Engineering and Technology Horizons

Predictive Maintenance of Banbury Mixer Using Machine Learning Methods: A Case Study of a Tire Manufacturing Factory

2024-11-27T14:55:49+07:00

This study investigates parameters of the Banbury mixer, focusing on Material Rubber Sheet (MRS) production duration, to develop a machine learning-based failure detection system. Proposed data preprocessing methods aim to uncover patterns of the relationship between current and previous batch parameters. While data differencing suffices for SVM and RF models, it minimally impacts ANN. Solely employing data cleaning renders RF suitable for model creation. SVM outperforms RF in failure detection but may produce occasional false alarms. Performance evaluation indicates RF accurately detects 13 out of 29 failures, surpassing SVM detecting 16 failures with 2 false alarms. The SVM model from data cleaning combined with differencing reduces waste by 55.17%, surpassing RF by 6.89%. Future research could explore advanced preprocessing for failure cause categorization and leverage sophisticated techniques.

Solar Irradiance Forecasting using Dynamic Mode Decomposition

2024-11-21T10:03:07+07:00

Reliable solar irradiance prediction is necessary for an easier transition from dependence on fossil fuels to renewable energy sources. The features of solar irradiance, such as its non-linearity and high variability, make predicting it a challenging task. This challenge is traditionally addressed by using regression and other ensemble models that require significantly large historical data to adequately train and rely on domain-specific knowledge. In this study, a data-driven framework that employed dynamic mode decomposition for solar irradiance forecasting was proposed. The efficiency of the dynamic mode decomposition-based framework was verified by employing it for short-term forecasting using two distinct datasets from geographically diverse locations. The comparative advantage over traditional regression was confirmed using performance assessment measures, including mean absolute error, mean bias error, and root mean square error. The resulting forecasts significantly outperformed the benchmark models, demonstrating that the proposed model could effectively forecast short-term solar irradiance with improved accuracy.

Effects of residual particles and multiply charged particles of a polystyrene latex particle on the particle filtration efficiency of a medical face masks

2024-11-21T10:22:14+07:00

The objective of this study is to elucidate the impact of residual particles and multiply charged particles of polystyrene latex (PSL) particles on the particle filtration efficiency of a medical face mask. Adhering to the ASTM F2299 test methodology, this investigation employed a PSL particle (Thermo ScientificTM Dri-CalTM) with a diameter of 100 nanometers and a geometric standard deviation of approximately 1.6%. The scanning mobility particle sizer (SMPS) was utilized to ascertain the size distribution of PSL particles both upstream and downstream of the test face mask, thereby determining the particle filtration efficiency of the face mask for residual particles and multiply charged particles at various PSL particle-to-DI water ratios. The PSL particle-to-DI water ratios employed in this study were 10, 20, and 100 drops per 200 mL, respectively. The findings revealed that PSL suspension residual particles with a mobility diameter less than 60 nanometers, as well as multiple charged particles with a mobility diameter exceeding 100 nanometers, introduced complexity to purportedly monodisperse particle aerosols. An increase in PSL suspension concentration resulted in a heightened relevance of the residual peak. At elevated concentrations, PSL multiply charged particles became discernible in the size distribution. Furthermore, it was demonstrated that as the PSL particle-to-DI water ratios increased, the particle filtration efficiency of the test face mask diminished for both single charged particles and multiply charged particles (+2, +3, +4, and +5).

Lean Improvement of the Fuzzy Overall Equipment Effectiveness for the Half-Bearing Production Processes using the System Dynamics Simulation

2024-12-23T09:08:51+07:00

This research presents the problem of decreasing the production efficiency of the Half-bearing production process resulting in parts production quantities not meeting the target. The OEE is used as a production efficiency index, which measures availability rate, performance efficiency and quality rate. The characteristic of production data in practice is dynamic behaviour. The objectives of this research are to analyze problems of the half-bearing production process to make improvements using OEE assuming time and loss to be TFNs. The research method is to develop a system dynamics model to analyze %OEE, whereas %OEE before improvements ranges between 67–72%. The improvement operations are waste reduction, planned maintenance, system testing and tool change. The results found that the %OEE after improvement range between 70–77% and the %improvement range between 3–11%.

Prediction of Vapor Pressure of Edible Vegetable Oils with Solvents at Temperature 348.15 K to 373.15 K from Activation Energy

2025-02-10T08:37:50+07:00

Edible vegetable oil is a food product extracted from plant seeds and is one of the key ingredients in global cuisine. It plays a vital role in enhancing the flavor, texture, and color of food. Additionally, vegetable oils contain natural antioxidants such as tocopherols, phenolic compounds, and phytosterols, which have health benefits. Vegetable oils play a crucial role in various industries, including food, chemicals, pharmaceuticals, cosmetics, and fuels. Predicting the vapor pressure of vegetable oils and solvents in a two-component phase equilibrium system is essential for solvent extraction processes and their subsequent reuse. This study presents a predictive equation for vapor pressure derived from Gibbs free energy, accounting for the influence of molecular interactions in the two-component phase equilibrium system. The research focuses on three types of vegetable oils: soybean oil, sunflower oil, and canola oil, along with three solvents: methanol, ethanol, and hexane, evaluated at temperatures of 348.15 K to 373.15 K across a molar concentration range of solvent = 0.2–1.0. The proposed equation is In T demonstrates high accuracy, with an average bias of -0.01 and an average absolute deviation (AAD) of 1.030.

The Impacts of Climate Change on Future Hydrometeorological Regimes in the Chao Phraya River Basin, Thailand

2025-01-20T08:47:06+07:00

The major aim of this study is to investigate the impacts of climate change on hydrometeorological regimes including air temperature, rainfall, potential evapotranspiration and consequently natural runoff. Chao Phraya basin was selected as the case of study area. Rainfall variation and change of the selected irrigation projects in the Chao Phraya basin was also examined. The historic records of air temperature and rainfall were collected from 41 meteorological stations located in and nearby the basin area. Climate change was investigated by using the Generalized Circulation Model (GCM) projection of A1B balance of all source scenario. The study used output data from the United Kingdom Hadley Center’s Global Climate Model “HadCM3 (10) (13)” with approximately 180 × 180 km. spatial resolution. Downscaling into the Chao Phraya basin was accomplished using the PRECIS Regional Climate Model with approximately 25 × 25 km. resolution along with the bias correction technique with reference to the mean observed time series. It was found that in the 20-year period during 2017–2036, the average monthly maximum and minimum air temperatures in the Chao Phraya basin trend to increase throughout the normal baseline values during 1981–2010 by 0.5–2.0 degree Celsius. The monthly rainfall under climate change in the Chao Phraya basin over 20-year period during 2017–2036 which is an average of every 5 years, shows significant deviation both higher and lower from the normal baseline values during 1981–2010 in the range of 10 mm per month to 50 mm per month. The monthly average runoff in the Chao Phraya basin over the 20-year period during 2017–2036, average every 5 years across all 4 periods during 2017–2021, 2022–2026, 2027–2031, and 2032–2036 are significantly varied. The highest values during the rainy season were found in the 2027–2031 average period. While the lowest values during the rainy season were observed in the 2022–2026 period. These findings should be useful for water allocation management and appropriate use of water resources under climate change condition in the Chao Phraya basin.

Development of a GIS-based Bird Map Application for the Royal Thai Air Force’s Aviation Safety Management

2025-01-30T08:44:10+07:00

This study aimed to assess bird strike risk and develop a GIS-based bird map application to enhance aviation safety for the Royal Thai Air Force. Five risk factors, including bird size, weight, density, activity range, and flight altitude were collected and analyzed using the data from the Royal Thai Air Force Software Center as well as the 6th Wing squadrons. Severity level and probability of occurrence were calculated and used to design a bird strike risk assessment application. The developed application allows users such as pilots and air operators to assess bird strike risk and display real-time maps and images of bird coordinates. This tool can increase safety and reduce bird strike accidents, serving both military and civilian aviation needs.

Rapid estimation method for span load distribution on a rectangular wing planform using tabular data from lifting-line theory

2025-01-16T07:51:18+07:00

The aim of this article is to present the rapid estimation method for span load distribution on a rectangular wing planform by using the tabular data from lifting-line theory. The data is generated by solving a monoplane equation to determine the coefficient of Fourier sine series that represents horseshoe strength distributed along the wingspan. Then the ratio of local lift coefficient to wing lift coefficient at the selected positions are determined and are presented in the tabular form. This set of data can be used to determine the spanwise lift distribution, the magnitude of total lift force and spanwise center of pressure location. The present method can be applied for a rectangular planform which has no sweep angle at quarter chord line and has no both geometric and aerodynamic twist along wingspan. The range of wing aspect ratio is 4 to 12 and the taper ratio is 0.1 to 1.0. Although the assumption of flow around wing planform is incompressible flow or at Mach number lower than 0.3, this present method can be extended to be used for subsonic compressible flow condition by using Prandtl-Glauert rule which is also discussed in this article. An illustrative example of the calculation process is presented in this article and the comparison of the result from the present method to the other calculation methods and experimental data are also discussed in this article.