Huu Tho Nguyen

This research aims to develop a coffee bean type detection model using convolutional neural networks (CNN), leveraging a dataset of 14,525 images from 116 types of Indonesian coffee beans. Pre-processing steps including resizing, rescaling, and augmentation were applied to improve the dataset quality. The dataset was split into training, validation, and testing sets with proportions of 80%, 10%, and 10%, respectively. Two model development approaches were used: transfer learning with Inception V3 in two scenarios and a model built from scratch. The transfer learning Inception V3 model in scenario 1 achieved the best performance, with a test accuracy of 0.87 and optimal evaluation metrics across precision, recall, and F1-score. This model was fine-tuned using pretrained weights, allowing it to adapt effectively to the coffee bean dataset. The results highlight that transfer learning, especially with Inception V3, provides a robust method for classifying coffee beans, offering potential applications in the coffee industry for improving classification efficiency and accuracy. The study demonstrates how deep learning can enhance the objectivity and precision of coffee bean classification, contributing to greater consistency in product sorting and quality assessment.

Gender classification has become an important application in the fields of system automation and artificial intelligence, having important implications across various fields. The main challenge in this classification task is the variation in illumination that affects the quality of facial images. This study presents a method for identifying genders with Convolutional Neural Networks (CNNs). To address this issue, various preprocessing methods are applied, including Self Quotient Image (SQI), Histogram Equalization, Locally Tuned Inverse Sine Nonlinear (LTISN), Gamma Intensity Correction (GIC), and Difference of Gaussian (DoG), to stabilize the effects of illumination variations before the images are processed by the CNN. The CNN architecture used consists of 5 convolutional blocks and 2 fully connected blocks, which have proven effective in image recognition. The results of the study show that the model trained with the DoG method achieved an accuracy of 91.07%, making it the best preprocessing technique compared to other methods such as SQI and HE, which achieved accuracies of 90.39% and 88.76%, respectively. These findings demonstrate that the application of SQI in CNN can improve the accuracy of gender classification on facial images, providing better performance than previous methods. These findings are expected to serve as a foundation for further developments in facial image classification and its applications in various fields.

The integration of unmanned aerial vehicles (UAVs) in logistics operations has attracted significant attention due to their potential to revolutionize delivery processes. This abstract provides an overview of the trends and advancements in utilizing UAVs for logistics. The study explores the current research landscape, identifies key challenges, and offers insights into the future applications of UAVs in logistics. By analyzing a range of scholarly articles, this abstract aims to illuminate the evolving role of UAVs in optimizing delivery efficiency, reducing costs, and addressing logistical complexities. Additionally, the abstract highlights the necessity for further research to address emerging issues and maximize the benefits of UAV technology in the logistics sector.

To select an air purifier that fully meets the needs of consumers, this study was conducted to provide the most objective and effective suggestions in making decisions and evaluating different smart purifier options that suit the tastes of users. Such decision-making is complicated for non-professionals when there are countless large and small brands on the market today with many different prices, features, and incentives. In such complex situations, decision-making with TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) can be used to eliminate risk and to better represent the preferences of decision-makers. From modeling to using the Entropy method and the TOPSIS method, all are done to serve the ultimate goal of choosing a good air purifier that meets the needs of customers.

In this paper, we consider a multi-crane scheduling problem in rail stations because their operations directly influence the throughput of the rail stations. In particular, the job is not only assigned to cranes but also the job sequencing is implemented for each crane to minimize the makespan of cranes. A dual cycle of cranes is used to minimize the number of working cycles of cranes. The rail crane scheduling problems in this study are based on the movement of containers. We consider not only the gantry moves, but also the trolley moves as well as the re-handle cases are also included. A mathematical model of multi-crane scheduling is developed. The traditional and parallel simulated annealing (SA) are adapted to determine the optimal scheduling solutions. Numerical examples are conducted to evaluate the applicability of the proposed algorithms. Verification of the proposed parallel SA is done by comparing it to existing previous works. Results of numerical computation highlighted that the parallel SA algorithm outperformed the SA and gave better solutions than other considered algorithms.

Using hands-on experiments in the classroom is a highly effective way to deepen students' understanding of theoretical concepts. By researching and developing a practical and cost-effective experiment using a pitot tube, we can transform the teaching of fluid mechanics from a content-based approach to one that promotes competency and practical application. The research includes exploring pitot tube theory, developing and evaluating the experimental design, simulating flow through the designed pitot tube using computational fluid dynamics simulation software, and implementing and evaluating learner satisfaction. The feedback we received from the students showed high satisfaction with the practical equipment used in the classroom and confirmed that this hands-on approach helped them achieve their learning objectives.

The EHA (electro hydraulic actuator) has a notable advantage over conventional hydraulic actuators as it uses a closed-loop circuit, reducing the size and volume of oil, and eliminates pressure losses caused by valve orifices. However, accurate control performance of EHA is difficult to achieve using a traditional PID (proportional integral derivative) controller due to the strongly nonlinear, time-varying, and unknown dynamics of the system. Hence this paper seeks to address this problem by proposing a design of an intelligent controller for the EHA. The proposed adaptive fuzzy sliding mode controller (AFSMC) is developed as a hybrid of the adaptive, fuzzy logic, and sliding mode algorithms. To reduce costs and time, a virtual prototype approach is also proposed instead of experimentations to evaluate the performance of the proposed controller. The virtual model of the EHA is built in Amesime software, and then embedded into Matlab/Simulink where the AFSMC is developed and tested to obtain the position responses of the EHA. The results show that the AFSMC is highly successful and more efficient than the traditional PID at controlling the position of the piston accurately.

Most of the world’s energy requirements are still derived from natural resources. This will result in a catastrophic energy crisis with negative environmental consequences. The increased energy supply will result in greater consumption of non-renewable sources. The production of biomass producer gas (BPG) from biomass gasification has received significant attention as an alternative fuel due to the depletion of non-renewable resources. This experimental study aimed to determine the flame propagation, flame propagation speed, and chamber pressure trace of BPG at different equivalence ratios. Understanding the characteristics of the BPG’s combustion, finding lower greenhouse gas emissions of BPG, and minimizing the use of fossil fuels is necessary to mitigate these problems. Using the direct visualization technique, an optical constant volume combustion chamber (CVCC) was developed to measure combustion characteristics. Liquid petroleum gas (LPG) was used to compare the flame propagation speed in the CVCC calibration. In comparison to wood pellet (WP), coconut husk (CH), and palm kernel shell (PKS), the chamber peak pressure at ϕ equal to 1 of CH for the combustion of BPG was the lowest at 20.84 bar. At ϕ of 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, and 1.3, the chamber peak pressure of CH was discovered to be around 17.77, 18.12, 18.81, 20.84, 20.39, 17.25, and 16.37 bar, respectively. Compared to the other two types of BPG, CH produced the lowest emissions of CO₂ and CO at 2.03% and 0.022%, respectively. In conclusion, the CH had the lowest chamber peak pressure and emissions due to the lower heating value (LHV) being relatively lower.

Homogeneous charge compression ignition (HCCI) is considered an advanced combustion method for internal combustion engines that offers simultaneous reductions in oxides of nitrogen (NO_x) emissions and increased fuel efficiency. The present study examines the influence of intake air temperature (IAT) and premixed diesel fuel on fuel self-ignition characteristics in a light-duty compression ignition engine. Partial HCCI was achieved by port injection of the diesel fuel through air-assisted injection while sustaining direct diesel fuel injection into the cylinder for initiating combustion. The self-ignition of diesel fuel under such a set-up was studied with variations in premixed ratios (0–0.60) and inlet temperatures (40–100 °C) under a constant 1600 rpm engine speed with 20 Nm load. Variations in performance, emissions and combustion characteristics with premixed fuel and inlet air heating were analysed in comparison with those recorded without. Heat release rate profiles determined from recorded in-cylinder pressure depicted evident multiple-stage ignitions (up to three-stage ignition in several cases) in this study. Compared with the premixed ratio, the inlet air temperature had a greater effect on low-temperature reaction and HCCI combustion timing. Nonetheless, an increase in the premixed ratio was found to be influential in reducing nitric oxides emissions.