Research On Deflection Change Monitoring Of Roller And Fault Diagnosis Of Rotary Kiln

Rotary kiln is widely used in building materials,metallurgy and other industries,which mainly consists of cylinder,supporting parts and transmission parts.It is the core equipment of cement plant.Its long-term operation in the harsh environment of high temperature and heavy load will inevitably lead to thermal bending deformation of cylinder and center line deviation.If the early failure of rotary kiln can not be found in time and allowed to develop,it will lead to further failure until the kiln is shut down,bringing huge economic losses to enterprises.At present,domestic cement plants lack the theory and technology of early fault diagnosis of rotary kiln,and remain in the post-test stage,which is not conducive to the maintenance of equipment.For this reason,this paper takes the supporting roller as the research object,studies how to monitor the fault state of rotary kiln in real time,and designs an on-line monitoring system on this basis.The main research contents are as follows:(1)Through the force analysis of the supporting roller,the influence of the bending deformation of the cylinder and the deviation of the center line on the deflection change of the supporting roller shaft is discussed.It is pointed out that KS(Kiln Shell)harmonics and KR(Kiln Roller)harmonics are included in the deflection change signal of the supporting roller shaft.The amplitudes of these two waveforms can reflect the two kinds of faults mentioned above.Through FFT analysis of the deflection change signal of the supporting roller shaft collected on the spot,it is found that there are two waveforms,which verify the correctness of the theoretical analysis and feasibility of reflecting the running state of rotary kiln by monitoring supporting roller.(2)In order to put forward the fault information of rotary kiln accurately,the decomposition effect of the three methods is analyzed by comparing the simulation signals,and it is found that CEEMD method has the best decomposition effect.This method is used to extract fault features from the measured data of a rotary kiln.KS and KR harmonics are effectively separated.The decomposition results are compared with those obtained by other two methods,and the correctness of this method is verified.In order to further realize the intelligent recognition of fault modes,the BP neural network model is established by using the characteristic parameters obtained from signal analysis and processing as the feature input vector.In view of the shortcomings of BP algorithm,genetic algorithm is used to optimize the network.Through comparison and verification,it shows that GA-BP network has better convergence speed and recognition effect.It provides a theoretical basis for on-line fault diagnosis.(3)In order to applying theoretical research to practical engineering,a monitoring system software is developed to monitor the operation status of rotary kiln on-line and in real time,based on LabVIEW platform.The overall framework and design mode of the software are briefly introduced.The implementation process of each functional module is explained in detail.The system test is carried out to verify the correctness and reliability of the software.The monitoring system provides a basic technical scheme for the operation condition monitoring of rotary kiln,and has theoretical research and practical engineering significance.