KMITL

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

Browse

Search Results

Now showing 1 - 10 of 27

Some of the metrics are blocked by your
consent settings
Enhanced forecasting of friction and cohesion of augmented unsaturated soil with nanostructured quarry fines (NQF) addition
(2026-12-01)
Kamchoom, Viroon
;
Van, Duc Bui
;
Hosseini, Shahab
;
Alimoradijazi, Mohammadreza
;
Amini-Khoshalan, Hasel
Shear strength parameters such as friction angle and cohesion are fundamental to solving geotechnical engineering problems related to slope stability, foundation design, and earthwork construction. This study presents the prediction of friction angle (Fi) and cohesion (Nc) of an unsaturated lateritic soil using three intelligent learning techniques: Support Vector Machine (SVM), Radial Basis Function (RBF), and Multilayer Perceptron (MLP), with Linear Multivariate Regression (LMR) adopted as a baseline model to evaluate agreement between input and output variables. The motivation for employing machine learning approaches stems from the limitations of complex laboratory testing and the need for reliable predictive tools that can support design and field applications. The investigated soil, classified as A-7-6 and poorly graded, exhibited coefficients of uniformity and curvature of 2.05 and 0.84, respectively. It was characterized by high plasticity and significant clay content, with a clay fraction of 23.02%, clay activity of 2, friction angle of 15°, maximum dry density of 1.84 g/cm3 at an optimum moisture content of 16.2%, and was tested under cyclic direct shear conditions. Multiple datasets were generated from varying treatment conditions and soil descriptors, forming the basis for model development. Eleven input parameters were used to predict Fi and Nc, and model performance was evaluated using Variance Accounted For (VAF), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R2). The results indicate that RBF and MLP outperformed SVM and LMR in both training and testing phases for predicting cohesion and friction angle, demonstrating superior generalization capability. Sensitivity analysis using the Cosine Domain Method revealed that unsaturated unit weight had the greatest influence on friction angle prediction, while clay content was the most influential parameter for cohesion. Among all models, MLP achieved the highest accuracy and overall predictive performance. Based on this optimal model, a Graphical User Interface was developed to enable users to input soil parameters and obtain rapid predictions, providing a practical tool for researchers and practitioners in geotechnical engineering.
Some of the metrics are blocked by your
consent settings
Rapid detection of potassium sorbate in coconut water using near infrared hyperspectral imaging
(2026-01-01)
Tantinantrakun, Achiraya
;
Kumpa, Benjaporn
;
Ainkast, Pranpriya
;
Thompson, Anthony Keith
;
Teerachaichayut, Sontisuk
Potassium sorbate may be illegally added to fresh coconut water in order to prolong its marketable life, but this adulteration may not be identified on the product label. The aim of this research was therefore to evaluate if samples of fresh coconut water that had been adulterated with measured amounts of potassium sorbate could be detected by near infrared hyperspectral imaging (NIR-HSI). Samples of coconut water with different potassium sorbate concentrations (N = 100) and pure coconut water samples (N = 100) were used in this study with their averaged spectral data used as independent variables. The smoothing spectral pretreatment gave the highest classification accuracy of 98.48% by partial least squares discriminant analysis (PLS-DA). While support vector machine regression (SVMR) with spectral pretreatment, using the 1st derivative combined with multiplicative scatter correction (MSC), achieved the optimum condition for developing the calibration model for determining potassium sorbate concentration with the correlation coefficient of prediction (R<inf>p</inf>) of 0.818 and the root mean square error of prediction (RMSEP) of 327.86 ppm. The results showed that NIR-HSI was able to be used as a fast, reliable, economic and environmentally friendly method of detecting potassium sorbate addition to coconut water.
Some of the metrics are blocked by your
consent settings
Forecasting maize yield from growth parameters using machine learning in a biochar-inorganic fertilizer amended soil under drip irrigation
(2025-12-01)
Faloye, Oluwaseun Temitope
;
Ajayi, Ayodele Ebenezer
;
Kamchoom, Viroon
;
Sinsamutpadung, Natdanai
;
Adeyeri, Oluwafemi
The combined application of biochar and inorganic fertilizers has demonstrated significant potential to enhance crop productivity under both rainfed and irrigated conditions. However, predictive modeling approaches utilizing machine learning (ML) to simulate field outcomes under diverse agronomic scenarios remain understudied. This study addresses two critical objectives: (i) to evaluate the efficacy of ML models—Support Vector Machine (SVM), Artificial Neural Network (ANN), and Boosted Trees (BT)—in predicting maize grain yieldin biochar-inorganic fertizer amended soil under drip irrigation; and (ii) to identify the growth stage(s) and ML models that deliver the most accurate predictions. A three-year factorial field experiment was conducted during dry seasons, testing five biochar rates (0, 3, 6, 10 and 20 t/ha), two fertilizer levels (0 and 300 kg/ha), and deficit irrigation treatments (60%, 80%, and 100% of full irrigation). Growth parameters were measured at vegetative (35 days after planting – DAP), flowering stage (49 DAP), and maturity stage (77 DAP), with grain yield recorded at harvest (90 DAP). The measured growth parameters at the different DAP were used for the grain yield forecast. 70, 15 and 15% of the dataset were used for model training, validation and testing, respectively. Field results revealed progressive increases in growth parameters from vegetative to maturity stages, with treatment efficacy following the order: control < biochar-only < fertilizer-only < combined biochar-fertilizer. ML predictions mirrored this hierarchy, with ANN achieving superior accuracy (R² = 0.73–0.85, RMSE = 0.43–0.76, NRMSE = 0.095–0.17 at maturity) compared to SVM and BT. Predictive performance was weakest at the vegetative stage (35 DAP) but improved during flowering (49 DAP) and maturity (77 DAP), underscoring the importance of later growth data for reliable yield forecasting. This study demonstrates that ML models, particularly ANN, can effectively predict maize yield using accessible growth metrics, offering a cost- and labor-efficient complement to traditional field research. By enabling rapid scenario analysis, such models empower stakeholders to optimize resource allocation and inform crop management decisions under varying irrigation and soil amendment strategies.
Some of the metrics are blocked by your
consent settings
Novel method for predicting the cracks of oxide scales during high temperature oxidation of metals and alloys by using machine learning
(2025-12-01)
Chawuthai, Rathachai
;
Promchan, Teeratat
;
Rojsanga, Jularak
;
Chandra-ambhorn, Somrerk
;
Nilsonthi, Thanasak
Material degradation is one of the main problems in various high-temperature processes, directly resulting in the failure of the material. Crack and protective oxide film spallation caused either by mechanical stress development in the oxidation process or thermal stress due to a mismatch of the thermal expansions of the formed oxide and alloy are common forms of failure in high-temperature processes. Typically, the Pilling-Bedworth ratio (PBR) is employed to predict crack and spallation of the oxide by determining the volume changes of oxide and alloy because of its simplicity. However, this approach provides poor crack and spallation predictions. Hence, machine learning was adopted in the present work to predict oxide formation and spallation in the temperature range of 600-1,200 °C. The inputs for the present developed model were alloy compositions, oxide formed during oxidation, and oxidation conditions and periods. Furthermore, the predicted results of the present developed machine learning model were compared to those obtained by the PBR method. The present results revealed that the accuracy of the oxide spallation prediction of the present model was better than that of the PBR method. The random forest with 15 estimators was the best machine learning model. Finally, it can be concluded that the machine learning model is essential for accurate material failure prediction.
Some of the metrics are blocked by your
consent settings
Near-infrared hyperspectral imaging for predicting the quality of SO2 pre-treated and dehydrated mango
(2025-08-01)
Aozora, Wayan Dipasasri
;
Tantinantrakun, Achiraya
;
Thompson, Anthony Keith
;
Teerachaichayut, Sontisuk
Prediction for quality indices of SO<inf>2</inf> pre-treated and dehydrated mango was accessed by NIR-HSI. Models for predicting TSS and SO<inf>2</inf> content achieved R = 0.82; RMSEP = 2.42% and R = 0.83; RMSEP = 56.40 mg/kg, respectively. Visualization of TSS and SO<inf>2</inf> content could be presented by predictive images.
Some of the metrics are blocked by your
consent settings
Assessing adulterated pineapple juice concentrate using electrical properties
(2025-01-01)
Tantinantrakun, Achiraya
;
Sinsamut, Varisara
;
Apairat, Nuengruthai
;
Smutrakalin, Thirapol
;
Thompson, Anthony Keith
The fraudulent addition of sugars to pineapple juice concentrate undermines consumer trust and satisfaction. Resistance (R), capacitance (C), dissipation factor (D), inductance (L), quality factor (Q), impedance (Z) and phase angle (θ) in the range of 0.012–200 kHz of juice adulterated with sugar increasing levels from 0 to 95% at 0.5% (w/w) intervals were tested to determine whether they could be used for detecting adulteration in pineapple juice concentrate using a LCR (inductance, capacitance, resistance) meter. A multiple linear regression (MLR) model was developed for predicting the concentration of additive sugars in samples. Linear discriminant analysis (LDA) was used for classifying pure pineapple juice concentrate and pineapple juice concentrate adulterated with added sugars. The most accuracy in the MLR model was obtained from θ, which achieved a correlation coefficient of prediction (R<inf>p</inf>) of 0.977 and a root mean square error of prediction (RMSEP) of 5.88% w/w. From the LDA analysis, the most accurate parameter for classification was C, which yielded a predictive classification accuracy of 94.57%. Therefore, this technique indicates its potential for use in the fruit juice industry a simple method for routinely testing in order to ensure the non-contamination of products offered for sale to consumers.
Some of the metrics are blocked by your
consent settings
Small scale method for estimation of genetic coefficients of photoperiod-insensitive rice using generalized likelihood uncertainty estimation
(2023-03-01)
Suanphrom, Nattawut
;
Khurnpoon, Lampan
;
Phakamas, Nittaya
Importance of the work: Genetic coefficients are important parameters for simulations of rice yield performance in crop growth models. Most genetic coefficients (GCs) are obtained from large experiments. Objectives: To estimate the GCs of seven photoperiod-insensitive rice cultivars for four planting dates in a pot experiment. Materials & Methods: Input data (soil, weather, management, plant parameters) were collected and used to calibrate the GCs of seven rice cultivars using the GLUE estimator in the DSSAT version 4.7 package. The data were collected from four planting dates: 1) 23 Nov 2019; 2) 23 Dec 2019; 3) 23 Jan 2020; and 4) 23 Feb 2020. The data from planting dates 1, 3 and 4 were used for calibration of the GCs, whereas the data from planting date 2 were used for evaluation of the GCs. Results: Good prediction qualities of the model for most cultivars were indicated for days to anthesis and days to physiological maturity; however, there were poor prediction qualities for almost all cultivars for their biomass and grain weight. Main finding: This information should be useful for further investigations of GCs in rice. Although the results were contrary to the initial hypothesis, the method showed promise for further use in rice modeling research if the method can be improved by experimental management, the use of suitable reference plants for each cultivar and running the model for an appropriate cycle. It was possible to obtain some reliable GCs from small-scale experiments, so the experiment should be improved to obtain better results. Further investigations should focus on the optimum scale and weather data specific to experimental sites.
Some of the metrics are blocked by your
consent settings
Effect of wood vinegar substitutes on acetic acid for coagulating natural para rubber sheets during the drying process
(2021-09-01)
Kalasee, Wachara
;
Dangwilailux, Panya
The coagulating properties of wood vinegar from para rubber wood, bamboo, and coconut shell used as a substitute for acetic acid in the production process of natural rubber (NR) sheets were investigated and considered. For the dirt and volatile content, the tensile strength at break, the percentage of elongation at break, and the 300% modulus, the results showed that the types of wood vinegar coagulants were not significantly different from acetic acid. However, the Mooney viscosity and plasticity retention index (PRI) properties were significantly different from those of acetic acid. The NR sheet temperature increased rapidly during the first hour after the drying process started due to heat transfer from the hot air. Afterward, the temperature of the NR sheet samples began to stabilize. When the drying process started, the drying temperature was increased, so the trend was reducing the drying time. For the yellowness index (YI) value, the increase in the YI value was related to the type of coagulating material, the increase in the airspeed, and the drying temperature. The dried sheet samples using para rubber wood vinegar as the coagulating material had a color value at the same level as acetic acid and the referent. However, the bamboo and coconut shell wood vinegars were at a lower level. In comparing the YI value data between the experimental results and prediction values, the second-degree model had a better fit in prediction than the zero-degree and first-degree models. This result was confirmed by the higher mean of the coefficient of determination. The dried sheet product coagulated by using wood vinegar had fungus growth prior to supplying it to the customer.
Some of the metrics are blocked by your
consent settings
Smoke particle, polycyclic aromatic hydrocarbons and total benzo[a]pyrene toxic equivalence emitted by palm oil sewage sludge bio‐char combustion
(2021-09-01)
Kalasee, Wachara
;
Dangwilailux, Panya
The size distribution, total particle mass concentration (TPMC), polycyclic aromatic hydrocarbons (PAHs) value, and total Benzo[a]pyrene Toxic Equivalence (BaPTE) concentration of smoke particles from palm oil sewage sludge (POSS) bio‐char combustion were studied. In this experiment, temperature data of the POSS bio‐char combustion were recorded in two parts: particle temperature (Tp) by using a two‐color pyrometer and temperature at 300, 500 and 800 mm, respectively, above the fire base by using K‐type thermocouples. The POSS bio‐char moisture content, clean air speed values, and burning period affected the change of temperature above the fire base. The mass median aerodynamic diameter (MMAD) values of the POSS bio‐char combustion were found to be 0.44 to 1.05 micron at various moisture contents and burning periods. The MMAD, TPMC, and PAHs values increased with increasing moisture content and decreased the POSS biochar combustion period. For the total BaPTE values, the results showed that the decrease in moisture content of the POSS bio‐char samples had a prime influence in decreasing the total BaPTE values. Meanwhile, with decreases in the clean air speed values, the total BaPTE values were increased. Comparing the total BaPTE data between the experimental results and predicted values, the first‐degree model had a better fit in predicting than the zero‐degree model; this result was confirmed by the higher mean of the coefficient of determination.
Some of the metrics are blocked by your
consent settings
Prediction of size distribution and mass concentration of smoke particles on moisture content and combustion period from para rubber wood burning
(2021-06-02)
Kalasee, Wachara
;
Dangwilailux, Panya
The size distribution and total particle mass concentration (TPMC) of smoke particles from para rubber wood (Hevea brasiliensis) combustion in the ribbed smoked sheet (RSS) process were studied. In this experiment, temperature data values of para rubber wood combustion were recorded at 500 mm above the base of the fire by K-type thermocouples. The wood moisture content and wood combustion period were used to find and improve an equation of smoke particle size distribution (SPSD) and TPMC by the response surface method (RSM). An eight-stage Andersen air sampler and a high-volume sampler were used to measure and calculate SPSD and TPMC, respectively. Resulting data in this experiment showed that TPMC ranged from 3.12 to 77.42 mg/m<sup>3</sup> . SPSD was single mode in which MMAD, mass median aerodynamic diameter, ranged from 0.64 to 1.27 microns for para wood with moisture content ranging from 31.5 to 89.7% dry weight basis. The combustion period and moisture content of para wood have a direct effect on the change of temperature data above the base of the fire and the TPMC and MMAD values. For predicting TPMC and MMAD values by the para wood moisture contents in each combustion period, the results found that the second-degree model was a better plot than the first-degree model, confirmed by higher values of the coefficient of determination (R<sup>2</sup> ).

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results