Research On State Identification Of Diamond Roller Radial Runout Dressing Based On Acoustic Emission Monitoring

High precision diamond roller is very important for precision machining in highend manufacturing field.In order to ensure the performance of diamond roller,the accuracy of radial runout should be ensured during grinding and dressing diamond roller.In the current production process of diamond roller,it is necessary to rely on the experience of workers to judge whether the radial circular runout of the roller meets the processing requirements,and then carry out manual detection.This not only greatly reduces the processing efficiency,but also can not effectively collect the data in the processing process,and can not realize the automatic process data collection and promote the iteration and optimization of the processing technology.In order to realize online detection of radial runout of diamond roller and lay a foundation for wheel processing and automation and intelligence,this paper carried out a study on state identification of radial circular runout dressing of diamond roller(hereinafter referred to as roller dressing)based on acoustic emission monitoring.The main work of this paper is as follows:(1)The dressing process of diamond roller is analyzed.The error of radial circular runout of diamond roller is analyzed,the dressing position of diamond roller is defined,and the dressing principle of diamond roller is expounded.The grinding mechanism of diamond roller was studied.(2)Common grinding state monitoring signals such as grinding force,spindle current,acoustic emission and acoustic signals in audible domain were compared and analyzed,and acoustic emission signals were selected to monitor the roller dressing state.Then the identification scheme of roller dressing state is determined.Then the source and transmission mechanism of acoustic emission signal in roller dressing process are analyzed.Finally,the analysis method of acoustic emission signal is described.(3)The acoustic emission testing experimental platform for diamond roller dressing was built,and the roller dressing states were divided and the acoustic emission signals under different dressing states were collected.Acoustic emission signal is analyzed in time and frequency domain.Combined with the characteristics of acoustic emission signal,wavelet transform is selected to process the signal.The wavelet basis function suitable for decomposition is selected,the scale of wavelet decomposition is determined,and the acoustic emission signal is decomposed by wavelet.Finally,five characteristic values,such as effective value,energy spectrum coefficient and margin factor,are extracted,which can accurately represent the three dressing states.(4)According to the extracted characteristic values,a data set is established,and the data set is input into the probabilistic neural network model optimized by the Sparrow search algorithm for training,and a model that can accurately identify the roller dressing state is constructed.Then,the established model is verified to prove that the model can accurately and effectively identify the dressing state of diamond roller,and realize the online detection of radial circular runout of diamond roller,which lays a foundation for the automation and intelligence of diamond roller processing.