Search results

The purpose of this research is to achieve multi-task autonomous driving by adjusting the network architecture of the model. Meanwhile, after achieving multi-task autonomous driving, the authors found that the trained neural network model performs poorly in untrained scenarios. Therefore, the authors proposed to improve the transfer efficiency of the model for new scenarios through transfer learning.

First, the authors achieved multi-task autonomous driving by training a model combining convolutional neural network and different structured long short-term memory (LSTM) layers. Second, the authors achieved fast transfer of neural network models in new scenarios by cross-model transfer learning. Finally, the authors combined data collection and data labeling to improve the efficiency of deep learning. Furthermore, the authors verified that the model has good robustness through light and shadow test.

This research achieved road tracking, real-time acceleration–deceleration, obstacle avoidance and left/right sign recognition. The model proposed by the authors (UniBiCLSTM) outperforms the existing models tested with model cars in terms of autonomous driving performance. Furthermore, the CMTL-UniBiCL-RL model trained by the authors through cross-model transfer learning improves the efficiency of model adaptation to new scenarios. Meanwhile, this research proposed an automatic data annotation method, which can save 1/4 of the time for deep learning.

This research provided novel solutions in the achievement of multi-task autonomous driving and neural network model scenario for transfer learning. The experiment was achieved on a single camera with an embedded chip and a scale model car, which is expected to simplify the hardware for autonomous driving.

Family policy is an area where policy transfer has garnered a lot of attention lately. A growing body of research demonstrates policymakers' interest in and willingness to adopt foreign family policies. However, previous studies have tended to neglect the second mechanism of policy transfer: resistance. This manuscript aims to address this research gap by exploring both the willingness and resistance to policy transfer in Czech and Korean childcare and leave policies.

This study employs a qualitative research design, incorporating structured expert interviews instrumental in in-depth thematic analysis.

The analysis shows that policymakers in both countries demonstrated interest and willingness to transfer family policies, albeit employing different strategies and to varying extents. Moreover, the two countries exhibited significant differences in resistance to family policy transfer, with resistance in the Czech Republic being more frequent and effective. Resistance is directed towards both forced and voluntary transfers, although it isn't always against transfers that require a paradigm change. Policy transfer and non-transfer can concurrently be perceived as threats.

The study concludes that integrating both policy transfer and resistance in the analyses helps to shed light on cross-national differences in family policy change and contributes to a more nuanced portrayal of the world of policy transfer in this policy field.

This study aims to examine the impact of knowledge and skills acquisition from youth leadership programs on the extent of training transfer. Additionally, it explores the role of self-esteem as a potential intervening mechanism in linking the acquired knowledge and skills to the transfer of training.

Using a non-probability purposive sampling, data were gathered from participants of youth leadership programs in Brunei. The study hypotheses were validated using multiple linear regression analysis and Hayes PROCESS macro.

The findings revealed that knowledge and skills acquisition positively affect the extent of training transfer among youth leadership program participants. Moreover, self-esteem is found to be a crucial mediator in the relationship between knowledge and skills acquired from youth leadership programs and the extent of training transfer.

The study suggests that to optimize the transfer of training in the youth leadership programs, relevant stakeholders – training providers, and practitioners – must prioritize not only the knowledge and skills acquisition but also the cultivation of participants’ self-esteem. To achieve this, a primary focus should be placed on the criticality of designing such programs to address these factors. Policymakers, particularly in Brunei, can enhance the leadership pipeline among the youth population and expedite progress toward achieving the national vision by aligning leadership development initiatives with the broader national development agenda.

Collectively, this study enhances understanding of training transfer in youth leadership development, an often-overlooked area in the literature.

The purpose of this paper is to explore the geometric parameter difference of the terrace-like structural transfer film under different working parameters [pressure and velocity (PV) values] and filled particle types (three fillers: SiO₂, TiO₂ and ZnO), and find the geometric parameter related to the wear of polytetrafluoroethylene (PTFE)-based composites.

PTFE composites were filled with SiO₂, TiO₂ and ZnO particles, and the morphology parameter of the PTFE composite transfer film under different PV values obtained from the rotary reciprocating pin-on-disk frictional tester was quantified by using a three-dimensional laser scanning microscope.

The results showed that the effective layer coverage rate and effective thickness of the transfer film had a good relationship with the wear of the three PTFE composites. On the whole, increasing the speed or load was helpful to increase the effective thickness of the three PTFE composite transfer films, but reduced the effective layer coverage rate. The greater the effective layer coverage rate and effective thickness of the transfer film, the better the wear resistance of the PTFE composites in the entire speed and load range.

This work will promote further understanding of the transfer film and lay a foundation for realizing its morphology regulation and improving the wear of the PTFE composites.

This paper aims to investigate how ball milling (BM) and load influence transfer film on counterbody and the correlation between transfer film and tribological properties of copper-based composites.

The copper-based mixed powders preprocessed by BM for different times were used to manufacture sintered materials. Specimens were tested by a custom pin-on-flat linear reciprocating tribometer and characterized prior and after tests by optical microscope, scanning electron microscope and energy-dispersive spectroscopy. Image J® and Taylor-hobson-6 surface roughness meter were used to quantify the coverage and thickness of the transfer film.

Main results show that an appropriate amount of BM time and applied load can contribute to the formation of the transfer film on counterbody and effectively improve the tribological properties of the copper-based material. The transfer film coverage is linearly related to the friction coefficient, thickness of transfer film and wear volume. As the transfer film coverage increases, the coefficient of friction decreases. As the thickness of the transfer film increases, the amount of wear increases.

This work intends to control and optimize the formation of transfer film, thereby helping improve the tribological properties of materials and providing a reference to guide the preparation of Cu-based composites with excellent tribological properties.

This paper aims to refine the mechanisms affecting the two-way technology spillover and carbon transfer interactions between supply chain enterprises, and to guide their reduction of carbon emissions.

This study formulates a supplier-led Stackelberg game model to explore the effects of the interactions between two-way technology spillover effects and carbon transfers in decentralized and centralized decision-making scenarios. The optimized Shapley value is introduced to coordinate across the supply chain and determine the overall profits lost in the decentralized scenario.

Emission reductions by the low-carbon manufacturer are negatively correlated with the carbon transfers. Vertical technology spillovers promote carbon reduction, whereas horizontal technology spillovers inhibit it. The vertical technology spillovers amplify the negative effects of the carbon transfers, whereas the horizontal technology spillovers alleviate these negative effects. When the vertical technology spillover effect is strong or the horizontal technology spillover effect is weak in the centralized scenario, the carbon reduction is negatively correlated with the carbon transfers. Conversely, when the vertical technology spillover effect is weak or the horizontal technology spillover effect is strong, the enterprise’s carbon reduction is positively correlated with the carbon transfers. An optimized Shapley value can coordinate the supply chain.

This study examines the effects of carbon transfers on enterprises from a micro-perspective and distinguishes between vertical and horizontal technology spillovers to explore how carbon transfers and different types of technology spillovers affect enterprises’ decisions to reduce carbon emissions.

Despite the knowledge transfer between projects has received increasing attention from scholars, few scholars still conduct comprehensive research on inter-project knowledge transfer from both horizontal and vertical perspectives. Besides, knowledge transfer is affected by multiple antecedent conditions, and these factors should be combined for analysis. Therefore, this paper aims to explore the key factors influencing knowledge transfer between projects using the fuzzy-set qualitative comparative analysis (fsQCA) method from both horizontal and vertical perspectives and how these factors combine to improve the effectiveness of knowledge transfer (EKT) between projects.

First, nine factors affecting knowledge transfer between projects were identified, which were from the four dimensions of subject, relationship, channel, and context, namely temporary nature (TN), time urgency (TU), transmit willingness (TW), receive willingness (RW), trust (TR), project-project transfer channels (PPC), project-enterprise transfer channels (PEC), organizational atmosphere (OA), and motivation system (MS). Then, the source of the samples was determined and the data from the respondents was collected for analysis. Following the operation steps of the fsQCA method, variable calibration, single condition necessity analysis, and configuration analysis were carried out. After that, the configurations of influencing factors were obtained and the robustness test was conducted.

The results of the fsQCA method show that there are five configurations that can obtain better EKT between projects. Configuration 3 (∼TN * ∼TU * TW * RW * TR * ∼PPC * PEC * MS) has the highest consistency, indicating that it has the highest degree of the explanatory variable subset. Configuration 1 (∼TN * ∼TU * TW * RW * PEC * OA * MS) has the highest coverage, meaning that this configuration can explain most cases. Also, the five configurations were divided into three types: vertical transfer, horizontal-vertical transfer, and channel-free transfer category.

Firstly, this study explores the key factors influencing knowledge transfer between projects from four dimensions, which presents the logical chain of influencing factors more clearly. Then, this study divided the five configurations obtained into three categories according to the transfer direction: vertical, horizontal-vertical, and channel-free transfer, which gives implications to focus on both horizontal knowledge transfer (HKT) and (VKT) when studying knowledge transfer between projects. Lastly, this study helps to realize the exploration of combined improvement strategies for EKT, thereby providing meaningful recommendations for enterprises and project teams to facilitate knowledge transfer between projects.

Supporting community college transfer students represents a critical strategy for broadening participation in STEM. In addition to being a racially diverse group, students who pursue STEM degrees by way of community college report frequent interests in graduate study and academic careers. Thus, supporting and expanding transfer students’ PhD interests can help to diversify the STEM professoriate. This study aims to identify the experiences that predict PhD interests among students who transferred into the computer science major from a community college.

Relying on longitudinal survey data from over 150 community college transfer students throughout their first year at their receiving four-year university, we used regression analysis to identify the post-transfer college experiences that predict early interest in PhDs.

We found that receiving information about PhDs from a professor strongly predicted PhD interest among transfer students. Relationships with other variables indicate that the provision of information about graduate school was more likely to occur for students who participated in undergraduate research experiences than for those participating in internships. Descriptive data document inequities in who has access to these types of experiences.

This paper provides new insight into how STEM departments can develop targeted efforts to ensure that information about PhD training is equitably available to all transfer students. Working to ensure that faculty equitably communicate with students about PhD opportunities may go a long way in countering potential deterrents among transfer students who may be interested in such pathways.

The purpose of this study is to numerically investigate the geometric influence of different corrugation profiles (rectangular, trapezoidal and triangular) of varying heights on the flow and the natural convection heat transfer process over isothermal plates.

This work is an extension and finalization of previous studies of the leading author. The numerical methodology was proposed and experimentally validated in previous studies. Using OpenFOAM® and other free and open-source numerical-computational tools, three-dimensional numerical models were built to simulate the flow and the natural convection heat transfer process over isothermal corrugation plates with variable and constant heights.

The influence of different geometric arrangements of corrugated plates on the flow and natural convection heat transfer over isothermal plates is investigated. The influence of the height ratio parameter, as well as the resulting concave and convex profiles, on the parameters average Nusselt number, corrected average Nusselt number and convective thermal efficiency gain, is analyzed. It is shown that the total convective heat transfer and the convective thermal efficiency gain increase with the increase of the height ratio. The numerical results confirm previous findings about the predominant effects on the predominant impact of increasing the heat transfer area on the thermal efficiency gain in corrugated surfaces, in contrast to the adverse effects caused on the flow. In corrugations with heights resulting in concave profiles, the geometry with triangular corrugations presented the highest total convection heat transfer, followed by trapezoidal and rectangular. For arrangements with the same area, it was demonstrated that corrugations of constant and variable height are approximately equivalent in terms of natural convection heat transfer.

The results allowed a better understanding of the flow characteristics and the natural convection heat transfer process over isothermal plates with corrugations of variable height. The advantages of the surfaces studied in terms of increasing convective thermal efficiency were demonstrated, with the potential to be used in cooling systems exclusively by natural convection (or with reduced dependence on forced convection cooling systems), including in technological applications of microelectronics, robotics, internet of things (IoT), artificial intelligence, information technology, industry 4.0, etc.

To the best of the authors’ knowledge, the results presented are new in the scientific literature. Unlike previous studies conducted by the leading author, this analysis specifically analyzed the natural convection phenomenon over plates with variable-height corrugations. The obtained results will contribute to projects to improve and optimize natural convection cooling systems.

In recent times, there has been a growing interest in buoyancy-induced heat transfer within confined enclosures due to its frequent occurrence in heat transfer processes across diverse engineering disciplines, including electronic cooling, solar technologies, nuclear reactor systems, heat exchangers and energy storage systems. Moreover, the reduction of entropy generation holds significant importance in engineering applications, as it contributes to enhancing thermal system performance. This study, a numerical investigation, aims to analyze entropy generation and natural convection flow in an inclined square enclosure filled with Ag–MgO/water and Ag–TiO2/water hybrid nanofluids under the influence of a magnetic field. The enclosure features heated slits along its bottom and left walls. Following the Boussinesq approximation, the convective flow arises from a horizontal temperature difference between the partially heated walls and the cold right wall.

The governing equations for laminar unsteady natural convection flow in a Newtonian, incompressible mixture is solved using a Marker-and-Cell-based finite difference method within a customized MATLAB code. The hybrid nanofluid’s effective thermal conductivity and viscosity are determined using spherical nanoparticle correlations.

The numerical investigations cover various parameters, including nanoparticle volume concentration, Hartmann number, Rayleigh number, heat source/sink effects and inclination angle. As the Hartmann and Rayleigh numbers increase, there is a significant enhancement in entropy generation. The average Nusselt number experiences a substantial increase at extremely high values of the Rayleigh number and inclination.

This numerical investigation explores advanced applications involving various combinations of influential parameters, different nanoparticles, enclosure inclinations and improved designs. The goal is to control fluid flow and enhance heat transfer rates to meet the demands of the Fourth Industrial Revolution.

In a 90° tilted enclosure, the addition of 5% hybrid nanoparticles to the base fluid resulted in a 17.139% increase in the heat transfer rate for Ag–MgO nanoparticles and a 16.4185% increase for Ag–TiO2 nanoparticles compared to the base fluid. It is observed that a 5% nanoparticle volume fraction results in an increased heat transfer rate, influenced by variations in both the Darcy and Rayleigh numbers. The study demonstrates that the Ag–MgO hybrid nanofluid exhibits superior heat transfer and fluid transport performance compared to the Ag–TiO2 hybrid nanofluid. The simulations pertain to the use of hybrid magnetic nanofluids in fuel cells, solar cavity receivers and the processing of electromagnetic nanomaterials in enclosed environments.