論文ID: ISIJINT-2024-169
Under certain conditions, the whole roll flatness meter outputs a bimodal waveform signal, which is clearly different from the conventional unimodal waveform signal. Since detection relies on the extraction of crest values and the values of the two waveforms do not have the same linearity, the presence of the two waveforms of different channels will clearly give rise to errors in the calculated flatness distribution. To develop an effective extraction method, it is necessary to accurately analyze the evolution of the waveforms. In this paper, the finite element method is used to calculate the load of the sensor, the stress distribution of each analysis surface and the deformation of the sensor mounting hole during the real-time detection to analyze the mechanism of the waveforms. The results show that unimodal and bimodal waveforms are produced under different strip tension and wrap angle conditions. In addition, the radial stress of the roll surface always presents two stress wave distributions. With increasing strip tension or wrap angle, the phase difference between the two waves increases. The stress distribution will change the deformation trend of the mounting hole and affect the stress distribution state of the sensor. When the phase difference of the stress waves exceeds the covering range of the sensor, the output signal changes from a unimodal waveform to a bimodal waveform. Finally, by setting up an experimental platform with variable tension and wrap angle, the relationship between the output waveforms and the working conditions in the simulation is reproduced.