论文部分内容阅读
针对BP神经网络存在局部极小值和收敛速度慢等问题,提出了一种RPROP的改进的BP神经网络。RPROP神经网络具有优良的非线性映射能力,可以很好地描述频率特征和诊断结果之间的关系。本文利用MATLAB结合齿轮箱故障建立了标准BP神经网络和本文提出的改进BP神经网络的两个故障诊断模型,并对其性能做了分析和对比。实验表明,基于改进的BP神经网络的齿轮箱故障诊断方法可以大大提高故障诊断的精确性,缩短了诊断时间。
Aiming at the problems of local minima and slow convergence of BP neural network, an improved BP neural network based on RPROP is proposed. RPROP neural network has excellent nonlinear mapping ability, which can well describe the relationship between frequency characteristics and diagnosis results. In this paper, MATLAB and gearbox fault were used to establish a standard BP neural network and the improved BP neural network proposed in this paper, two fault diagnosis model, and its performance analysis and comparison. Experiments show that the gearbox fault diagnosis method based on improved BP neural network can greatly improve the accuracy of fault diagnosis and shorten the diagnosis time.