Research On On-line Prediction And Diagnosis Method For Broken Shaft Fault Of Rotating System

With the development of society and the progress of science and technology,there are more and more applications of the automatic rotating system,however,its failure reduces the efficiency of automation equipment applications.Especially the fault of the rotating shaft will further expand the fault degree of the equipment.Rotating mechanical system is one of the most commonly used transmission system,widely used in various fileds,such as chemical,power,petroleum,aerospace,etc.Rotating machine shaft failure can cause huge economic losses,or even casualties.Therefore,strengthening the monitoring of the operating state of the rotating machinery system,especially the prediction of the shaft fracture,is conductive to production safety,personal safety,reduction of the economic losses and national economy improving.Although there are a lot of ways to detect the fault of rotating shaft,such as offline and online,these methods have their own defects,bring great difficulties to judge the running state of the rotating shaft.In China,study on rotating machinery shaft breaking predictions are not mature enough,although there are many papers on the study of broken shafts,generally limited to the theoretical analysis of the causes of broken shafts.So the actual research can be applied to the field of broken shafts test is very few,we lack a simple and effective online prediction system,which can deal with the shaft fracture.This paper aim at the insufficient of existing online detection method,based on the mechanical shaft is an elastic body,and there is a certain relationship between the strain generated by the torque transmission process and the torsion stiffness,proposed a new method for online prediction of rotating shaft fault of rotating machinery,and completed the preliminary research and testing of the whole system.This paper firstly introduces the fault condition of the rotating machinery and the harm of the broken shaft,analyzes the existing fault detection methods and their shortcomings,summarizes the demand and development of the fault diagnosis.Secondly,the change of stiffness and dynamics in the running state of the rotating shaft are analyzed from the perspective of the axial stress,and describe the feasibility of the new online prediction method.Thirdly,analyzed the principle of the new type of broken shaft on-line prediction system,and designed the structure of the whole prediction system,finally,the real-time detection of the fault axis of the actual system is carried out by using the detection system,and the effectiveness of the proposed method is proved.