欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对电机定子绕组匝间短路故障，提出了基于递归小波神经网络（RWNN）的故障在线诊断方法。该方法采用两个RWNN监测匝间故障，一个用于估算故障的严重度，另一个用于确定故障匝数。针对RWNN的训练，研究了Levenberg—Marquardt（LM）学习算法，以减少训练中的计算量，确保网络模型的快速收敛。根据此方法，设计了试验系统，试验结果表明，基于RWNN的诊断模型可精确确定短路故障匝数，与前馈神经网络（FFNN）相比，能更有效地监测匝间短路故障的缓慢变化情况。
英文摘要：
A recurrent wavelet neural network（RWNN}-based online stator winding turn fault detection method for induction motors was presented, in which two RWNNs were employed to detect turn fault, one was used to estimate the fault severity, the other was used to determine the exact number oI fault turns： In the course of training, Levenberg - Marquardt （LM} algorithm Was introduced to make the RWNN converging more quickly. Experiments were carried out on a special rewound laboratory induction motor. The results show that the RWNN-based diagnosis model determines the shorted turns exactly, and is more effective than the feed-forward neural network { FFNN}-based detection model under the condition of detecting a slowly developing turn fault.
同期刊论文项目
同项目期刊论文
Simulation of grain boundary effect on characteristics of ZnO thin film transistor by considering th
Four optimal design approaches of high-order finite-impulse response filters based on neural network
期刊信息
位置：