R.O.C Military Academy

The present study is numerically investigated the effects of successively various location and tilted angle of piezoelectric fan on the heat transfer performance of a uniform heated cylindrical surface by piezoelectric fan. Since the blade vibrates periodically, three dimensional laminar, pseudo-steady-state natural convection flow is assumed. The discrete structure of fluid domain adopts the dynamic meshing scheme to simulate the three-dimensional time-varying geometries of the vibrating blade. Thus, numerical analysis can define precisely the motion of piezoelectric fan along the length of the blade at each time step. The numerical results infer that the heat transfer augmentation by piezoelectric fan is mainly contributed with two factors: One is the scale of produced vortex region by the vibrated blade; the other is the direction of induced jet-like stream impinging to the heat transfer surface. To obtain the optimal heat transfer augmentation, a total of 12 piezoelectric fans with various blade orientations from the bottom section of the cylinder to the flat top surface are evaluated. The cases of P4 ( y / H = 0.625) and P2 ( y / H = 0.375) provide the best heat transfer augmentation ratios 1.92 and 1.41 for the fan tip dimensionless amplitude equal to 2 and 1, respectively. However, the worst heat transfer performances occur at the positions P12 and P11 for dimensionless amplitude equal to 2 and 1.

BackgroundDetecting liver tumors via computed tomography (CT) scans is a critical but labor-intensive task. Extensive expert annotations are needed to train effective machine learning models. This study presents an innovative approach that leverages federated learning in combination with a teacher‒student framework, an enhanced slice-aware network (SANet), and semisupervised learning (SSL) techniques to improve the CT-based liver tumor detection process while significantly reducing its labor and time costs.MethodsFederated learning enables collaborative model training to be performed across multiple institutions without sharing sensitive patient data, thus ensuring privacy and security. The teacher–student SANet framework takes advantage of both teacher and student models, with the teacher model providing reliable pseudolabels that guide the student model in a semisupervised manner. This method not only improves the accuracy of liver tumor detection but also reduces the dependence on extensively annotated datasets.ResultsThe proposed method was validated through simulation experiments conducted in four scenarios, and it demonstrated a model accuracy of 83%, which represents an improvement over the original locally trained models.ConclusionsThis study presents a promising method for enhancing the CT-based liver tumor detection while reducing the incurred labor and time costs by utilizing federated learning, the teacher-student SANet framework, and SSL techniques. Compared with previous approaches, the proposed method achieved a model accuracy of 83%, representing a significant improvement.Trial registrationNot applicable.

A 2.45-GHz rectifying antenna (rectenna) using a circularly polarized (CP) square-ring patch antenna with an RF to dc conversion circuit board and a transmitting CP antenna array are presented. Two pairs of square patches are added orthogonally on the four corners of the square-ring patch with an aperture-coupled feed to produce frequency perturbation in the WLAN band. For avoiding polarization mismatch and losses, a same sense of polarization antenna array is proposed and used as transmitting antenna at 2.45 GHz. A five-stage voltage doubler is used as RF to dc conversion rectifying circuit, and an LC impedance matching circuit between the receiving antenna and the rectifying circuit is also designed.

ABSTRACT Acceptance sampling plans represent an economical method of quality verification. One approach is the single sampling plan; however, scholars have developed more advanced sampling techniques to address its poor cost‐effectiveness, including the multiple‐dependent‐state sampling plan (MDSP) with lot disposition based on historical quality information and the repetitive group sampling plan (RGSP) with resampling. Some researchers have integrated these to design an efficient sampling scheme incorporating lot‐backtracking and resampling mechanisms: the modified RGSP (MRGSP). Studies have highlighted flaws in the MDSP and RGSP sampling mechanisms. The performance of MDSP diminishes with increased lot records, whereas the RGSP may risk infinite resampling. Although improvements have been proposed for these shortcomings, the MRGSP, integrating both mechanisms, still suffers from these drawbacks. We design an integrated sampling plan (ISP) featuring adaptable lot‐backtracking and resampling mechanisms based on the process capability index to address these deficiencies. The proposed ISP reduces the inspection sample size significantly to enhance the cost‐effectiveness of the sampling inspection compared to existing plans. Moreover, because the ISP involves numerous adjustable parameters, creating a complex optimization model, we develop a cloud‐calculation application to assist in obtaining the optimal plan design. Finally, we present a practical case to demonstrate the applicability of ISP.

This study presents an innovative, low-cost miniaturized tag antenna with an artificial substrate, optimized for use on the human body. Perfect impedance matching between the proposed antenna and the microchip was obtained by adjusting the gap's capacitive reactance. A design approach based on current distribution along the dipole antenna arms was applied to further reduce the antenna's size. As a result, the proposed antenna configuration not only achieved miniaturization but also enhanced radiation efficiency to 33.2 %, attained a gain of -1.68 dB, and reached a power transmission coefficient of approximately 99.4 %. Additionally, it provided an extended reading distance (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$&gt;\mathbf{1 2} \mathbf{\ m}$</tex>) and a wide impedance bandwidth of 30 MHz, spanning nearly the entire UHF RFID frequency range used in North America, regardless of the tag's placement on the human body.

In the field of industrial automation, the diversity of products and protocols poses significant challenges for system integration. OPC UA (Open Platform Communications Unified Architecture) is a standardized communication protocol that aims to unify and support interoperability among various systems and devices. This paper explores the applications and challenges of Open62541 in implementing OPC UA. It discusses the implementation of the server/client and publish/subscribe communication models within OPC UA. By comparing these communication models, this paper provides insights into their respective advantages and limitations. Ultimately, Open62541 offers a reliable and flexible solution for addressing system integration issues in industrial automation.

Eye movement data can be utilized to observe attention distribution, offering valuable insights into how flight cadets engage with their environment and their underlying cognitive processes. This study evaluates the impact of eye-tracking technology on simulated takeoff and landing flight training. The experiment involved two simulated flights, with eye-tracking data provided to the experimental group between sessions. Results indicated significant improvements in the experimental group’s flight performance, particularly in attention allocation to critical instruments, evidenced by decreased deviations in altitude, speed, and heading. This suggests that eye-tracking data can enhance training effectiveness by providing cadets with insights into their visual attention patterns, thereby improving decision-making and learning outcomes.

Military academy education is a part of university education. This study aims to explore the characteristics and achievements of the learning network for military academies students in Taiwan order to understand the main reasons for the differences and to identify special students. Moreover, this study analyzed the social network to explain the patterns of the students’ learning network. The findings are as follows. (1) An extensively connected learning network exists among the students of different classes. Students who comply with social norms have close relationships and interact frequently with each other. (2) Different cognition of social norms is the main reason leading to differences among students in their learning network and performance. Compliance with social norms is effectively conducive to the access and sharing of knowledge and improvement of learning performance. (3) The academic performances of those who have aloof relationships with their peers are polarized with either extreme good or extreme poor, and most of them comply with no notice or reflect the behavior of rejecting social norms. The academic contribution can support administrator as an important reference for planning education strategy.

Within the framework of sociocultural theories, this study investigated how affordances in various beyond the classroom settings supported the development of agency and English proficiency among advanced EFL learners. The participants were 10 Taiwanese college students, each with at least B2 proficiency according to the Common European Framework of Reference for Languages. Data were gathered through a qualitative approach encompassing semi-structured interviews, writing samples, and a 23-item questionnaire. A grounded theory approach was adopted to analyze and interpret the data. The findings revealed that contextual affordances were not always immediately accessible or apparent to learners and exhibited considerable variability across different locations, which could impede or facilitate the learners’ development of agency and English proficiency. The mediator played an indispensable role in guiding learners toward effectively utilizing these resources. Furthermore, the development of learners' English proficiency in beyond-the-classroom settings was intricately intertwined with that of their agency, both of which were deeply rooted in and shaped by sociocultural contexts. As learners enhanced their English proficiency through the mediator’s strategic engagement with contextual affordances, their agency concurrently flourished, prompting them to proactively seek additional affordances and explore further opportunities for autonomous decision-making and action.