Study On Condition Monitoring And Fault Diagnosis Method Of Ship Shafting Based On Variational Mode Decomposition

In recent years,with the rise of smart ship concept and the development of internet and information technology,such as big data,cloud computing and Internet of Things,the intelligent ship has become the trend of global shipping.Currently,research on smart ship has been carried out worldwide.One of the main purposes of smart ships to analyze and evaluate the operation status and health status of machinery and equipment in marine engine room,and make operational decisions and maintenance plans for the marine mechanical equipment.Marine machinery and equipment mainly include marine power system,marine auxiliary engines,generators and so on.Among these machinery,the ship power system which is the heart of the ship undertakes the mission of energy transmission,power transmission and ship propulsion.Ship propulsion shafting system is an important component of ship propulsion system.The vibration of shaft obviously increases under the deformation of hull and the external loads of the wind and waves,and the failure of the shaft will be generated.Therefore,it is necessary to focus on the reliability and safety of the ship shafting operation.The condition monitoring and fault diagnosis of the ship propulsion shafting is an effective method to ensure the reliability of the marine power plant and security of the ship navigation.How to identify and diagnose various abnormal conditions or faults of the ship shafting system effectively,timely and accurately,is the key to the study of ship shafting condition monitoring and fault diagnosis,which plays an important role on improving the reliability and effectiveness of ship shafting operation.This dissertation takes the ship propulsion shafting as the research project.The vibration characteristics of shafting under the hull deformation and external load excitation are gained.On this based,a intelligent method of fault feature extraction and intelligent diagnosis of ship shaft vibration signal was established.This intelligent diagnosis method not only provide a reasonable basis for ship shaft fault diagnosis and development of smart ships,but also provide methods and technical supports for promoting the optimal design,installation,performance monitoring and maintenance of ship shafting.The main research contents and results are as follows.1)The transverse vibration model of the hull under the deformation excitation was established and the influence of different factors on the vibration characteristics of the ship propulsion shaft was studied.By numerical analysis,the influences of shaft position,propeller,host,support stiffness,excitation position and different scales on the lateral vibration of the hull under the deformation excitation were analyzed.Taking 8530 TEU container as the research object,a real ship experimental study was carried out to verify the analysis results in some degree.2)A multi-excitation test for the vibration characteristics of the ship shafting is carried out.The influence of different excitation on the vibration characteristics of the hull-shaft is analyzed.Taking ship proportion shaft test rig as the research object,the influence of hull excitation and propeller excitation on hull-shaft vibration is discussed.Hull-shaft vibration characteristics under forced displacement,sinusoidal excitation force and frequency conversion excitation force were analyzed.3)Variational Mode Decomposition(VMD)was proposed to extract the fault feature of ship shaft.Based on the study of the mechanism of typical shaft faults(unbalance,misalignment,friction and oil film whirl),a simulation experiment of shaft faults was launched.The VMD method was used to extract the fault features of shaft faults.The results show that VMD method can extract the fault feature of the ship shaft system effectively.To reduce the difficulty on VMD parameter selection,a parameter selection method is proposed,which can effectively decompose the fault signal without experience and prediction.The method to extract fault features of shaft system with VMD-energy entropy was proposed and a great deal of shaft fault feature data were obtained,which provides data support for the subsequent fault diagnosis research.4)In order to improve the efficiency and accuracy of fault identification and diagnosis of ship shafting system,a fault diagnosis method of ship shafting system based on deep learning is proposed.The method of deep belief network is applied to the fault diagnosis of shaft system.Compared with the SVM method,the deep belief network can obtain higher recognition and diagnosis rate with fewer fault features,enhances the intelligent of the shaft fault identification and diagnosis.In conclusion,according to the vibration characteristics of the ship shafting and the problems existing in the feature extraction and fault diagnosis of vibration signal of the ship shafting,a combination method of theoretical analysis and experimental research are carried out,a fault diagnosis method of ship shafting based on variational mode decomposition and deep belief network was proposed.The proposed method enhance the intelligent fault identification and diagnosis of ship shafting with fewer fault features,and achieve higher recognition and diagnosis rate.