In-Line Measurement of Particle Weight of Pulses Such as Chickpea by Piezoelectric Sensor

Abstract

Piezoelectric-based sensors are used as well as optical sensors for homogeneous sowing of seeds in the fields and for calculating the grain loss rate in harvesting and cleaning with harvesters in agricultural production. In this study, a piezoelectric-based sensor design has been developed that can measure properties such as size and weight during the storage of grains in silos in addition to these applications. Two different impact sensors using piezoelectric disks and piezoelectric ceramic elements were designed. Impact experiments were performed with chickpea grains in three different weight and size classes, and the energies of the electrical signals generated by the piezo elements because of the impact were calculated. This energy data was used in the training of ten various classification algorithms and their prediction success was calculated. While the highest classification accuracy for sensor data using piezo disk elements was obtained with the neural network algorithm with 92%, the highest classification accuracy for sensor data using piezo ceramic elements was obtained with the neural network and adaptive boosting algorithms with 68%. These data show that the sensor design using piezo disk elements could be used to measure the weights of grains such as chickpea during storage in silos.

Keywords

• Classification algorithms
• ceramic
• disc
• particle weight
• piezoelectric