Study on the Detection Model of Cigarettes' Dense-end Position Based on Density Probability Distribution

The density distribution of cigarettes was analyzed in order to improve the detection precision of cigarettes' dense-end position. The distribution pattern of the density under the ideal production condition was combined, and a dense-end position detection model was constructed. The model parameters were determined on the basis of data optimization. The model was validated by MW-T and cigarettes. We got conclusions as below. 1) For the measurement of dense-end with the Dense-end Position Standard Parts, the average error of the measurement was 0.1646 mm and the maximum error was -0.2735 mm. 2) The average error between the measured value of the dense-end position with ordinary cigarettes and that of the MW4420 was 0.79 mm, which meets the actual production demand (≤1 mm). The model has high accuracy and good reliability, and can provide a favorable technical support for the dense-end position detection of cigarettes. It is helpful to improve the level of quality control in cigarette production, and inject the new vigor for the sustainable development of cigarette industry.