水果模擬振動平台最適化隨機控制曲線求取之探討

Abstract

水果於採收後因為人為或機械因素，會造成損傷導致品質降低，其中部分外力是導因於運輸時的振動。因此以振動平台進行水果模擬實運振動試驗，陸續被學者們探討分析。惟前人研究中，對於將試驗所得的功率頻譜密度(power spectral Density)曲線轉換為振動模擬可用之隨機控制曲線(random profile)的方法，乃以概略性的描繪方式找出數段直線，作控制器所用之隨機控制曲線，並無特別詳述其求取步驟或演算法則。本研究透過田間試驗量測振動資料進行分析，以平均法計算，將各小區的功率頻譜密度加以平均，求得1 Hz ~ 100 Hz頻率範圍內的平均功率頻譜密度。將此實際輸振動資料，藉由MATLAB程式發展的多段式變動樣本數二次多項式迴歸技術，計算出應用於振動控制器之隨機振動控制曲線，進行水果模擬振動平台模擬試驗。結果由頻譜圖的比對，所有的田間試驗振動之功率頻譜密度曲線形狀以及峰值發生之頻率值皆與振動平台模擬振動之量測資料相符；惟位於較高頻率範圍內之功率頻譜密度模擬之峰值稍有不足。由本研究發展之技術將可找出模擬運過程之適用隨機控制曲線，並發展山將田間振動資料轉換為實驗室進行運振動模擬所須參數的可行方法，以便應用於未來的水果模擬振動與損傷之試驗分析。 Fruit after harvest result in considerable degradations in quality. Damages are generally caused by manhandling or mechanical operations. One of these causes is vibrations endured during transportation. Therefore, the topic of how to simulate the real Situation of fruits during transportation in the lab by the shaker has been explored by researchers over the last. few decades. However, the method to convert the experimental power spectral density (PSD) curves into random profiles used by the shaker's controller is either unclear or not mentioned. One method to get the random profile is by tracing the envelope of the spectrum on a transparency subjectively. The objective of this study is to investigate the proper techniques of developing the random profiles. The average PSD from the accelerometer mounted on the chassis of a small fam transporter is calculated by average method, in which the frequency domain from I Hz to 100 Hz is of major interest„ A "multi-segments 2-degree polynomial regression with increasing sampling variability" technique is developed and carried out by MATLAB program in the study. Comparing the filed data and the results of simulation, it shows that satisfactory coincidence exists in the shapes of the average PSDs and the frequency of peak signals. However, the peak values in the higher frequency range show some difference. The method can find the optimum random profile for the electro-dynamic shaker to simulate the vibration of fruit during transportation. Furthermore, the achievement can be applied to studies of fruits' Injury during transportation.