基于GWO-LSTM的航空零件质量预测方法  

Quality Prediction Mthod of Aerospace Parts Based on GWO-LSTM

作　　者：尹佳 张思超 贾保国 段晓蕊 张茜[3] YIN Jia;ZHANG Sichao;JIA Baoguo;DUAN Xiaorui;ZHANG Qian(AVIC Xi'an Aircraft Industry Group Company Ltd.,Xi'an 710089,China;School of Electro Mechanical Engineering,Xidian University,Xi'an 710071,China;The 10th Research Institute of China Electronics Technology Group Corporation,Chengdu 610036,China)

机构地区：[1]中航西安飞机工业集团股份有限公司,陕西西安710089 [2]西安电子科技大学机电工程学院,陕西西安710071 [3]中国电子科技集团公司第十研究所,四川成都610036

出　　处：《机械制造与自动化》2025年第2期170-176,共7页Machine Building & Automation

基　　金：陕西省科技重大专项项目(2019zdzx01-01-02)。

摘　　要：为了提高航空零件的加工质量,使用深度学习算法对原始加工误差数据集进行增强,构建长短记忆神经网络预测未来零件的误差,并使用灰狼算法对预测模型关键参数进行优化。实验结果表明:与传统的深度学习质量预测方法对比,质量预测精度有明显提升。因此,GWO-LSTM可以有效提高航空零件质量预测精度,提高产品生产质量,具有一定的工程实践价值。In order to improve the machining quality of aerospace parts,the deep learning algorithm was used to enhance the original machining-errors data set,the long short-term memory was constructed to predict the errors of future parts,and the grey wolf algorithm was applied to optimize the key hyperparameters of the prediction model.The experimental result shows that the accuracy of the improved deep learning method for quality prediction is significantly raised compared with the traditional one.Thus,GWO-LSTM can effectively improve the accuracy of quality prediction of aerospace parts as well as the production quality,which has a certain engineering practice value.

关 键 词：航空零件 质量预测 参数优化 样本增强 长短记忆神经网络 灰狼算法 

分 类 号：V262.3[航空宇航科学与技术—航空宇航制造工程]