Research On Health Condition Monitoring System Of Electric Monorail Crane

Electric monorail crane is an important implement in the underground auxiliary transportation system of coal mine.It is mainly used to transport production materials,personnel and equipment,accomplish lifting and hoisting of underground equipment as well.During the working of electric monorail crane,battery pack and driving motor have been always in a high load operation,which is prone to engender various health problems,bringing extreme hidden dangers to the security production of coal mine.Under the above present situation,this thesis designs a health monitoring system of electric monorail crane integrating functions of online monitoring and fault prediction,which combined with embedded technology,Internet of things technology,cloud computing and edge computing,artificial intelligence algorithm.Since electric monorail crane hold in moving state during its operation,monitoring data cannot be transmitted through wired networks.Conventional wireless communication modes,such as WiFi,have limited data transmission distance as a result of various disturbances in the underground environment of coal mine.In addition,although the communication distance of Low-Power Wide Area Network(LPWAN)is long,it is difficult for LPWAN to transmit audio and other massive data information.Therefore,deep neural network is deployed on the embedded platform to analyze audio data intelligently.Analysis results then sent to the monitoring center through LPWAN,which avoids network congestion and delay caused by massive data information transmission,thereby improves monitoring efficiency of the designed system.Drawbacks generally remain in the current fault diagnosis methods of electric monorail crane,such as the difficulties existing in the installation of corresponding sensors,the complexity of signals' acquisition,low accuracy and so forth.In view of the above problems,a fault diagnosis method of electric monorail crane motor based on audio deep temporal features is proposed.This method take advantage of MFCC to preprocess the collected audio data and obtain 13 dimensional features of it Moreover,considering the temporal characteristics of audio signals,GRU and MLP are utilized to construct a deep neural network to process and classify audio features and realize motor fault prognosis.On this basis,audio data set of motor fault is built up to train and verify the proposed model.Functions of the designed system are experimental analyzed and tested on the spot at the ending of this dissertation.Results demonstrate that the system is well able to realize online monitoring of electric monorail crane battery pack and driving motor data,which achieves the expected design goal.The model proposed can precisely identify four typical faults of electric monorail crane driving motor,which provides an effective solution for the safe and reliable operation of electric monorail crane and has certain practical application value.Figure[32]Table[14]Reference[83]...