为提高铸机故障诊断能力,设计利用BP神经网络实现故障与特征的非线性映射方法。在该映射关系中,分析了BP网络故障诊断思路,并以搬运机械手为例,通过BP神经网络对其网络控制、液压驱动、回转及行走配合过程中的故障进行诊断。通过分析表明,BP网络能解决复杂装备诊断困难等问题。 In order to improve the fault diagnosis ability of caster, a nonlinear mapping method of fault and characteristic is designed by BP neural network. In this mapping relationship, the idea of ​​BP network fault diagnosis is analyzed. Taking the handling robot as an example, the BP neural network is used to diagnose the faults in the process of network control, hydraulic drive, slewing and walking. The analysis shows that BP network can solve the problems of complex equipment diagnosis.