Abstract
We developed behavior modeling method of bill in non-pinched belt conveyors at high speed. Non-pinching belt conveyors are commonly used in bill sensing system since flapping of bills degrades the performance of sensing. However, predicting the complex behaviors of bills in these conveyors at high speed is especially difficult because of the thin thickness and low stiffness. To improve the prediction accuracy, here we introduce probability model and Bayesian inference into analysis model. The proposed model is evaluated by comparing with experimental data obtained with laser displacement meter, and the prediction accuracy of the model is 98.4%.
