• Geomechanics and Engineering
• /
• 제31권6호
• /
• pp.545-556
• /
• 2022

Evaluation and optimization of tunnel wall convergence (TWC) plays a vital role in preventing potential problems during tunnel construction and utilization stage. When convergence occurs at a high rate, it can lead to significant problems such as reducing the advance rate and safety, which in turn increases operating costs. In order to design an effective solution, it is important to accurately predict the degree of TWC; this can reduce the level of concern and have a positive effect on the design. With the development of soft computing methods, the use of deep learning algorithms and neural networks in tunnel construction has expanded in recent years. The current study aims to employ the long-short-term memory (LSTM) deep neural network predictor model to predict the TWC, based on 550 data points of observed parameters developed by collecting required data from different tunnelling projects. Among the data collected during the pre-construction and construction phases of the project, 80% is randomly used to train the model and the rest is used to test the model. Several loss functions including root mean square error (RMSE) and coefficient of determination (R²) were used to assess the performance and precision of the applied method. The results of the proposed models indicate an acceptable and reliable accuracy. In fact, the results show that the predicted values are in good agreement with the observed actual data. The proposed model can be considered for use in similar ground and tunneling conditions. It is important to note that this work has the potential to reduce the tunneling uncertainties significantly and make deep learning a valuable tool for planning tunnels.

• Proceedings of the Korea Water Resources Association Conference
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.520-520
• /
• 2023

제주도 동부 중산간 지역은 화산암으로 구성된 지하지질로 인해 지하수위의 변동폭이 크고 변동양상이 복잡하여 인공신경망(Artificial Neural Network, ANN) 모델 등을 활용한 지하수위의 예측이 어렵다. ANN에 적용되는 활성화함수에 따라 지하수의 예측성능은 달라질 수 있으므로 활성화함수의 비교분석 후 적절한 활성화함수의 사용이 반드시 필요하다. 본 연구에서는 5개 활성화함수(sigmoid, hyperbolic tangent(tanh), Rectified Linear Unit(ReLU), Leaky Rectified Linear Unit(Leaky ReLU), Exponential Linear Unit(ELU))를 제주도 동부 중산간지역에 위치한 2개 지하수 관정에 대해 비교분석하여 최적 활성화함수 도출을 목표로 한다. 또한 최적 활성화함수를 활용한 ANN의 적용성을 평가하기 위해 최근 널리 사용되고 있는 순환신경망 모델인 Long Short-Term Memory(LSTM) 모델과 비교분석 하였다. 그 결과, 2개 관정 중 지하수위 변동폭이 상대적으로 큰 관정은 ELU 함수, 상대적으로 작은 관정은 Leaky ReLU 함수가 지하수위 예측에 적절하였다. 예측성능이 가장 낮은 활성화함수는 sigmoid 함수로 나타나 첨두 및 최저 지하수위 예측 시 사용을 지양해야 할 것으로 판단된다. 도출된 최적 활성화함수를 사용한 ANN-ELU 모델 및 ANN-Leaky ReLU 모델을 LSTM 모델과 비교분석한 결과 대등한 지하수위 예측성능을 나타내었다. 이것은 feed-forward 방식인 ANN 모델을 사용하더라도 적절한 활성화함수를 사용하면 최신 순환신경망과 대등한 결과를 도출하여 활용 가능성이 충분히 있다는 것을 의미한다. 마지막으로 LSTM 모델은 가장 적절한 예측성능을 나타내어 다양한 인공지능 모델의 예측성능 비교를 위한 기준이 되는 참고모델로 활용 가능하다. 본 연구에서 제시한 방법은 지하수위 예측과 더불어 하천수위 예측 등 다양한 시계열예측 및 분석연구에 유용하게 사용될 수 있다.

• Journal of the Korea Society for Simulation
• /
• 제19권4호
• /
• pp.169-178
• /
• 2010

Agent based simulation is an approach for the analysis of a system's long term behavior where the entities in the system behave independently by their own judgement and memory, but influence each other to cope with given environment. In this paper we developed an agent based simulation model for the analysis of behavioral mechanism of team formation. In the process of team formation members' mutual preference is an important factor although each member can join up with one's own will. Also a team performance can vary by the member's own experience. We implemented the developed model using Netlogo 4.1, and verified the model by simulation. From the simulation results we found that the model successfully performed necessary functions using behavioral rules, judgments, and evolutionary processes by memory. As a further study we will be able to apply the model for analyzing various ecological behavior of team formation.

• Acta Via Serica
• /
• 제8권1호
• /
• pp.1-30
• /
• 2023

Traditional Kurdish weavings are among the world's most ancient living textile traditions. One of the largest regional ethnic and linguistic groups, Kurds have inhabited a significant part of Western Asia for millennia. Historically, Kurdish territories were crisscrossed by old and important trade routes, including the Silk Roads. This led to the formation of some of the most significant Kurdish artistic and cultural traditions, including textiles, which influenced and were influenced by those of other non-Kurdish ethnic groups from Caucasia to Central Asia and beyond. One example of Kurdish carpet traditions born in the eighteenth century at the cross-sections of Safavid (1501-1736) urban carpets workshops and centuries-old indigenous Kurdish tribal/rural weaves is senneh gelim or sojaee. A finely flatwoven carpet that was exchanged regionally and internationally as a diplomatic gift and a highly prized commodity. Although in decline, senneh gelims continue to be made by Kurdish women weavers in their original birthplace Sanandaj, the provincial capital of Iranian Kurdistan to date. This study adopts an inter-disciplinary approach to present an image of senneh gelim and women gelim weavers, tracing the developmental trajectories of the craft from the eighteenth century to the present time by drawing on extant art-historical and social scientific studies along with primary ethnographic data collected in Iranian Kurdistan (2018-2019). It investigates the craft tradition's historical origin, various aspects such as techniques, materials, aesthetics, functions, and meanings, and how these transformed over time. Additionally, the paper looks at the social contexts of production, focusing on women carpet weavers and how their socioeconomic and cultural situation has formed senneh carpet production in the past and present and the implications for long-term preservation.

• The Bulletin of The Korean Astronomical Society
• /
• 제44권2호
• /
• pp.69.2-69.2
• /
• 2019

To construct a coronal force-free magnetic field, we must impose the boundary normal current density (or three components of magnetic field) as well as the boundary normal field at the photosphere as boundary conditions. The only method that is known to implement these boundary conditions exactly is the method devised by Grad and Rubin (1958). However, the Grad-Rubin method and all its variations (including the fluxon method) suffer from convergence problems. The magnetofrictional method and its variations are more robust than the Grad-Rubin method in that they at least produce a certain solution irrespective of whether the global solution is compatible with the imposed boundary conditions. More than often, the influence of the boundary conditions does not reach beyond one or two grid planes next to the boundary. We have found that the 2D solenoidal gauge condition for vector potentials allows us to implement the required boundary conditions easily and effectively. The 2D solenoidal condition is translated into one scalar function. Thus, we need two scalar functions to describe the magnetic field. This description is quite similar to the Chandrasekhar-Kendall representation, but there is a significant difference between them. In the latter, the toroidal field has both Laplacian and divergence terms while in ours, it has only a 2D Laplacian term. The toroidal current density is also expressed by a 2D Laplacian. Thus, the implementation of boundary normal field and current are straightforward and their effect can permeate through the whole computational domain. In this paper, we will give detailed math involved in this formulation and discuss possible lateral and top boundary conditions and their meanings.

• Asia Marketing Journal
• /
• 제14권4호
• /
• pp.23-33
• /
• 2013

This case focuses on WE CAN Cookies, a social enterprise in South Korea that was founded in 2001 with the support of the Korean Roman Catholic Church. WE CAN Cookies specializes in the making of high quality organic cookies. As a nonprofit organization that uses a labor force of mostly mentally disabled workers, the company faces many challenges that normal companies do not experience. The company had to initially overcome the social prejudice that the handicapped cannot make good cookies. Despite the religious background and social agenda of the company, it started making inroads as a cookie-making business only after its managers, including the nuns who run it began adopting modern management philosophies and practices. The WE CAN Cookies case illustrates three main marketing-related concepts: One, WE CAN Cookies is a good example of how social enterprises face a broader spectrum of challenges when compared to conventional profit-seeking enterprises. Two, WE CAN Cookies demonstrates that social enterprises need flexibility in formulating their business strategies. Even though WE CAN Cookies is subject to many constraints, as a social enterprise it can also take advantage of new opportunities for obtaining support from the government and from the private sector. Three, WE CAN Cookies shows that these types of operations need to create greater balance in their social and business competencies to ensure the long term viability. Social enterprises are certified by governments with the stated goal of improving the lives and the wellbeing of special interest group. As important as achieving these objectives are, social enterprises also must additionally be able to build their operational capabilities not only in manufacturing but also in functions such as marketing.

• Safety and Health at Work
• /
• 제14권3호
• /
• pp.303-308
• /
• 2023

Background: Occupational workers at altitudes are more prone to falls, leading to catastrophic outcomes. Acrophobia, height-related anxiety, and affected executive functions lead to postural instabilities, causing falls. This study investigated the effects of repeated virtual height exposure and training on cognitive processing and height-related anxiety. Methods: Twenty-eight healthy volunteers (age 20.48 ± 1.26 years; mass 69.52 ± 13.78 kg) were recruited and tested in seven virtual environments (VE) [ground (G), 2-story altitude (A1), 2-story edge (E1), 4-story altitude (A2), 4-story edge (E2), 6-story altitude (A3), and 6-story edge (E3)] over three days. At each VE, participants identified occupational hazards present in the VE and completed an Attitude Towards Heights Questionnaire (ATHQ) and a modified State-Trait Anxiety Inventory Questionnaire (mSTAIQ). The number of hazards identified and the ATHQ and mSTAIQ scores were analyzed using a 7 (VE; G, A1, A2, A3, E1, E2, E3) x 3 (DAY; DAY 1, DAY 2, DAY 3) factorial repeated measures analysis of variance. Results: The participants identified the lowest number of hazards at A3 and E3 VEs and on DAY 1 compared to other VEs and DAYs. ATHQ scores were lowest at G, A1, and E1 VEs. Conclusion: Cognitive processing is negatively affected by virtual altitudes, while it improves with short-term training. The features of virtual reality, such as higher involvement, engagement, and reliability, make it a better training tool to be considered in ergonomic settings. The findings of this study will provide insights into cognitive dual-tasking at altitude and its challenges, which will aid in minimizing occupational falls.

• Journal of the Korean Society for information Management
• /
• 제40권4호
• /
• pp.33-51
• /
• 2023

The main objective of this study is to explore the forward-looking role of a National Hangeul Museum Library that can actively respond to changes in the era and to propose development strategies for this purpose. To achieve this, the current functions, resources, materials, online and offline services, and the library's website were thoroughly analyzed. We conducted research on the operational status and best practices of libraries within museums, art galleries, and advanced libraries both domestically and internationally. Ultimately, we aimed to establish medium to long-term development strategies for the Korean library and derive step-by-step detailed implementation plans. The results of this study can serve as foundational data to help the National Hangeul Museum Library effectively fulfill its central role as a library related to the Korean language and culture.

• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.98-103
• /
• 2023

As technology advanced dramatically in the late 20th century, a new era of science arrived. The emerging era of scientific discovery, variously described as e-Science, cyberscience, and the fourth paradigm, uses technologies required for computation, data curation, analysis, and visualization. The emergence of the fourth research paradigm will have such a huge impact that it will shake the foundations of science, and will also have a huge impact on the role of data-information infrastructure. In the digital age, the roles of data-information professionals are becoming more diverse. As eScience emerges as a sustainable and growing part of research, data-information professionals and centeres are exploring new roles to address the issues that arise from new forms of research. The functions that data-information professionals and centeres can fundamentally provide in the e-Science area are data curation, preservation, access, and metadata. Basically, it involves discovering and using available technical infrastructure and tools, finding relevant data, establishing a data management plan, and developing tools to support research. A further advanced service is archiving and curating relevant data for long-term preservation and integration of datasets and providing curating and data management services as part of a data management plan. Adaptation and change to the new information environment of the 21st century require strong and future-responsive leadership. There is a strong need to effectively respond to future challenges by exploring the role and function of data-information professionals in the future environment. Understanding what types of data-information professionals and skills will be needed in the future is essential to developing the talent that will lead the transformation. The new values and roles of data-information professionals and centers for 21st century researchers in STEAM are discussed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권2호
• /
• pp.348-369
• /
• 2024

With the advancement of Industry 4.0 and Industrial Internet of Things (IIoT), manufacturing increasingly seeks automation and intelligence. Temperature and vibration monitoring are essential for machinery health. Traditional abnormal state detection methodologies often overlook the intricate frequency characteristics inherent in vibration time series and are susceptible to erroneously reconstructing temperature abnormalities due to the highly similar waveforms. To address these limitations, we introduce synergistic, end-to-end, unsupervised Frequency-Time Domain Memory-Enhanced Autoencoders (FTD-MAE) capable of identifying abnormalities in both temperature and vibration datasets. This model is adept at accommodating time series with variable frequency complexities and mitigates the risk of overgeneralization. Initially, the frequency domain encoder processes the spectrogram generated through Short-Time Fourier Transform (STFT), while the time domain encoder interprets the raw time series. This results in two disparate sets of latent representations. Subsequently, these are subjected to a memory mechanism and a limiting function, which numerically constrain each memory term. These processed terms are then amalgamated to create two unified, novel representations that the decoder leverages to produce reconstructed samples. Furthermore, the model employs Spectral Entropy to dynamically assess the frequency complexity of the time series, which, in turn, calibrates the weightage attributed to the loss functions of the individual branches, thereby generating definitive abnormal scores. Through extensive experiments, FTD-MAE achieved an average ACC and F1 of 0.9826 and 0.9808 on the CMHS and CWRU datasets, respectively. Compared to the best representative model, the ACC increased by 0.2114 and the F1 by 0.1876.