RMUTL Engineering Journal

Toxicity Posts and Hateful Comments Detection in Thai Language Using Supervised Ensemble Classification

2024-05-30T09:38:26+07:00

Social media platforms are the community people gather in where they can generally express their free willing opinions to others on any topics they attend. However, on many occasions, the cause of violating arguments or an unpleasant atmosphere in the community is initiated by negative, toxic, and hateful posts or comments. For that reason, monitoring post systems on social media is an essential topic in the natural language processing area, especially in multi-linguistics research. In this study, we proposed a method of improvement for the Thai language's toxic and hateful classification that was trained on the dataset of 2,160 posts from the Thai toxicity Twitter corpus for training and verifying. Therefore, we designated the ensemble approach which includes the combination of XGBoost, multinomial naive Bayes, logistic regression, support vector machine, and random forest for classifiers. In summary, the ensemble classifier improved the previous study in the same dataset with 0.7808 precision, 0.7778 recall, and 0.7721 average accuracies in the weighted F1 scoring with an accuracy of 0.8235 in the F1 binary scoring.

Exploring Waste Heat Recovery: A Case Study of a Batch Centrifugal Machine

2024-06-25T22:31:46+07:00

Recovering waste heat is crucial for improving energy efficiency in industrial processes. As the demand for sustainable and eco-friendly solutions grows, industries are seeking various methods to reclaim and reuse the heat generated during their operations. The primary objective of this study is to explore the potential for waste heat recovery in batch centrifugal machines by analyzing their current energy consumption and waste heat generation, investigating the feasibility and efficiency of various recovery methods, assessing the potential energy savings and environmental benefits of implementing these systems, and providing practical recommendations for industries to optimize their energy use, thereby demonstrating the significant role of waste heat recovery in enhancing the overall energy efficiency and sustainability of industrial operations. Through an analysis of empirical data collected from the operation of batch centrifugal machines, this study quantifies the magnitude of waste heat generation across different operating conditions. By examining power and current values obtained from the tables, the study identifies opportunities for energy recovery and optimization. Leveraging innovative techniques, the research demonstrates how proactive waste heat management can enhance energy efficiency and sustainability in industrial operations. The findings underscore the importance of integrated approaches to waste heat management, emphasizing its role in reducing environmental impact and improving overall operational performance.

Estimating the Length of Right-Turn and U-Turn Pocket Lane Using Binomial Distribution

2024-08-16T11:10:15+07:00

Delays at U-turns and signalized intersections often occur due to insufficient right turns and U-turn pockets. It is significant to assess and determine the lengths of the pocket lane for right turns and U-turns to improve vehicle efficiency in the area. The right-turn and U-turn pocket designs are based on a binomial distribution and comply with the standards of the DOH of Thailand. Research demonstrates that the flow becomes congested and optimized of right-turn and U-turn pockets, which vary with the volume of turning vehicles. Once the lane does not lack the capacity of a pocket lane, the flow rate of vehicles in both directions decreases, but the flow for right turns and U-turns increases. Therefore, determining the appropriate turn pocket can boost the flow rate and reduce travel delays in those areas.

Reducing Sink Mark Defect of Cup Plastic Injection by Biodegradable Material with an Injection Rate

2025-01-09T13:31:12+07:00

In this research, the issue of shrinkage in coffee cup sleeves filled with biodegradable plastic materials is investigated. Polylactic acid (PLA) and coffee grounds are two combined materials that are difficult to form by injection molding. The quality of the coffee cup sleeves declined as their surface was reduced. This study aims to investigate the effects of varying plastic injection molding parameters, including clamping force, injection pressure, injection rate, and cooling time, on shrinkage in the molded parts. The goal is to determine the optimal conditions for reducing shrinkage in the molded parts. The experimental results indicate that adjusting the injection speed rate to 40 cm³/s and the injection pressure to 54 bar significantly reduces shrinkage in the molded parts, thereby markedly improving the surface quality of the product. The results provided an important guideline for the development of injection molding technology employing bioplastic materials with higher viscosity than generally synthetic plastics. Other bioplastic goods can be produced using this method with favorable outcomes.

Static Responses of Fifth-Order Polynomial Shaped Shell under Hydrostatic Pressure

2024-08-14T16:30:47+07:00

This paper presents the static responses of the fifth-order polynomial shaped shell that supports the hydrostatic pressure. The geometry of the fifth-order polynomial shaped shell was computed by the differential geometry. The model of fifth-order polynomial shaped shell was desighed by using one-dimensional beam elements, which divided along the shell radius. In this study, the shell was separeted into 2 regions for preventing "dZ/dr=∞" and reducing the errors of result at equator plane. At the junction of two regions, was defined the function values, of displacements and continuous slopes The energy function of the shell can be derived by the principle of virtual work, and the static responses of the shell can be obtained by the finite element method.The results indicate that the sea level, shell thickness, and elastic modulus affec the displacement of the fifth-order polynomial shaped shell.

Concrete Paving Block Using Bagasse Ash and Calcium Carbide Residue with Sand and Calcite Residue as Fine Aggregate

2024-10-12T13:59:51+07:00

This research focuses on the development of concrete paving blocks using bagasse ash and calcium carbide residue as binders, with sand and calcite residue as fine aggregates. Bagasse ash is a by-product from biomass power plants, calcium carbide residue is a waste material from acetylene gas production, and calcite residue is obtained from limestone mining. The concrete paving blocks were produced using a binder-to-aggregate ratio of 1:3 by weight and a water-to-binder ratio from 0.45 to 0.63. The ratios of bagasse ash to calcium carbide residue used were 90:10, 70:30, and 50:50 by weight. Additionally, ordinary Portland cement was incorporated at 10%, 20%, and 80% by weight of the binder for the 70:30 mixture. The effects of using calcite residue as a fine aggregate replacement for sand were investigated using sand-to-calcite residue ratios of 100:0, 75:25, 50:50, 25:75, and 0:100 by weight, based on a binder ratio of bagasse ash to calcium carbide residue of 70:30. The concrete paving blocks were hand-formed and tested for compressive strength at the ages of 3, 7, 14, and 28 days, and for water absorption at 28 days. The results indicate that the 70:30 mixture of bagasse ash to calcium carbide residue with 80% cement replacement achieved the highest compressive strength and lowest water absorption. When calcite residue was used as the fine aggregate, the 0:100 sand-to-calcite residue ratio provided the highest compressive strength among the replacement levels, although it remained lower than when using sand alone. Although the compressive strength of these concrete paving blocks did not meet the standard requirements of TIS 827-2565, they still exceeded 17 MPa, making them suitable for non-load-bearing applications. This study demonstrates a practical approach for utilizing industrial byproducts, reducing cement consumption, lowering production costs, and promoting environmental waste management.

Economic Suitability Analysis of Solar Photovoltaic Modules and Energy Storage Systems Installation in Residential Units Group with Optimization Method

2024-12-19T05:19:19+07:00

This article presents an economic feasibility analysis of installing solar photovoltaic (PV) modules and energy storage systems for a residential complex of 11 buildings with 220 rooms. In total. The analysis employed a mathematical model and the Generalized Reduced Gradient (GRG) method to solve nonlinear optimization problems, by finding the minimum total project cost under specified conditions. The analysis reveals that installing a standalone 10 kW solar PV module results in the lowest total project cost and the optimal production capacity. Installing solar PV modules combin with an energy storage system is found to be economically unfeasible due to the high recent cost per unit of energy storage systems. Energy storage is applied for using the 100% solar energy efficiency by storing excess produced energy over the demand.The optimal product of the solar PV module and the energy storage system are 11 kW and 8 kWh, respectively. The installation of solar PV module has a payback period of 2 years and 9 months, which reduce the consumption of grid electricity by 23,158 kWh per year, decrease the electricity costs by 99,579 Baht per year, and diminish the CO₂ emissions by 27,790 kilograms per year. While the solar PV module with an energy storage system, has the payback period in 4 years, with the reduction of grid electricity consumption by 25,689 kWh per year and the reduction of CO₂ emissions by 30,827 kilograms per year. These findings are crucial for decision-making and policy formulation regarding renewable energy adoption as an alternative to fossil fuels. The key contribution of this study is the methodology for optimization using an objective function derived from relevant variables.

Baseflow Analysis of the Pasak River Basin

2025-03-19T15:12:28+07:00

This research aims to determine the appropriate parameter values for baseflow analysis of the Pasak River Basin and to identify suitable methods for baseflow separation. The study focuses on monitoring stations S.33 and S.4B, located in the upper part of the Pasak River Basin. The BFI+ model was used to obtain suitable parameter values and to evaluate the accuracy of five selected methods. These methods include four iterative digital filter methods (Lynie & Hollick, Chapman, EWMA, and Boughton two-parameter) and one graphical digital

method (Local Minimum), based on daily hydrological data. The results indicate that all five methods effectively

separate the baseflow at both monitoring stations in the Pasak River Basin. Among them, the Lynie & Hollick method demonstrated higher accuracy compared to the other four methods. Additionally, the optimal parameter values for both monitoring stations were found to be similar. This research highlights the importance of selecting appropriate baseflow separation methods and accurately calibrating parameters to ensure precise and reliable baseflow analysis in the Pasak River Basin.