摘要
为了保证水下爆炸容器在服役期间的安全性,有必要进行容器的动态响应预测。对服役期的水下爆炸容器在不同载荷条件下进行动态响应测试试验,选取时间、药量、加载静水压和应变片位置的11个亚变量,共14个影响因素作为输入变量,容器的最大应力作为输出变量,建立基于GRNN的水下爆炸容器动态响应预测模型,仿真载荷与容器应变的映射关系,并通过10折交叉验证法验证了该模型具有较好的预测性能。同时对比基于BPNN的预测模型,GRNN模型的拟合与预测性能明显优于BPNN模型,进一步说明了GRNN方法在水下爆炸容器动态响应预测过程中的有效性。
In order to ensure the safety of the underwater explosion vessels in service,it is necessary to predict the dynamic response of the container.The dynamic response tests were carried out under different load conditions on the underwater explosion vessel in the service period.14 influencing factors were selected as input variables,including time,the amount of explosives,hydrostatic pressures,and 11 dummy variables about strain gauge positions,and the maximum stress of the vessel was identified as output variable.The dynamic response prediction model of the vessel based on GRNN was established to simulate the mapping relationship between the load and the strain of vessel. Moreover,the model was verified as a better predictive performance through the 10-fold cross-validation method.At the same time,compared with the BPNN prediction model,the fitting and prediction performance of the GRNN model were significantly better than the BPNN model,which further demonstrates the effectiveness of the GRNN method in the dynamic response prediction process of underwater explosion vessels.
作者
李琳娜
李甜
钟冬望
涂圣武
刘洋
LI Lin-na;LI Tian;ZHONG Dong-wang;TU Sheng-wu;LIU Yang(College of Science,Wuhan University of Science and Technology,Wuhan 430065,China;Hubei Key Laboratory of Process System Science in Metallurgy Industry,Wuhan University of Science and Technology,Wuhan 430065,China;Department of Electrical Engineering,Wuhan Electric Technology College,Wuhan 430079,China)
出处
《爆破》
CSCD
北大核心
2018年第4期141-146,共6页
Blasting
基金
国家自然科学基金资助项目(51404175
51574184)
冶金工业过程系统科学湖北省重点实验室开放基金资助项目(Y201712)
武汉科技大学国防预研基金资助项目(GF201708)
关键词
水下爆炸容器
动态响应
广义回归神经网络
underwater explosion vessels
dynamic response
general regression neural network