Research On Fault Detection Method Of Wind Turbine And Design Of Remote Monitoring Device

In order to solve the problems of energy and environment,countries all over the world have focused on new energy generation technology,With the rapid development of wind powe,the problem of wind turbine fault is gradually prominent.Bearing is the key component of wind turbine,and has a high failure rate.Therefore,the fault diagnosis of wind turbine bearing has important research significance.This thesis first extracts the fault feature of the wind turbine bearing vibration signal.In view of the non-linear and non-stationary characteristics of vibration signals of wind turbine bearings,the fault features of wind turbine bearings are extracted by using wavelet packet decomposition,the number of decomposition layers of the wavelet packet is set to 3 layers.and the proportion of energy in eight subbands to the total energy is calculated.This energy ratio is used as the fault eigenvector of the wind turbine bearing.Secondly,the fault diagnosis analysis of wind turbine bearing based on neural network is completed.Fault diagnosis of wind turbine is based on the combination of BP network and PNN network.The BP network model and PNN network model are created,and then the parameters such as the number of hidden layer neurons,transfer function and spreadvalue are analyzed and determined to get the wind turbine bearing fault diagnosis neural network model.Then,the fault diagnosis method of wind turbine bearing is simulated and analyzed.The simulation results show that both BP and PNN neural networks achieve ideal results,but PNN neural network is better than BP neural network in terms of model design difficulty and total time-consuming of fault diagnosis.Finally,this thesis designs the wind turbine bearing fault monitoring system based on the Internet of things.The system uses CC2530 as the main control chip.The Zig Bee terminal node is installed on the wind turbine to monitor the vibration parameters of the wind turbine bearing in real time.The vibration parameters collected by each node is summarized and packaged on the coordinator and sent to the web server through the gateway module.The vibration data of the wind turbine can be displayed and stored in the upper computer of the monitoring center.The upper computer carries out fault diagnosis according to the vibration signal and displays the running state of the wind turbine The manager can query and manage the data on the upper computer...