Analysis Of Vibration Failure Causes And Recognition Methods Of Main Reducer Of Three Gorges Ship Lift

During the operation of the main reducer of the Three Gorges Ship Lifter,due to its complicated operating conditions,some potential gear failures,such as tooth surface wear,pitting,and tooth root cracks,will increase the vibration response of the reducer,reduce the life of the reducer and affect Normal operation of the shiplift.At present,the main reducer only has regular oil detection and no real-time monitoring system.Therefore,carrying out research on the impact of various fault factors on the vibration of the main reducer of the Three Gorges ship lifter and subsequent failure identification research will help to troubleshoot the main reducer and optimize the operation and maintenance management of the equipment.This paper focuses on the three kinds of faults,such as tooth surface wear,pitting and root cracks,which are common in transmission gears.Based on the consideration of internal excitation changes,this paper studies and analyzes the influence of such fault factors on the vibration of the main reducer.Based on the vibration signal,a fault pattern recognition platform is built for future fault diagnosis of the main reducer.The main research work of the paper is as follows:(1)The influence of tooth surface wear on the internal excitation and vibration characteristics of the gear is analyzed.Based on the related principles of tooth surface wear and the calculation method of the amount of wear,a reasonable uniform tooth surface wear amount of the main reducer of the Three Gorges ship lift is obtained.The original model of the ship’s main reducer was established,and four different models of tooth surface wear were established based on this model.The effects of different wear levels on the internal excitations of gears such as time-varying meshing stiffness,transmission error,and meshing impact force were studied,and the effects of tooth surface wear on the vibration response of the reducer were calculated and analyzed.(2)The effects of tooth surface pitting on the internal excitation and vibration characteristics of the gear were studied.According to the related principles and distribution laws of tooth surface pitting,the positions and sizes of tooth surface pitting that may occur in the reducer were predicted,and four types of tooth surface pitting models were established.The effects of different pitting degrees on the internal excitation of gears such as time-varying meshing stiffness,transmission error,and meshing impact force were studied,and the effects of tooth surface pitting on the vibration response of the reducer were calculated and analyzed.(3)The effects of tooth root cracks on the internal excitation and vibration characteristics of the gear were explored.Based on the related principles and expansion laws of tooth root cracks,the possible crack angles and depths of the main reducer were estimated,and four tooth root crack models with different depths were established.The effects of different crack depths on internal excitations such as time-varying meshing stiffness,transmission error,and meshing impact force of the gear were studied,and the effects of crack depths on the vibration response of the reducer were analyzed.(4)A convolutional neural network model suitable for fault identification of the main reducer is built.The neural network structure and basic model required for the main reducer fault diagnosis were determined.Taking the vibration frequency domain signals obtained under different faults as input,the input signal is repeatedly integrated and trained,considering the influence of different network parameters on the diagnosis results,and the convolution network structure is optimized.The diagnostic accuracy is high,which can be used to monitor the reducer in real time Network model of operating status.