Study Of Vibration Spectrum Characteristics On Different Internal Qualities Of Korla Pear

Nondestructive detection of internal qualities of Korla pear has been paid more attention in the process of sorting. In this work, the focus on rough peel pear, geographical difference and internal defect detection of Korla pear based on vibration spectrum in order to realize the nondestructive detection of the internal qualities. Firstly, the stability of testing system, suitable excitation conditions and the effects of mass, fruit shape and firmness of pear to testing were determined by means of vibration frequency excitation test. Secondly, The vibration spectrum characteristics of normal pear, rough peel pear from Korla pear and Aksu pear were tested by pulse excitation. The differences of vibration spectrum characteristics between the normal and the rough peel ones, between Korla pear(the normal) and Aksu pear were analyzed. The microstructures of peel and flesh tissue of three kinds of pear were observed. Finally, uneven development of internal defective tissue on vibration spectrum response were discussed. Furthermore, the relationship between development degree of internal defective tissue and the response of vibration spectrum was made clearly. Finally detection threshold with vibration spectrum method was determined. The main contributions are as follows:(1) The exponential window function is suitable to truncation progressing of dynamic signal of pear because the processed signal has high SNR(signal to noise ratio) and less information leakage. The dynamic signals are stable and the responses of dynamic spectrum are consistent at different excitation points on the equatorial region and same detection point on the line of equator with single excitation and detection. The different excitation materials and forces have no significant effect on the response dominant frequency. When the rubber was used for force hammer, the dynamic signal is strong with the range of excitation force in 9-12 N and the difference is big on the amplitude of between the dominant response frequency and the second response frequency, thus the dynamic signal can be analyzed easily. The dominant response frequency rose monotonically as linear correlation with the firmness increased, downed monotonically as linear correlation with the mass and maximum height to diameter ratio of fruit decreased. Therefore, the dominant response frequency can be the important index to evaluate the firmness, meanwhile the mass and fruit shape must be considered in measuring fruit firmness with the dominant response frequency.(2) The response dominant frequency of the rough peel and the normal pear are 687.5±13.1 Hz and 775±19.4 Hz, respectively. The response dominant frequency of the Korla pear and Aksu pear are 740.5±27.3 Hz and 775±16.1 Hz, respectively. And there is significant difference of dominant response frequency. Therefore the dominant resonant frequency of vibration spectrum could possibly be used to for the nondestructive recognition of rough peel pear and the identification of different producing area samples. Main reasons of the rough peel pear with the biggest response dominant frequency and Aksu pear with the smallest response dominant are the significant of the variety of fruit firmness. And the cell turger was the essential reason of changing of fruit firmness.(3) There was no effect on the dominant response frequency of different distances from defect tissue border to the excitation points on the peel of pear. The dominant response frequency of the defect pear decreased as exponential order while the dominant response frequency of the sound pear linearly deceased. The relationship between the dominant response frequency(fd) and the percentage of internal defect tissue mass(ω) of pear can be characterized by the regression equation fd=410.649e-0.0833ω+261.947. The detection threshold of internal defect is 1.807-2.281%. Consequently, the discrimination rates of detection using vibration spectrum technique are 12.121%, 18.75%, 91.667% when the percentages of internal defect tissue mass of pear are <1.807%, 1.807-2.281%, >2.281%, respectively. These results can provide guidance for choosing suitable detection conditions to increase the detection accuracy of internal defect of Korla pear.