KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 5 of 5

Some of the metrics are blocked by your
consent settings
Moisture content prediction in durian husk biomass via near infrared spectroscopy coupled with aquaphotomics and explainable machine learning
(2025-12-15)
Shrestha, Zenisha
;
Shrestha, Bijendra
;
Sirisomboon, Panmanas
;
Pun, Umed Kumar
;
Bajracharya, Tri Ratna
Accurate determination of moisture content is essential for energy efficiency and biomass management for fuel materials such as durian husk. Traditional methods of determining biomass moisture content are time-consuming and require specialized expertise, posing challenges for continuous monitoring. To address this limitation, this study applies Near Infrared Spectroscopy (NIRS) combined with machine learning models to rapidly and accurately assess moisture content. Both linear Partial Least Squares Regression (PLSR) and non-linear approaches were used, including Support Vector Machines (SVM), Artificial Neural Networks (ANN), and Extreme Gradient Boosting (XGB). The application of preprocessing techniques, notably the Savitzky-Golay second derivative (SD) and Standard Normal Variate (SNV), significantly augmented the predictive performance, highlighting the importance of data preprocessing in spectral analysis. Synthetic spectral augmentation using Gaussian noise revealed that while SVM and ANN exhibited near-perfect performance, SVM demonstrated quantifiable reliability. This study also demonstrates SVM as the most sensitive and reliable method for detecting and quantifying moisture content in durian husk. This research contributes novel insights to biomass analysis, highlighting the benefits of integrating NIRS and feasibility of explainable machine learning techniques to identify water related spectral parameters to advance aquaphotomics, thereby advancing rapid and accurate biomass characterization.
Some of the metrics are blocked by your
consent settings
In-line near infrared spectroscopy for the prediction of moisture content in the tapioca starch drying process
(2019-03-01)
Phetpan, Kittisak
;
Udompetaikul, Vasu
;
Sirisomboon, Panmanas
Moisture content is an important parameter measured in tapioca starch production as this parameter has been shown to correlate strongly with the quality of the finished product. However, there is currently no in-line sensor which can be used to directly measure the moisture content of the product in real time. The objective of the present work was to study the use of an in-line measurement which can be introduced at the end of the drying process for tapioca starch moisture content evaluation. Either in-line NIR data or at-line NIR data was used to develop the necessary calibration models for evaluating the moisture content. Furthermore, calibration models were also developed by pooling the in-line and at-line data. Its performance was then verified using additional in-line data. The NIR model developed using 100% of the at-line data and 50% of the in-line data was validated using the unused 50% of the inline data. This model was shown to provide better performance in moisture content prediction with an SEP of 0.61% and a bias of 0.001%. In addition, the results showed that the at-line spectrum can also be used for the calibration model development to predict the moisture content of the samples scanned by an in-line spectrometer. However, the in-line spectrometer installation on a pneumatic conveying circular tube where tapioca starch and air mixed was found to be complicated due to significant vibration. This caused additional variation in the data with time. Therefore, it is concluded that the most suitable place for installing a spectrometer would be at a position involving a low pressure, or where the stream flow of a product is steadier in order to avoid the dynamic mixing of the product within the drying tube affecting the uncertainty of NIR scattering during the measurement.
Some of the metrics are blocked by your
consent settings
Rapid non-destructive evaluation of moisture content and higher heating value of Leucaena leucocephala pellets using near infrared spectroscopy
(2016-07-15)
Posom, Jetsada
;
Shrestha, Amrit
;
Saechua, Wanphut
;
Sirisomboon, Panmanas
The MC (moisture content) and HHV (higher heating value) of Leucaena leucocephala pellets using NIR (near infrared) spectroscopy was investigated in this study. The MC of the pellets was adjusted by subjecting the samples to different relative humidity environments. The samples were scanned in diffuse reflection mode at wavenumbers of 12,500-4000 cm-1. Partial least squares regression models correlating the MC and HHV with the NIR spectra were developed and validated by full cross validation. The model for MC and HHV provided coefficients of determination (R2) of 0.995 and 0.964, a root mean square error of cross validation (RMSECV) of 0.187%wb and 79.2 J g-1, bias of -0.0008%wb and 1.29 J g-1 and a RPD (ratio of prediction to deviation) of 13.9 and 5.30, respectively. The models had excellent accuracy. This rapid quality evaluation method may be used for trading of biomass pellets. An equation related MC and HHV was also developed.
Some of the metrics are blocked by your
consent settings
Evaluation of the moisture content of tapioca starch using near-infrared spectroscopy
(2015-03-25)
Phetpan, Kittisak
;
Sirisomboon, Panmanas
The purpose of this study was to develop a calibration model to evaluate the moisture content of tapioca starch using the near-infrared (NIR) spectral data in conjunction with partial least square (PLS) regression. The prediction ability was assessed using a separate prediction data set. Three groups of tapioca starch samples were used in this study: tapioca starch cake, dried tapioca starch and combined tapioca starch. The optimum model obtained from the baseline-offset spectra of dried tapioca starch samples at the outlet of the factory drying process provided a coefficient of determination (R2), standard error of prediction (SEP), bias and residual prediction deviation (RPD) of 0.974, 0.16%, -0.092% and 7.4, respectively. The NIR spectroscopy protocol developed in this study could be a rapid method for evaluation of the moisture content of the tapioca starch in factory laboratories. It indicated the possibility of real-time online monitoring and control of the tapioca starch cake feeder in the drying process. In addition, it was determined that there was a stronger infl uence of the NIR absorption of both water and starch on the prediction of moisture content of the model.
Some of the metrics are blocked by your
consent settings
Evaluation of the moisture content of Jatropha curcas kernels and the heating value of the oil-extracted residue using near-infrared spectroscopy
(2015-02-01)
Posom, Jetsada
;
Sirisomboon, Panmanas
The use of near-infrared spectroscopy for evaluation of moisture content of Jatropha curcas kernels and heating value of its residue after oil extraction were studied. In total, 100 samples of whole kernels from green, yellow and black fruits and oven-dried kernels scanned in diffuse reflection mode using a Fourier transform NIR spectrometer at wave numbers of 1,250,000-400,000m-1 were used to develop moisture-predicting models. The corresponding residues after the oil extraction of the samples scanned in transflection mode using the same spectrometer and wave number range were used to develop the heating-value-predicting models. The models correlating the spectral data and the corresponding values measured using the reference method were developed by partial least squares regression and were validated using a test set. For the moisture content and heating value, coefficients of determination (R2) were 0.969 and 0.860, root mean square errors of prediction (RMSEP) were 4.0%wb and 360Jg-1, biases were -0.7%wb and -17Jg-1 and ratios of prediction to deviation (RPD) were 5.7 and 2.6, respectively. In addition, vibration bands of fibre and cellulose had important effects on the prediction of the heating value.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results