Journal of Science and Transport Technology

Evaluation of Carbon Emission Reduction in Concrete Using Fly Ash and Slag: Case Studies from Vietnam

2025-07-16T08:15:27+00:00

Concrete is one of the most widely used construction materials; however, its high cement content is a major contributor to global carbon emissions. This study applies Life Cycle Assessment (LCA) to evaluate carbon emissions from various cement replacement scenarios using fly ash and slag, based on multiple studies in Vietnam. Experimental results show that replacing 20%–40% of cement with fly ash reduces emissions by 9.14%–40%, while 40% slag replacement achieves a 31.84% reduction. A combined mix with 20% fly ash and 20% slag offers the best balance, reducing emissions by 32.16% while maintaining a compressive strength of approximately 50 MPa. These findings highlight the potential of industrial by-products in reducing carbon footprints while maintaining concrete performance. This study provides insights for optimizing sustainable concrete mix designs in Vietnam, promoting greener construction practices.

Real-time Vehicle Behavior Classification using Single 3-Axis Magnetic Field Sensor and Neural Network

2025-08-01T04:30:56+00:00

This paper presents a real-time method for classifying vehicle behaviors, particularly related to lane violation behaviors by analyzing the output of magnetic sensors. The system assumes the sensor is installed beneath lane markings to detect magnetic field disturbances as vehicles approach. The feature extraction process emphasizes the vertical (Z-axis) magnetic field component and its declination angle, both of which demonstrate robust discriminative characteristics across different vehicle types and positions. A lightweight neural network classification model, based on those features, is trained on these features and deployed on embedded hardware to ensure rapid response and minimal power consumption. The proposed model achieves an overall accuracy of 89.75%, in distinguishing between legal (in-lane) and illegal (lane-crossing) vehicle behaviors. This work introduces a novel integration of simplified signal processing and efficient machine learning suitable for real-time deployment in low-cost Intelligent Transportation Systems (ITS), especially in dense or infrastructure-limited environments.

System design and Flow simulation of a Blowdown Sliding-block Supersonic Wind tunnel

2025-06-12T09:17:58+00:00

This study presents the design, simulation and performance evaluation of a blowdown sliding-block supersonic wind tunnel having a test section sized 20 × 20 cm, capable of operating in the range of Mach number from 2 to 4. This sliding-block mechanism allows precise control of Mach number by adjusting the nozzle throat area. A computational fluid dynamics (CFD) simulation was performed at Mach 2.5 to evaluate the flow characteristics. The 3D geometry was simplified into a 2D axisymmetric model, and a structured quadrilateral mesh was generated using ANSYS Meshing. The density-based solver in ANSYS Fluent with the RNG k-epsilon turbulence model was employed to capture supersonic flow phenomena. Results demonstrate that the system achieves the target Mach number with a relative error of less than 0.4%, indicating excellent flow quality, and no major observed shock wave. The study validates the wind tunnel's performance, providing a reliable foundation for experimental aerodynamic testing.

Development and Evaluation of DUT Vibro: A High-Precision Vibrating Wire Sensor Readout

2025-05-19T09:35:43+00:00

This study presents the development and evaluation process of a vibrating wire sensor readout named DUT Vibro, with both hardware and software entirely developed by Vietnamese researchers. The primary objective of this paper is to evaluate the efficacy of two methods to determine the resonant frequency of the vibrating wire to facilitate precise strain calculations. In this study, parabolic interpolation is investigated to improve the accuracy of the resonant frequency of a steel wire, addressing the limitation of Fast Fourier Transform (FFT) constrained by the storage capacity of the microcontroller. In addition, the determined resonant frequency of DUT Vibro is compared with a commercial DIGIANGLE (DAS). Furthermore, two stimulation signals—sine and square waves—were employed to compare their impact on measurement accuracy. The results indicate that the parabolic interpolation method yields the lowest standard deviation, closely aligning with the DAS readout, and demonstrates stability across both low and high load conditions. In contrast, the FFT method exhibits greater error variability, particularly in the medium load range, due to the influence of noise and non-linearities in the response signal. The sine wave stimulus combined with parabolic interpolation achieves the highest accuracy. The measurement system maintains high linearity, with linearity errors below 0.5% of full scale (FS), and the lowest linearity error is 0.129% FS when using a sine wave stimulus. Linear regression analysis reveals a slope coefficient of approximately 0.052, reflecting a linear relationship between load and measured strain. Based on these findings, the parabolic interpolation method has been integrated into the DUT Vibro readout, meeting stringent accuracy requirements for strain measurement applications.

Accurate and Interpretable Prediction of Marshall Stability for Basalt Fiber Modified Asphalt Concrete using Ensemble Machine Learning

2025-06-18T09:33:40+00:00

Marshall Stability (MS), a parameter that reflects the load-bearing capacity and deformation resistance of asphalt concrete, is critical for pavement performance and durability. This study assesses the predictive capability of five tree-based machine learning (ML) algorithms - Decision Tree Regression, CatBoost Regressor, Random Forest Regression, Extreme Gradient Boosting Regression, Light Gradient Boosting Machine - in estimating the MS of basalt fiber - modified asphalt concrete (BFMAC). A compiled database of 128 samples was used for model training. Models were optimized with GridSearchCV and 5-fold cross-validation (CV), assessed via multiple statistical metrics, while SHAP analysis provided model interpretability. Among the tested models, Random Forest Regression (RFR) demonstrated the highest predictive accuracy (R² ≈ 0.922, RMSE ≈ 0.748 on the test set) and exhibited strong generalization capability. Interpretability analysis revealed that aggregate gradation (specifically, percentage of aggregate passing 2.36 mm and 4.75 mm sieves) and binder penetration were the most significant factors influencing MS prediction, followed by fiber content. This research underscores the potential of interpretable ML models, such as RFR, in accurately predicting MS, offering a viable alternative to conventional experimental methods for pavement material assessment.

Investigation into the behavior of ballasted railway track foundations through numerical analysis

2025-06-02T07:31:22+00:00

Railways are now the most widely used form of public transport due to recent traffic congestion on highways in several countries worldwide, which has raised demand for quicker and larger trains. Modern railway traffic has led to the beginning of hefty wheel loads and high-speed trains, which has increased track layer stresses and resulted in excessive vibrations under dynamic loading. So, there is now a major rise in the risk linked to train operations in terms of train safety, track foundation degradation, and track damage. Studying how ballasted railway track foundations behave under various train speeds is essential to ensuring the dependability and safety of high-speed trains. The railway tracks were also stabilized with geo-grid to reduce the settlement of the track. This research paper presents a three-dimensional (3D) numerical approach to imitate the dynamic reaction to the ballasted railway's track subgrade systems for high-speed trains (HST). This study determined the time history curves for rail vertical displacement for the different elastic moduli of track layers. Also, the position of the geo-grid in the ballast layer (from 0.1 to 0.3 m) is investigated. Geo-grid stabilization can reduce about 24% vertical displacement for train dynamic load in the position of 0.1 m from the top of the ballast layer. The obtained results offer valuable insights into the dynamic reaction of ballast railway track subgrade systems, which can be utilized to enhance the design and maintenance of modern railways.

Vibration analysis of a mobile military repair vehicle during movement

2025-09-11T09:32:26+00:00

Military mobile repair vehicles are specialized vehicles capable of lifting, holding, and moving loads on complex terrain. The article presents a dynamic model of military mobile repair vehicles when holding and moving loads on road surfaces with random or harmonic profiles. The dynamic model considers the elastic deformation of tires, suspension systems, hoisting cables, and wind resistance acting on the loads during vehicle movement. This is a unique dynamic model due to the combination of crane and three-axle truck models. Lagrangian equations are applied to establish a system of differential equations describing the oscillations of the system and solve them using simulation methods in Matlab software. The results of the article indicate the displacements, velocities, and accelerations of all components in the system, particularly demonstrating the oscillation of the load when the vehicle moves on road surfaces with different profiles and velocities. The results of the article provide a basis for evaluating the process of holding and moving loads by military mobile repair vehicles across various terrains. The recommendation is to hold and move loads at a slow speed ranging from 2 m/s to 3 m/s. The findings also serve as input for proposing stable control solutions for the base vehicle and the load during movement.

Reuse of coal combustion ashes as alternatives to cement and natural fine aggregate in green building bricks production for sustainable development

2025-09-15T14:59:50+00:00

The coal thermal power plant generates a lot of ashes, including fly ash (FA) and bottom ash (BA), which potentially pollute the environment. Thus, to promote sustainable development, this research assesses the feasibility of converting these ashes into green building bricks (GB). In detail, BA was utilized to fully replace natural fine aggregate in the GB mixes, and FA was utilized as a binder material to partially replace cement at various weight percentages of 0%, 30%, 50%, 70%, and 85%. The designed grade of GB in this study is M7.5, as classified in the TCVN 6477:2016, which is commonly used in non-loading bearing wall applications. The influence of FA replacement level on the GB’s properties, such as compressive strength, ultrasonic pulse velocity (UPV), electrical resistance (ER), and thermal conductivity (TC), was investigated. Moreover, the mineralogy change and microstructure of the GB samples were identified using X-ray diffraction and scanning electron microscopy techniques. The experimental findings demonstrate that replacing cement with FA had a substantial effect on all of the GB sample performance. At 28 days, GB samples had compressive strength, UPV, ER, and TC values ranging from 3.1 to 14.8 MPa, 1526 to 3360 m/s, 5.1 to 18.3 kΩ.cm, and 0.29 to 0.69 W/mK, respectively. These findings illustrate the viability of employing FA and BA in the manufacturing of GB for sustainable construction. As a result, the GB with ≤30% FA replacement met the target strength of ≥7.5 MPa, qualifying for use in a non-loading bearing wall in real practice.

Capillary Wick Irrigation Technique: A Sustainable Hydraulic Innovation for Water-Efficient and Climate-Resilient Infrastructure in Arid Regions

2025-06-20T14:43:08+00:00

Groundwater is a vital resource supporting agriculture, industry, and rural livelihoods. However, changing climatic patterns, erratic rainfall, and unsustainable human activities have accelerated groundwater depletion, posing major challenges to sustainable water management. In response, this study introduces the Capillary Wick Irrigation Technique (CWIT) an innovative, passive irrigation system designed to enhance water use efficiency and promote sustainable agricultural infrastructure, particularly in arid and saline-prone environments. Unlike conventional drip systems, CWIT utilizes capillary action through specially engineered wick structures embedded in a subsurface pipe network, eliminating the need for external energy or technical operation. Experimental trials on fennel crops under controlled saline conditions revealed a distinct hemispherical wetting front, extending vertically up to 50 cm and horizontally up to 30 cm, with soil moisture retained for up to 12 days without additional irrigation. Field studies conducted in Velavadar, Surendranagar District, Gujarat, further validated the technique, showing approximately 17.4% water savings over drip irrigation and nearly 85% compared to traditional surface methods. CWIT also enhanced crop yield efficiency, reduced evaporation and runoff, and supported soil conservation and groundwater recharge. Offering a low-cost, scalable, and environmentally resilient solution, CWIT presents strong potential for integration into rural water systems and climate-resilient farming, particularly in water-scarce regions.

Optimizing Semantic Segmentation for Autonomous Vehicle Scene Understanding in Unstructured Indian Traffic through Reinforced Active Learning

2025-08-18T04:02:34+00:00

Autonomous vehicles (AVs) offer a radical leap in transportation, delivering safer and more efficient mobility options. The capacity to interpret complicated surrounding traffic scenarios in real-time is central to their effectiveness. Scene awareness, especially semantic segmentation, is vital in allowing AVs to successfully comprehend and navigate their environments. However, limited labelled data availability and dataset biases restrict the effectiveness of semantic segmentation models, especially in specific contexts such as Indian driving scenarios. This study presents a novel approach employing reinforced active learning to overcome the aforementioned difficulties. Reinforced active learning integrates reinforcement learning into the active learning framework, allowing the model to select samples for annotation based on model operations and uncertainty estimation. By augmenting the segmentation model with annotation effort, our approach enhances performance in real-world driving scenarios in India. Rigorous testing and validation on the Indian Driving Dataset (IDD) demonstrate improvements in segmentation precision and effectiveness compared to training methods. Reinforcement Active Learning (RAL) using Inception-Unet outperforms Inception-Unet models trained solely on labeled data (DL), achieving a score of 0.615. However, it falls slightly behind the performance of Inception-Unet models trained on fully labeled datasets (DF). Our findings indicate that reinforced learning excels over strategies in selecting samples and substantially boosts segmentation accuracy.