基于神经网络激光修整砂轮去除效率的预测  

Prediction of Removal Efficiency of Laser Dressing Wheels Based on Neural Networks

作　　者：朱毅 梅丽芳 黄佳成 周伟[3,4] 陈根余 王昊[3,4] Zhu Yi;Mei Lifang;Huang Jiacheng;Zhou Wei;Chen Genyu;Wang Hao(College of Mechanical and Automotive Engineering,Xiamen University of Technology,Xiamen 361024,Fujian,China;Fujian Key Laboratory of Advanced Design and Manufacture of Passenger Cars,Xiamen 361024,Fujian,China;College of Mechanical and Vehicle Engineering,Hunan University,Changsha 410082,Hunan,China;Laser Research Institute of Hunan University,Hunan University,Changsha 410082,Hunan,China)

机构地区：[1]厦门理工学院机械与汽车工程学院,福建厦门361024 [2]福建省客车先进设计与制造重点实验室,福建厦门361024 [3]湖南大学机械与运载工程学院,湖南长沙410082 [4]湖南大学激光研究所,湖南长沙410082

出　　处：《应用激光》2025年第1期58-67,共10页Applied Laser

基　　金：国家科技重大专项(2012ZX04003101)。

摘　　要：精准的去除效率预测对纳秒激光修整青铜金刚石砂轮具有重大意义,以BP神经网络建立激光修整青铜金刚石砂轮工艺参数的预测模型,利用CIGWO算法进行改进。首先,基于BP神经网络模型的拓扑关系,确定出模型的输入层、隐含层及输出层节点数,构建工艺参数与砂轮单位时间去除量之间的映射关系;然后,采用Cricle混沌映射自适应权重灰狼算法对建立的预测模型实现工艺参数寻优;最后,根据反向传播训练结果对比试验测量真实值。结果表明,与传统BP网络模型相比,提出的CIGWO-BP预测算法精度提高了3.32%,真实值与CIGWO-BP预测值的平均相对误差在5.4%以内。综合说明该优化模型为建立激光修整砂轮去除效率的预测提供了一种方式。This paper presents a prediction model for the process parameters of laser dressing bronze diamond grinding wheels,utilizing a BP neural network enhanced by the CIGWO algorithm.Based on the topological relationship of the BP neural network model,the number of nodes in the input layer,implicit layer and output layer of the model was determined firstly,and the mapping relationship between the process parameters and the removal amount per unit time of the grinding wheel was constructed,then the Cricle chaotic mapping adaptive weight grey wolf algorithm was used to optimise the established prediction model for the process parameters,and finally the real values of the experimental measurements were compared according to the back propagation training results.The results showed that the CIGWO-BP prediction algorithm achieves a 3.32%higher accuracy than the traditional BP network model,with an average relative error of less than 5.4%between the predicted and actual values.In summary,the optimized model offers a robust approach for predicting the removal efficiency of laser-dressed grinding wheels.

关 键 词：神经网络 灰狼算法 激光加工 青铜金刚石砂轮 

分 类 号：TN249[电子电信—物理电子学]