摘要
针对火炮身管烧蚀磨损量预测中存在的数据采样时间间隔不均匀、采样难度大、成本高、数据量小,常规数据拟合和预测方法难以处理等问题,提出一种基于改进的非等间隔灰色理论和BP神经网络的组合预测方法。通过组合预测模型对某型火炮身管的烧蚀磨损量进行预测,实例分析表明该组合预测模型具有较高的预测精度,为身管内膛磨损量的预测提供了一种新的技术途径。
General data fitting and prediction methods are constrained by the unequal time interval, difficult sampling, high cost and small amount of data in predicting of the wear degrees of gun barrel. A combined prediction method based on the improved unequal interval grey model and neural network is proposed. The proposed method is used to predict the wear of gun tube. The predicted results agree well with the experimental values. The results show that the combined prediction method has high prediction accuracy, and therefore can be used to effectively predict the gun bore wear.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
2016年第12期2220-2225,共6页
Acta Armamentarii
基金
江苏省自然科学基金项目(BK20140789)
中央高校基本科研业务费专项资金项目(30915118826)
关键词
兵器科学与技术
身管磨损
非等间隔
灰色模型
BP神经网络
组合预测模型
ordnance science and technology
gun barrel wear
unequal Interval
grey model
BP neural network
combined prediction method
