Research On Fault Test And Diagnosis Method Of Harmonic Gear Reducer

As a new type of transmission structure,harmonic reducer has the characteristics of high precision,large transmission ratio,small volume,light weight,smooth transmission and high transmission efficiency.It has a wide application prospect in industrial robots and aerospace manipulators.It is an important component to ensure the normal action of industrial robots and the on-orbit service of spacecraft.Once the failure occurs,it will cause inestimable losses.Therefore,it is of great significance to fully study the working characteristics and typical fault modes of harmonic reducer and to monitor and diagnose the fault.In this paper,the typical fault diagnosis methods of harmonic reducer are studied by means of theoretical deduction,mechanical simulation analysis and simulation test.First of all,the research status of harmonic reducer and related fault diagnosis methods are reviewed and discussed in detail.It is found that the reason why it is difficult to make a breakthrough in the study of reducer fault diagnosis is that it is difficult to build the overall dynamics simulation modeling.The working ideas of reducer fault diagnosis mainly through simulation tests are clarified in this paper.Then,the main structure and working principle of the harmonic reducer are defined,and the tooth profile design,wave generator contour design and gear tooth meshing characteristics are analyzed.By constructing the mathematical model of the movement of the flexible bearing,the theoretical derivation of the fault characteristic frequency of the flexible bearing and the rigid wheel and the flexible wheel meshing is carried out.The failure mechanism of the typical failure modes of harmonic reducer in industrial practice is analyzed,and the failure modes and causes are explained with the finite element simulation results of flexwheel assembly process.Secondly,the simulation test platform of harmonic reducer is designed by Solid Works 3D entity software.Based on the test purpose and functional requirements,the models and specifications of various equipment and parts on the test platform are selected and the construction of the test platform is completed.The sensor layout and test scheme were designed,and the original signals were obtained by simulating the typical fault modes of harmonic reducer,such as the friction of rigid wheel teeth,the dislocation of wave generator and its compound faults.Finally,respectively,to obtain all kinds of model corresponding to the original signal in time domain,frequency spectrum in time and frequency analysis,made clear the fault signal characteristic,to build the relationship in failure mode and failure characteristics,extract the fault feature,through the fault diagnosis method based on support vector machine(SVM)to the original fault signal in the fault pattern recognition,The results show that SVM is effective in the diagnosis of typical faults of harmonic reducer.