Primary study of temperature distribution measurement in stored grain based on acoustic tomography

Abstract

A common and traditional method for detecting stored grain deterioration is the measurement of grain temperature. This study proposes a temperature monitoring method for stored grain based on acoustic tomography. A model describing the relationship between grain temperature and sound travel time in stored grain is established. A speed conversion coefficient is defined and used to convert the measured velocity of sound in the stored grain into the sound velocity in free space, which has the same gas components and temperature as those of the stored grain. The influence factors of the speed conversion coefficient are analyzed, and the calibration method of the coefficient is introduced. The time-delay estimation method based on triple correlation and wavelet de-noising and the temperature reconstruction algorithm based on Markov radial basis function and singular value decomposition are used to measure the temperature distribution of the grain bulk. Two experimental systems are built to verify the acoustic model and reconstruct the temperature distribution of grain. The experimental result shows that the acoustic model can be used for temperature measurement in soybeans. The temperature distributions with different hot spot locations in soybeans are all well reconstructed, and the coordinates of the hot spots are reconstructed with better accuracy than that in thermocouple measurement. Thus, acoustic temperature measurement is useful in the early prediction of temperature anomalies in stored grain.