Construction of Distributed Sensing System for Peristaltic Continuous Mixing Conveyor Imitating Intestines: Internal State Measurement Using Accelerometer

Continuous mixing and conveying technologies for processing solid-liquid mixtures and highly viscous fluids are required in the manufacture of food and medicine. A device featuring such a technology, namely the peristaltic continuous mixing conveyor, has been developed by the authors previously. This device, imitating the movement of intestines, is capable of mixing and conveying content by inflating a rubber tube using air pressure. Herein, we aim to estimate the degree of mixing of the contents via machine learning using data from pneumatic factor sensors installed in this device. However, the low measurement accuracy of these sensors when dealing with compressible fluids makes the mixing degree estimation difficult. Therefore, in this study, to improve the measurement accuracy of the sensing system, we installed an accelerometer that could directly measure the behavior of the rubber tube inside the device without any contact with the contents being required. Upon operation of the device with various objects within it, the sensor angle of the rubber tube was obtained from the accelerometer values for each input. The sensor angle at steady state after air supply was 22.5° greater for the rigid bodies than for the mixtures of powder and liquid. For the powder-liquid mixtures, a linear relationship was demonstrated between the percent concentration of mass and the sensor angle at a steady state. The results suggest that the content of the device can be determined with greater accuracy by using the attitude angle obtained from the accelerometer than from the conventional pneumatic factor sensors.