and RBF Neural Network in Mending in the 3D Incomplete Point Clouds of Complex Surface Parts

作　　者：王春香[1] 张勇[1] 梁亮 王岩辉 WANG Chun-xiang;ZHANG Yong;LIANG Liang;WANG Yan-hui(School of Mechanical Engineering,Inner Mongolia University of Technology,Baotou Inner Mongolia,014010,China)

机构地区：[1]内蒙古科技大学机械学院,内蒙古包头014010

出　　处：《组合机床与自动化加工技术》2018年第3期118-120,共3页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：内蒙古自治区高等学校科学研究项目(NJZY16167);内蒙古自治区自然科学基金项目(2017MS(LH)0530)

摘　　要：在难以利用软件成功修补复杂型面的较大漏洞情况下,为了获得精确和完整复杂型面零件点云的三维模型,应用BP和RBF神经网络对精度要求高的挖掘机斗齿内腔人为漏洞修补,误差对比分析表示,BP修补效果较RBF更佳。考虑到工程实际中应用,精度要求和效率上,以复杂型面点云机架为例,实验表明,BP算法取得了很好的修补效果,该修补方法在漏洞修补上比软件修补和RBF修补效果好且效率高,为后续复杂型面点云数据处理提供了参考依据。In case of applying to the software isn′t success to mend the large 3D incomplete point clouds,in order to obtain accurate and whole 3D point clouds model of complex surface parts,application of BP and RBF neural network mend excavator bucket teeth lumen of the high accurancy requirement,error analysis show that BP repair effect is better than RBF.Considering the partical engineering application,accurancy and efficency,taking complex surface rack of point clouds for example,the experimental results show that BP algorithm has achieved very good repair effect.the method of mending point clouds hole than software repair and RBF repair good effect and efficiency by using the method for complex surface point cloud provide some reference base.

关 键 词：复杂型面 机架 BP和RBF神经网络 漏洞修补 

分 类 号：TH166[机械工程—机械制造及自动化] TG506[金属学及工艺—金属切削加工及机床]