摘要
提出一种基于改进BP算法的异步电机故障诊断方法,不仅能够对电机的四种故障状态做分类识别,而且能够提高网络的收敛速度并避免其陷入局部极小。首先,根据故障特征向量与异步电机故障类别之间的映射关系,建立基于BP神经网络的故障诊断模型,然后利用故障样本对该模型进行训练与测试。仿真结果表明,该方法能有效地识别异步电机的四种故障类别,且具有较高的故障诊断准确率。
A fault diagnosis method of asynchronous motor based on improved BP algorithm is proposed in this paper.It can not only classify and identify the four fault states of the motor,but also improve the convergence speed of the network and avoid falling into local minimum.Firstly,according to the mapping relationship between fault feature vector and fault category of asynchronous motor,a fault diagnosis model based on BP neural network is established.Then,the model is trained and tested by using fault samples.The simulation results show that the method can effectively identify four fault categories of asynchronous motor,and has high fault diagnosis accuracy.
出处
《工业控制计算机》
2021年第2期71-72,76,共3页
Industrial Control Computer
基金
山西省高等学校科技创新项目(201804034)。
关键词
BP神经网络
异步电机
故障诊断模型
改进算法
BP neural network
asynchronous motor
fault diagnosis model
improved algorithm