基于LM-BP神经网络的液压支架顶梁疲劳寿命预测及应用  

Fatigue life prediction and application of hydraulic support roof beam based on LM-BP neural network

作　　者：李世科[1] LI Shike(School of Computer Engineering,Henan Institute of Economics and Trade,Zhengzhou 450046,China)

机构地区：[1]河南经贸职业学院计算机工程学院,河南郑州450046

出　　处：《中国矿业》2019年第5期92-96,共5页China Mining Magazine

摘　　要：本文提出基于LM-BP神经网络进行液压支架顶梁疲劳寿命预测方法,选取主筋板厚度、柱窝上方中心处横板厚度、两侧横板厚度、导向套筒孔半径、顶板厚度作为输入参量,将样本的液压支架顶梁疲劳寿命作为输出量,在进行训练时采用LM算法对BP神经网络进行改进,得到基于LM的BP神经网络模型,利用该模型进行液压支架顶梁疲劳寿命预测。研究结果表明:基于LM的BP神经网络模型的计算结果与测试样本拟合精度较高,具有广泛的应用前景。The method for predicting the fatigue life of hydraulic support roof beam is presented based on LM-BP neural network in this paper.The main stiffener thickness,the thickness of the transverse plate above the center of the column socket,the thickness of the transverse plates on both sides,the radius of the guide sleeve hole and the thickness of the roof are selected as input parameters,the fatigue life of the sample hydraulic support roof beam is taken as output,LM algorithm is used to improve BP neural network in training,and a BP neural network model based on LM is obtained.The model is used to predict the fatigue life of hydraulic support roof beam.The results show that the fitting accuracy of the BP neural network model based on LM is high,and it has wide application prospects.

关 键 词：LM算法 BP神经网络 液压支架 顶梁疲劳寿命 

分 类 号：TD355[矿业工程—矿井建设]