基于LSTM和贝叶斯网络的枪械交验合格率预测  

Prediction of Firearms Acceptance Rate Based on LSTM and Bayesian Network

作　　者：王宪升 胡瑶 姜黎明 郝佳[2] 孙嘉伟 张晓宁 陈东阳 Wang Xiansheng;Hu Yao;Jiang Liming;Hao Jia;Sun Jiawei;Zhang Xiaoning;Chen Dongyang(Institute of Process Technology,Chongqing Jianshe Industry(Group)Co.,Ltd.,Chongqing 400054,China;School of Mechanical Engineering,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]重庆建设工业(集团)有限责任公司工艺技术研究所,重庆400054 [2]北京理工大学机械与车辆学院,北京100081

出　　处：《兵工自动化》2025年第4期26-31,共6页Ordnance Industry Automation

摘　　要：为准确定位影响成枪一次交验合格率的关键加工环节,选取贝叶斯网络构建加工参数与合格率之间的因果模型。通过选取长短期记忆(long short-term memory,LSTM)神经网络模型作为成枪一次交验合格率的时间序列预测模型,能较准确地预测下一批次的成枪一次交验合格率,进一步定位到关键加工环节。结果表明,该预测可为下一步有针对性地改进生产工艺提供理论参考。In order to accurately locate the key processing links affecting the pass rate of the first delivery of the finished gun,the Bayesian network is selected to construct a causal model between the processing parameters and the pass rate.By selecting the long short-term memory(LSTM)neural network model as the time series prediction model for the pass rate of the first delivery of guns,the pass rate of the first delivery of guns in the next batch can be predicted more accurately,and the key processing links can be further located.The results show that the prediction can provide a theoretical reference for the next targeted improvement of the production process.

关 键 词：交验合格率 预测 LSTM模型 贝叶斯网络 

分 类 号：TJ2[兵器科学与技术—武器系统与运用工程]