BP和RBF神经网络在起重机钢丝绳断丝数量预测中的比较研究  

Comparison Study of BP and RBF Neural Network in Prediction of Number of Broken Wires of Wire Ropes for Cranes

作　　者：曹玲芝[1] 安治恒 

机构地区：[1]郑州轻工业学院,郑州450002

出　　处：《煤矿机械》2016年第1期187-189,共3页Coal Mine Machinery

摘　　要：利用MATLAB工具箱函数分别建立了起重机钢丝绳断丝数目检测的BP神经网络和RBF神经网络模型,并对2种模型的结构、预测精度和训练过程进行了对比研究。结果表明,在一定的样本集和训练条件下,BP和RBF神经网络均能对钢丝绳的断丝数目进行很好预测,可以解决峰值、阀值波宽、小波能量和波形下面积对钢丝绳断丝数目的非线性映射关系,能够满足工程预测的需要。但RBF神经网络较BP神经网络在迭代次数、收敛速度和网络结构方面更具优势,因此其预测能力和泛化能力都优于BP神经网络。The BP neural network and RBF neural network model for the detection of the broken wires of the crane＇s wire ropes are established by using the MATLAB toolbox function, and the structure, prediction accuracy and training process of the two models were compared. The results show that in the sample set and training conditions, BP and RBF neural network can predict the number of wire ropes, and also can solve the peak, threshold wave width, area under the wavelet energy and waveform of wire rope broken wire number of nonlinear mapping relationship, which can meet the needs of engineering prediction. However, the RBF neural network has more advantages in the number of iterations, the convergence speed and the network structure, so the prediction ability and generalization ability of BP neural network is better than that of BP neural network.

关 键 词：BP神经网络 RBF神经网络 预测 

分 类 号：TD183[矿业工程—矿山地质测量]