Parameter Identification And Fault Diagnosis Of Marine Auxiliary Low Voltage Induction Motor

With the rapid development of the ship industry in China,the requirements of network,information and intelligence of ship management are becoming higher and higher.Because of its low price and high reliability,especially the low-voltage induction motor is the main auxiliary equipment of the ship,its operation state is directly related to the performance of the local equipment of the ship.It is necessary to monitor and diagnose the status of the auxiliary low-voltage induction motor.It can not only eliminate the potential safety risks of the induction motor,but also improve the performance of the local equipment in the cabin,It can ensure the safety of the ship in navigation.According to the monitored status information,the auxiliary low voltage induction motor is evaluated and predicted to achieve fault warning,which improves the management level of ship equipment and the reliability of the ship in operation.(1)This thesis has done the following work for the parameter identification and fault diagnosis of low voltage induction motor of ship auxiliary machine: firstly,according to the research of the identification method of the induction motor,this thesis establishes the identification model suitable for the least square method in the two-phase static coordinate system,and then compares with the traditional off-line identification method of induction motor parameters,The recursive least square method can be used to identify the stator resistance,rotor resistance,stator inductance and mutual inductance of induction motor at one time before the motor is running.Finally,the simulation results show that the method used in this thesis is more accurate and more efficient than the traditional method of off-line identification of induction motor parameters.(2)In terms of induction motor fault diagnosis,this paper first describes the characteristics and mechanism of induction motor fault and the current methods of fault diagnosis of induction motor,analyzes the causes of the fault,and obtains several common failure frequencies.Secondly,wavelet packet transform is used to eliminate noise from the original data of the bearing fault test of the induction motor in the West storage University of America,Then,the feature vector of feature band is extracted,and the conclusion of wavelet packet transform is good local time frequency and multi-resolution processing ability.(3)According to the large amount of data of induction motor vibration measured in the company,the eigenvector of the feature band in the event of fault is extracted by wavelet packet transform as the input sample of neural network for training.When the precision meets our needs,the fault diagnosis system of the input signal is established by using the improved BP neural network.After many experiments,the experiment proves the effectiveness of the method,which makes the fault diagnosis of induction motor more intelligent by wavelet neural network.