摘要
针对BP神经网络在提升机制动系统故障诊断中存在收敛速度慢、诊断精度低和鲁棒性较差等缺点,提出了一种基于粒子群神经网络的故障诊断方法。建立了以提升机制动系统的故障特征参数为输入,以制动系统的主要故障类型为输出的故障诊断模型;采用粒子群算法优化BP神经网络的参数,加快了神经网络的收敛速度。通过对提升机制动系统典型故障的诊断研究表明,该诊断方法改善了提升机制动系统故障诊断的精度和速度。
For the limitations of BP neural network in the hoist braking system fault diagnosis, such as slow convergence rate, low diagnosis accuracy and poor reliability. A neural network fault diagnosis method based on particle swarm optimization algorithm is proposed. The hoist braking system fault diagnosis model is established, which uses braking system fault characteristic parameters as input variables and adopts fault modes for output variables. Particle swarm optimization algorithm is applied to optimize the neural network parameters, and the optimized neural network is used for braking system fault diagnosis, which can increase the convergence rate of neural network. Research on the hoist typical faults diagnosis indicates that the diagnosis strategy improves the fault diagnosis precision and speed of the hoist braking system.
出处
《控制工程》
CSCD
北大核心
2016年第2期294-298,共5页
Control Engineering of China
基金
河南省科技计划项目(094300510015)
关键词
矿井提升机
故障诊断
神经网络
粒子群算法
Mine hoist
fault diagnosis
neural network
particle swarm optimization algorithm
