Research On Removal Model And On-line Monitering Of Fluid Magnetic Abrasive Hole Finishing

The surface roughness,burr,and other structural defects left by the machined parts after machining have an important impact on their performance,even on the overall equipment performance,stability,reliability,and life.As a common feature of mechanical parts,it is very important for the surface finishing of small holes in order to improve the surface quality.Supported by fund projects such as National Natural Science Foundation of China(51075294),Natural Science Foundation of Shanxi Province(201601D011060),Key Laboratory of Precision and Special Processing Ministry of Education(JMTZ201603),etc.,the fluid magnetic abrasive hole finishing technology has achieved a rapid development,and it has unique advantages in the finishing of the complex hole surface.The difficulty of surface testing after processing and the selection and control of the process parameters mainly depend on the current situation of empirical judgment.This paper has studied the following:(1)Based on the related theories of rheological properties,magnetohydrodynamics,slurry mechanics,tribology,and mechanical processing of fluid magnetic abrasive,the flow characteristics and material removalmechanism of the abrasive tool in the workpiece are analyzed.Then a corresponding mathematical model is established: Slip model and material removal rate model.(2)In view of the problems existing in the previous equipment,the structure and principle analysis are carried out.A peristaltic double loop equipment is designed,and the problems of non wear resistance,fluid leakage and pressure attenuation in the previous generation of equipment are solved.The effects of various components of the fluid magnetic abrasive is analyzed.Different grades of abrasive tools are prepared.The important role of xanthan gum in improving the stability of fluid magnetic abrasives is verified.Theoretical model verification experiments,machining tests,and online monitoring tests are designed.(3)The effects of processing time,magnetic field strength,inlet pressure,and the material on the surface of the hole after machining are obtained through the analysis of the test results.(4)Based on the artificial neural network technology,a BP network prediction model for the roughness of fluid magnetic abrasive with is constructed.The structure of the network is 5-11-1.On the basis of the experiment,a training sample and a test sample for the prediction model are constructed,and the prediction network model is trained and tested.(5)The feasibility of on-line acoustic emission monitoring of the machining process is verified by analyzing the acoustic emission signalwaveforms,energy RMS values,and ringing counts of the machine under conditions of no-load and addition of different abrasive components to the base fluid.The processing test is performed by changing the parameters,and the characteristics of the acoustic emission signals of the processing are compared and analyzed.Finally the correlation between the acoustic emission signal characteristics and process parameters is derived.