Level Measurement in Grain Silos with Extreme Learning Machine Algorithm
Hüseyin Duysak, Enes Yi̇ği̇t · 2019
In this study, a new method based on a machine learning technique to detect the amount of grain in silos with radar is proposed. In order to carry out the measurements in the laboratory environment, a model silo corresponding to 8% of the actual commercial silo was produced by a 3D printer and the radar measurements were performed in the frequency band of 18 - 40 GHz. In the proposed method, 2000 back scattering information belonging to different levels of grain were collected and they were used as input data to the extreme learning machine (ELM). The accuracy of the algorithm was obtained by K-fold cross validation technique. As a result, the ELM algorithm was determined the grain amount with an accuracy rate of 84%. The results show that the measurement system based on machine learning is more practical than traditional methods.