Prediction of Oil Content in Fresh Palm Fruit based on an Ultrasonic Technique

Authors

  • Sutthawee Suwannarat Intelligent Systems Laboratory, Department of Computer Engineering, Prince of Songkla University, Songkhla 90110, Thailand.
  • Thanate Khaorapapong Intelligent Systems Laboratory, Department of Computer Engineering, Prince of Songkla University, Songkhla 90110, Thailand.
  • Mitchai Chongcheawchamnan Intelligent Systems Laboratory, Department of Computer Engineering, Prince of Songkla University, Songkhla 90110, Thailand.

Keywords:

ultrasonic testing, oil content, palm fruit, polynomial regression, neural network

Abstract

An ultrasonic technique was proposed to predict the oil content in fresh palm fruit by measuring the attenuation based on the ultrasonic transmission mode. Several palm fruit samples with known oil content determined by Soxhlet extraction (ISO9001:2008) were tested using ultrasonic measurement. Amplitude attenuation data results for all palm samples were collected. The Feedforward Neural Network (FFN) technique was proposed to apply to predict the oil content of the samples. The root mean square error and mean absolute error of the FNN model for predicting the oil content percentage were 5.8672 and 3.4731, respectively, with a correlation coeffi cient of 0.8891.

Downloads

Published

2012-04-30

How to Cite

Suwannarat, Sutthawee, Thanate Khaorapapong, and Mitchai Chongcheawchamnan. 2012. “Prediction of Oil Content in Fresh Palm Fruit Based on an Ultrasonic Technique”. Agriculture and Natural Resources 46 (2). Bangkok, Thailand:318-24. https://li01.tci-thaijo.org/index.php/anres/article/view/242817.

Issue

Vol. 46 No. 2 (2012): March-April

Section

Research Article

License

online 2452-316X print 2468-1458/Copyright © 2022. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/),
production and hosting by Kasetsart University of Research and Development Institute on behalf of Kasetsart University.