• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2012.10a
• /
• pp.31-32
• /
• 2012

• Safety and Health at Work
• /
• v.6 no.4
• /
• pp.268-278
• /
• 2015

Background: This study aimed to assess the whole-body vibration (WBV) exposure among large blast hole drill machine operators with regard to the International Organization for Standardization (ISO) recommended threshold values and its association with machine- and rock-related factors and workers' individual characteristics. Methods: The study population included 28 drill machine operators who had worked in four opencast iron ore mines in eastern India. The study protocol comprised the following: measurements of WBV exposure [frequency weighted root mean square (RMS) acceleration ($m/s^2$)], machine-related data (manufacturer of machine, age of machine, seat height, thickness, and rest height) collected from mine management offices, measurements of rock hardness, uniaxial compressive strength and density, and workers' characteristics via face-to-face interviews. Results: More than 90% of the operators were exposed to a higher level WBV than the ISO upper limit and only 3.6% between the lower and upper limits, mainly in the vertical axis. Bivariate correlations revealed that potential predictors of total WBV exposure were: machine manufacturer (r = 0.453, p = 0.015), age of drill (r = 0.533, p = 0.003), and hardness of rock (r = 0.561, p = 0.002). The stepwise multiple regression model revealed that the potential predictors are age of operator (regression coefficient ${\beta}=-0.052$, standard error SE = 0.023), manufacturer (${\beta}=1.093$, SE = 0.227), rock hardness (${\beta}=0.045$, SE = 0.018), uniaxial compressive strength (${\beta}=0.027$, SE = 0.009), and density (${\beta}=-1.135$, SE = 0.235). Conclusion: Prevention should include using appropriate machines to handle rock hardness, rock uniaxial compressive strength and density, and seat improvement using ergonomic approaches such as including a suspension system.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 1990.05a
• /
• pp.38-39
• /
• 1990

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.317-317
• /
• 2021

효율적인 물관리를 위한 댐 유입량 대한 연구는 필수적이다. 본 연구에서는 다양한 머신러닝 알고리즘을 통해 40년동안의 기상 및 댐 유입량 데이터를 이용하여 소양강댐 유입량을 예측하였으며, 그 중 고유량과 저유량예측에 적합한 알고리즘을 각각 선정하여 머신러닝 알고리즘을 결합한 CombML을 개발하였다. 의사 결정 트리 (DT), 멀티 레이어 퍼셉트론 (MLP), 랜덤 포레스트(RF), 그래디언트 부스팅 (GB), RNN-LSTM 및 CNN-LSTM 알고리즘이 사용되었으며, 그 중 가장 정확도가 높은 모형과 고유량이 아닌 경우에서 특별히 예측 정확도가 높은 모형을 결합하여 결합 머신러닝 알고리즘 (CombML)을 개발 및 평가하였다. 사용된 알고리즘 중 MLP가 NSE 0.812, RMSE 77.218 m3/s, MAE 29.034 m3/s, R 0.924, R2 0.817로 댐 유입량 예측에서 최상의 결과를 보여주었으며, 댐 유입량이 100 m3/s 이하인 경우 앙상블 모델 (RF, GB) 이 댐 유입 예측에서 MLP보다 더 나은 성능을 보였다. 따라서, 유입량이 100 m3/s 이상 시의 평균 일일 강수량인 16 mm를 기준으로 강수가 16mm 이하인 경우 앙상블 방법 (RF 및 GB)을 사용하고 강수가 16 mm 이상인 경우 MLP를 사용하여 댐 유입을 예측하기 위해 두 가지 복합 머신러닝(CombML) 모델 (RF_MLP 및 GB_MLP)을 개발하였다. 그 결과 RF_MLP에서 NSE 0.857, RMSE 68.417 m3/s, MAE 18.063 m3/s, R 0.927, R2 0.859, GB_MLP의 경우 NSE 0.829, RMSE 73.918 m3/s, MAE 18.093 m3/s, R 0.912, R2 0.831로 CombML이 댐 유입을 가장 정확하게 예측하는 것으로 평가되었다. 본 연구를 통해 하천 유황을 고려한 여러 머신러닝 알고리즘의 결합을 통한 유입량 예측 결과, 알고리즘 결합 시 예측 모형의 정확도가 개선되는 것이 확인되었으며, 이는 추후 효율적인 물관리에 이용될 수 있을 것으로 판단된다.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.371-382
• /
• 2019

Machine reading comprehension is the task of understanding a given context and finding the correct response in that context. A simple recurrent unit (SRU) is a model that solves the vanishing gradient problem in a recurrent neural network (RNN) using a neural gate, such as a gated recurrent unit (GRU) and long short-term memory (LSTM); moreover, it removes the previous hidden state from the input gate to improve the speed compared to GRU and LSTM. A self-matching network, used in R-Net, can have a similar effect to coreference resolution because the self-matching network can obtain context information of a similar meaning by calculating the attention weight for its own RNN sequence. In this paper, we construct a dataset for Korean machine reading comprehension and propose an $S^2-Net$ model that adds a self-matching layer to an encoder RNN using multilayer SRU. The experimental results show that the proposed $S^2-Net$ model has performance of single 68.82% EM and 81.25% F1, and ensemble 70.81% EM, 82.48% F1 in the Korean machine reading comprehension test dataset, and has single 71.30% EM and 80.37% F1 and ensemble 73.29% EM and 81.54% F1 performance in the SQuAD dev dataset.

• Electronics and Telecommunications Trends
• /
• v.37 no.5
• /
• pp.44-53
• /
• 2022

The Internet of Things (IoT) is growing exponentially, with the number of IoT devices multiplying annually. Accordingly, the paradigm is changing from cloud computing to edge computing and even tiny edge computing because of the low latency and cost reduction. Machine learning is also shifting its role from the cloud to edge or tiny edge according to the paradigm shift. However, the fragmented and resource-constrained features of IoT devices have limited the development of artificial intelligence applications. Edge MLaaS (Machine Learning as-a-Service) has been studied to easily and quickly adopt machine learning to products and overcome the device limitations. This paper briefly summarizes what Edge MLaaS is and what element of research it requires.

• KIEE International Transaction on Electrical Machinery and Energy Conversion Systems
• /
• v.5B no.3
• /
• pp.215-223
• /
• 2005

An electrical machine for a high-speed flywheel for energy storage in large hybrid electric vehicles is described. Design choices for the machine are motivated: it is a radial-flux external-rotor permanent-magnet synchronous machine without slots in the stator iron and with a shielding cylinder. An analytical model of the machine is briefly introduced whereafter optimization of the machine is discussed. Three optimization criteria were chosen: (1) torque; (2) total stator losses and (3) induced eddy current loss on the rotor. The influence of the following optimization variables on these criteria is investigated: (1) permanent-magnet array; (2) winding distribution and (3) machine geometry. The paper shows that an analytical model of the machine is very useful in optimization.

• Vending industry
• /
• v.8 no.3
• /
• pp.115-115
• /
• 2008

• Journal of Korean Institute of Industrial Engineers
• /
• v.30 no.4
• /
• pp.306-316
• /
• 2004

We present a closed form approximate analytical model to estimate the expected waiting times for the storage and retrieval requests of an automated storage/retrieval (AS/R) system, assuming that the storage/retrieval (S/R) machine idles either at the rack or at the input/output point. The expected waiting times (and the associated mean queue lengths) can play an important role to decide whether the performance of a stable AS/R system is actually acceptable, to determine buffer size (or length) of the input conveyor, and to compute the number of the rack openings which is required to hold the loads which are requested by processing machines but waiting in the rack to be retrieved by the SIR machine. This model can be effectively used in the early design stage of an AS/R system.

• Journal of Korean Institute of Industrial Engineers
• /
• v.18 no.2
• /
• pp.23-29
• /
• 1992

To compute the expected travel time of storage and retrieval (S/R) machine in automated warehousing systems most of the previous studies assumed that equal number of rack openings are required regardless of the nature of storage assignment rules. It is known that randomized storage assignment rule usually needs less storage to space than needed for full turnover-based assignment rule. The objective of this paper is compute the expected travel time of each assignment rule more equitably by taking into account the storage space required for each rule. First, the rack storage space is determined which satisfies a given service level. Then based on the standard Economic Ordering Quantity model, trade-off analysis is carried out which relates the storage space to the expected travel time of the S/R machine. Finally, example problems are solved to compare the performance of each assignment rule under varying conditions of demand pattern and service level.