KMITL

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

Browse

Search Results

Now showing 1 - 6 of 6

Some of the metrics are blocked by your
consent settings
Explainable machine learning for predicting thermogravimetric analysis of oxidatively torrefied spent coffee grounds combustion
(2025-04-01)
Pambudi, Suluh
;
Jongyingcharoen, Jiraporn Sripinyowanich
;
Saechua, Wanphut
Understanding the combustion behavior of oxidatively torrefied spent coffee grounds (SCG) is crucial for advancing sustainable fuel technologies. This study introduces a novel, explainable machine learning (ML) framework as a cost-effective alternative to traditional thermogravimetric analysis (TGA) that is designed to accelerate the evaluation of oxidatively torrefied SCG combustion properties. Four ML models: artificial neural network (ANN), k-nearest neighbor (k-NN), random forest (RF), and decision tree (DT), were compared to predict TGA data using proximate analysis and combustion temperature (CT). Among the evaluated models, k-NN exhibited the highest performance, achieving near-perfect R2 values that exceeded 0.9904 and RMSE values below 0.9552 on the validation set for both TG (mass loss) and DTG (derivative mass loss). It also accurately predicted key combustion properties, including ignition, peak, and burnout temperature when tested on unknown data. LIME (Local Interpretable Model-agnostic Explanations) analysis revealed that CT was the most influential predictor for TG and DTG, enhancing model interpretability. The results highlight the effectiveness of the k-NN-LIME approach in analyzing the combustion of oxidatively torrefied SCG, offering a robust and explainable model with significant implications for bioenergy research and sustainable fuel development.
Some of the metrics are blocked by your
consent settings
Machine learning based prediction and iso-conversional assessment of oxidatively torrefied spent coffee grounds pyrolysis
(2024-12-01)
Pambudi, Suluh
;
Jongyingcharoen, Jiraporn Sripinyowanich
;
Saechua, Wanphut
This research focused on developing a predictive model for mass loss during the pyrolysis of oxidatively torrefied spent coffee grounds (SCG) using machine learning techniques. Four algorithms were employed: artificial neural networks (ANN), k-nearest neighbors (k-NN), random forest (RF), and decision tree (DT), with the RF model demonstrating superior performance (R2 > 0.9981, RMSE <1.346) for both training and testing sets. The pyrolysis behavior, kinetics, and thermodynamics of SCG were also investigated using thermogravimetric analysis (TGA) under an inert atmosphere at different heating rates. Higher heating rates in TGA cause T<inf>peak</inf> values to shift to higher temperatures with increased DTG<inf>peak</inf> values, while also resulting in lower T<inf>onset</inf> and higher T<inf>offset</inf>. Kinetic analysis, using the Flynn-Wall-Ozawa (FWO) method, was identified as the most suitable approach for determining activation energy (E<inf>a</inf>), with values ranging from 192.66 to 288.13 kJ mol−1, indicating differences in energy requirements for pyrolysis across samples. Thermodynamic analysis further revealed that both raw SCG and oxidatively torrefied SCG pyrolysis were endothermic reactions. These findings contribute valuable insights into the optimization of biomass conversion technologies, highlighting the potential of machine learning in improving predictive accuracy and efficiency in thermal behavior modeling. This research advances sustainable bioenergy production by promoting the use of SCG, an abundant waste material, as a renewable feedstock in pyrolysis-based processes.
Some of the metrics are blocked by your
consent settings
A thermogravimetric assessment of eco-friendly biochar from oxidative torrefaction of spent coffee grounds: Combustion behavior, kinetic parameters, and potential emissions
(2024-02-01)
Pambudi, Suluh
;
Saechua, Wanphut
;
Jongyingcharoen, Jiraporn Sripinyowanich
This study investigated how temperature and residence time variations impact the combustion behavior, kinetic, and potential emissions of biochar produced through oxidative torrefaction of spent coffee grounds (SCG). The study examined the characteristics of combustion and kinetics by conducting thermogravimetry analysis under heating rates of 10 °C·min−1. The Coats-Redfern model was utilized to calculate kinetic parameters. While the emission indices were approximated using the data obtained from elemental analysis. The results indicated that biochar's comprehensive combustion index (C<inf>ci</inf>) from oxidative torrefaction was lower than that of raw SCG, suggesting stable combustion behavior. Moreover, with the escalation of torrefaction intensity, the activation energy (E<inf>a</inf>) values exhibited an upward trend for the char combustion stage, ranging from 22.08 kJ mol−1 to 38.46 kJ mol−1. Concurrently, the E<inf>a</inf> values pertaining to the oxidative pyrolysis stage decreased from 64.26 kJ mol−1 to 52.65 kJ mol−1. Besides, this study emphasized that the ash content of the biochar was lower than that of coal and remained consistent with the ash content of raw SCG (p > 0.05). Moreover, the study revealed that biochar from oxidative torrefaction emitted less CO<inf>2</inf> (67.35 g MJ−1) than lignite coal (76.55 g MJ−1). Additionally, biochar exhibited up to 27 times lower dust emissions than bituminous coal, emphasizing its eco-friendly fuel potential.
Some of the metrics are blocked by your
consent settings
On-line measurement of activation energy of ground bamboo using near infrared spectroscopy
(2019-04-01)
Sirisomboon, Panmanas
;
Posom, Jetsada
On-line measurement of activation energy (Ea) is very important in supporting the thermal conversion process. The main objective of this study was to evaluate the Ea of ground bamboo using near infrared spectroscopy in real time. 80 bamboo samples with different diameters were selected using random sampling. Ea was determined using the Coats-Redfern method, and Ea of reaction order (n) at n = 1 and n≠1 was investigated. The performance of on-line measurement predicted by PLS modelling for Ea at n = 1 and Ea at n≠1 showed coefficients of determination of 0.781 and 0.714, respectively; standard error of prediction of 5.249 and 6.858 kJ/mol, respectively; and bias values of −1.0628 and −1.871 kJ/mol, respectively. Both PLS models were found to be fair and could be applied toward screening. The results showed that the vibration bands of lignocellulosic components (CH<inf>2</inf>, hemicellulose, cellulose, and lignin) highly influenced model development. Moreover, internal relationships were identified among Ea, the pre-exponential factor (A), and n, such as A (1/min) = 63251 × e0.2200×Ea (at n = 1), A (1/min) = 33719 × e0.2267×Ea (at n≠1), and n = 0.008 × Ea+0.254. These relationships can be used to evaluate A and n if Ea is known. In the case of this study, Ea was forecasted using an NIR model.
Some of the metrics are blocked by your
consent settings
Evaluation of pyrolysis characteristics of milled bamboo using near-infrared spectroscopy
(2017-01-01)
Posom, Jetsada
;
Saechua, Wanphut
;
Sirisomboon, Panmanas
This paper reports the development of a rapid and low-cost method based on near-infrared spectroscopy as an alternative for thermogravimetric determination of the pyrolysis characteristics, including T<inf>onset</inf>, T<inf>sh</inf>, T<inf>peak</inf>, T<inf>offset</inf>and DTG<inf>peak</inf>, of milled bamboo. T<inf>onset</inf>is the extrapolated onset temperature that is calculated from the partial peak resulting from the decomposition of the hemicellulose component, T<inf>sh</inf>is the temperature corresponding to the overall maximum of the hemicellulose decomposition rate, DTG<inf>peak</inf>is the overall maximum of the cellulose decomposition rate, T<inf>peak</inf>is the temperature corresponding to the overall maximum of the cellulose decomposition rate and T<inf>offset</inf>is the extrapolated offset temperature of the DTG<inf>peak</inf>curves determined using thermogravimetric analysis (TGA). The models may be used to control the pyrolysis processes of bamboo to achieve the most economical and environmental conditions. 80 samples of bamboo with various circumferences of culms in the ranges of approximately 16–18, 18–20, 20–22, 22–24, 24–26, 26–28, 28–30, 30–32, 32–34, 34–36, 36–38 and 38–40 cm were randomly collected for optimization of the models. The models were optimized by partial least squares regression (PLSR) with 80% of samples for the calibration set and 20% for the validation set. For T<inf>onset</inf>, T<inf>sh</inf>, T<inf>peak</inf>, T<inf>offset</inf>and DTG<inf>peak</inf>the models showed coefficients of determination (R2) of 0.566, 0.845, 0.917, 0.973, and 0.671; root mean square errors of prediction (RMSEP) of 9.7 °C, 4.36 °C, 3.77 °C, 2.66 °C, and 0.428 wt loss %/min; ratios of prediction to deviation (RPD) of 1.52, 2.58, 3.48, 3.55, and 1.75; and biases of −0.344 °C, −0.765 °C, 0.349 °C, −5.41 °C, and 0.045 wt loss %/min, respectively. In addition, the results showed that pyrolysis characteristics did not depend on the circumference. The vibrational bands of water and CH<inf>3</inf>, O[sbnd]H stretch, first overtones of Ar–OH, CH<inf>2</inf>and HC[dbnd]CH in the cellulose and lignin structures, O[sbnd]H hydrogen bonds of polyvinyl alcohol and C[sbnd]H stretch corresponding to the first overtone of CH<inf>2</inf>had the highest influence on the values of T<inf>onset</inf>, T<inf>sh</inf>, T<inf>peak</inf>, and T<inf>offset</inf>, respectively. The vibrational band of the C[sbnd]O[sbnd]C asymmetrical stretches of cellulose and hemicellulose, and the combination of O[sbnd]H stretch and HOH bend of polysaccharides influenced the DTG<inf>peak</inf>value. These results are beneficial for studying the thermal behaviour of milled bamboo as a potential resource for producing biofuels, especially in the pyrolysis process.
Some of the metrics are blocked by your
consent settings
Grass blade-like microparticle MnPO4·H2O prepared by a simple precipitation at room temperature
(2010-11-01)
Boonchom, Banjong
;
Baitahe, Rattanai
;
Joungmunkong, Zongporn
;
Vittayakorn, Naratip
Grass blade-like microparticle MnPO<inf>4</inf>·H<inf>2</inf>O was synthesized by a simple precipitation at room temperature using a mixture of manganese sulphate monohydrate, phosphoric acid and water at pH=7. The thermogravimetric study indicates that the synthesized compound is stable below 500°C and its final decomposed product is Mn<inf>2</inf>P<inf>2</inf>O<inf>7</inf>. The pure monoclinic phases of the synthesized MnPO<inf>4</inf>·H<inf>2</inf>O and its final decomposed product Mn<inf>2</inf>P<inf>2</inf>O<inf>7</inf> are verified by XRD data. FTIR spectra indicate the presences of the PO<inf>4</inf>3- ion and water molecules in the MnPO<inf>4</inf>·H<inf>2</inf>O structure and the P<inf>2</inf>O<inf>7</inf>4- ion in the Mn<inf>2</inf>P<inf>2</inf>O<inf>7</inf> structure. The thermal stability, crystallite size, and grass blade-like microparticle of MnPO<inf>4</inf>·H<inf>2</inf>O in this work are different from previous reports, which may be caused by the starting reagents and reaction condition for the precipitation. © 2010 Elsevier B.V.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results