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。 Aiming at the shortcomings of the BP neural network in the fault diagnosis of the hoist braking system, such as slow convergence rate, low diagnostic accuracy and poor robustness, a fault diagnosis method based on the PSO neural network is proposed. The fault diagnosis model based on the fault characteristic parameters of the brake system of the elevator is established and the main fault type of the braking system is output. The particle swarm optimization algorithm is used to optimize the parameters of the BP neural network to speed up the convergence of the neural network. The research on typical fault diagnosis of hoist brake system shows that this method improves the accuracy and speed of fault diagnosis of hoist brake system.