摘要
为了解决液压支架前连杆可靠性评估困难的问题,提出了一种基于径向基神经网络的液压支架前连杆可靠性评估的方法。首先使用ANSYS参数化语言APDL对液压支架前连杆进行静力分析,确定与前连杆可靠性相关的设计变量;再使用ANSYS/PDS模块结合拉丁超立方抽样法对前连杆进行可靠性分析,获取多组前连杆不同设计变量的可靠度;最后使用径向基(RBF)神经网络拟合设计变量与可靠性之间的函数关系,建立前连杆可靠性评估模型,预测前连杆的可靠度。计算结果表明,前连杆可靠度计算结果的最大相对误差为4.46%,最小误差为1.59%。证明了径向基神经网络应用于液压支架前连杆可靠性评估的可行性,为液压支架前连杆可靠性评估提供了新的方法与思路。
In order to solve the problem that the reliability evaluation of the front connecting rod of the hydraulic support was difficult,a method based on radial basis function(RBF)neural network was proposed to evaluate the reliability of the front connecting rod of the hydraulic support.Firstly,by using the ANSYS parametric language APDL,the static analysis of the front connecting rod of the hydraulic support was carried out to determine the design variables related to the reliability of the front connecting rod.Then,the reliability analysis on the front connecting rod was carried out by using the ANSYS/PDS module and the Latin hypercube sampling method,so as to obtain the reliability of the front connecting rods with different design variables.Finally,the RBF neural network was used for fitting the functional relationship between design variables and reliability,establishing the reliability evaluation model of the front connecting rod,and predicting its reliability.The results showed that the maximum relative error of the calculation of the front connecting rod reliability was 4.46%and the minimum error was1.59%.It was proved that the RBF neural network was applied to the reliability evaluation of the front connecting rod of the hydraulic support,and provided a new method and idea for the reliability evaluation of the front connecting rod of the hydraulic support.
作者
钱鹏
陆金桂
朱正权
QIAN Peng;LU Jinggui;ZHU Zhengquan(School of Mechanical and Dynamic Engineering,Nanjing Tech University,Nanjing,Jiangsu 211816,China)
出处
《矿业研究与开发》
CAS
北大核心
2019年第1期110-113,共4页
Mining Research and Development
基金
国家"十二五"科技支撑计划资助项目(2013BAF02B11)
关键词
液压支架
前连杆
有限元分析
径向基神经网络
可靠性
Hydraulic support
Front connecting rod
Finite element analysis
Radial basis function (RBF)neural network
Reliability
