• 한국산업융합학회 논문집
• /
• 제21권5호
• /
• pp.207-217
• /
• 2018

Recently, the importance of energy demand management, particularly peak load control, has been increasing due to the policy changes of the Second Energy Basic Plan. Even though the installation of distributed generation systems such as Photovoltaic and energy storage systems (ESS) are encouraged, high initial installation costs make it difficult to expand their supply. In this study, the power consumption of a university building was measured in real time and the measured power consumption data was used to calculate the optimal installation capacity of the Photovoltaic and ESS, respectively. In order to calculate the optimal capacity, it is necessary to analyze the operation methods of the Photovoltaic and ESS while considering the KEPCO electricity billing system, power consumption patterns of the building, installation costs of the Photovoltaic and ESS, estimated savings on electric charges, and life time. In this study, the power consumption of the university building with a daily power consumption of approximately 200kWh and a peak power of approximately 20kW was measured per minute. An economic analysis conducted using these measured data showed that the optimal capacity was approximately 30kW for Photovoltaic and approximately 7kWh for ESS.

• 디지털융복합연구
• /
• 제19권7호
• /
• pp.199-208
• /
• 2021

기계번역 사후교정이란, 기계번역 문장에 포함된 오류를 자동으로 교정하기 위해 제안된 연구 분야이다. 이는 번역 시스템과 관계없이 번역문의 품질을 높이는 오류 교정 모델을 생성하는 목적을 가진 연구로, 훈련을 위해 소스문장, 번역문, 그리고 이를 사람이 직접 교정한 문장이 활용된다. 특히, 최신 기계번역 사후교정 연구에서는 사후교정 데이터를 통한 학습을 진행하기 이전에, 사전학습된 다국어 언어모델을 활용하는 방법이 적용되고 있다. 이에 본 논문은 최신 연구들에서 활용되고 있는 다국어 사전학습 언어모델들과 함께, 해당 모델을 도입한 각 연구에서의 구체적인 적용방법을 소개한다. 나아가 이를 기반으로, 번역 모델과 mBART모델을 활용하는 향후 연구 방향을 제안한다.

• Journal of information and communication convergence engineering
• /
• 제13권3호
• /
• pp.197-204
• /
• 2015

In this paper, we present an algorithm for the detection of illegally parked vehicles based on a combination of some image processing algorithms. A digital camera is fixed in the illegal parking region to capture the video frames. An adaptive Gaussian mixture model (GMM) is used for background subtraction in a complex environment to identify the regions of moving objects in our test video. Stationary objects are detected by using the pixel-level features in time sequences. A stationary vehicle is detected by using the local features of the object, and thus, information about illegally parked vehicles is successfully obtained. An automatic alarm system can be utilized according to the different regulations of different illegal parking regions. The results of this study obtained using a test video sequence of a real-time traffic scene show that the proposed method is effective.

• 융복합기술연구소 논문집
• /
• 제8권1호
• /
• pp.1-4
• /
• 2018

In this study, design and manufacture of LFR (Linerar Fresnel Reflector) system was performed for solar thermal absorption cooling. The LFR system was designed considering the expansion and convenience to be installed according to the cooling capacity of the applicable building. Twelve LFR modules with a total reflection area of $204m^2$ were installed. The automatic tracking system was applied to track the sun during the daytime.

• Korean Journal of Radiology
• /
• 제20권8호
• /
• pp.1275-1284
• /
• 2019

Objective: To develop algorithms using convolutional neural networks (CNNs) for automatic segmentation of acute ischemic lesions on diffusion-weighted imaging (DWI) and compare them with conventional algorithms, including a thresholding-based segmentation. Materials and Methods: Between September 2005 and August 2015, 429 patients presenting with acute cerebral ischemia (training:validation:test set = 246:89:94) were retrospectively enrolled in this study, which was performed under Institutional Review Board approval. Ground truth segmentations for acute ischemic lesions on DWI were manually drawn under the consensus of two expert radiologists. CNN algorithms were developed using two-dimensional U-Net with squeeze-and-excitation blocks (U-Net) and a DenseNet with squeeze-and-excitation blocks (DenseNet) with squeeze-and-excitation operations for automatic segmentation of acute ischemic lesions on DWI. The CNN algorithms were compared with conventional algorithms based on DWI and the apparent diffusion coefficient (ADC) signal intensity. The performances of the algorithms were assessed using the Dice index with 5-fold cross-validation. The Dice indices were analyzed according to infarct volumes (< 10 mL, ≥ 10 mL), number of infarcts (≤ 5, 6-10, ≥ 11), and b-value of 1000 (b1000) signal intensities (< 50, 50-100, > 100), time intervals to DWI, and DWI protocols. Results: The CNN algorithms were significantly superior to conventional algorithms (p < 0.001). Dice indices for the CNN algorithms were 0.85 for U-Net and DenseNet and 0.86 for an ensemble of U-Net and DenseNet, while the indices were 0.58 for ADC-b1000 and b1000-ADC and 0.52 for the commercial ADC algorithm. The Dice indices for small and large lesions, respectively, were 0.81 and 0.88 with U-Net, 0.80 and 0.88 with DenseNet, and 0.82 and 0.89 with the ensemble of U-Net and DenseNet. The CNN algorithms showed significant differences in Dice indices according to infarct volumes (p < 0.001). Conclusion: The CNN algorithm for automatic segmentation of acute ischemic lesions on DWI achieved Dice indices greater than or equal to 0.85 and showed superior performance to conventional algorithms.

• 한국기계가공학회지
• /
• 제20권3호
• /
• pp.40-46
• /
• 2021

An automatic transmission of automobiles enables comfortable driving experience with lower transmission shifting jerks. However, the assembly structure is more complicated and requires additional components with lower efficiency than the manual transmission system. Extensive research has been conducted to improve the overall transmission efficiency by optimizing each component of the automatic transmission assembly. This study focuses on enhancing the friction torque of double angular contact ball bearings used in automatic transmission. The friction torque of the bearing varies with the operating conditions such as the operational load and rotating speed. Since reducing the friction torque of the bearing tends to deteriorate the durability of the bearing, it is necessary to design the bearing having a minimum required friction torque by determining the durability life of an automatic transmission assembly, In this study, the theoretical life and friction torque of conventional and newly-developed bearings are calculated. The difference in the friction torque between the new and existing bearings are also evaluated.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2023년도 제68차 하계학술대회논문집 31권2호
• /
• pp.661-664
• /
• 2023

이 프로젝트의 목표는 공유 공간에서의 고객 이용 시간 관리를 향상하기 위해 라즈베리 파이와 센서 기술을 활용한 자동 시간 체크 시스템의 개발입니다. 이 시스템은 로드셀, 진동 감지 센서, 그리고 LiDAR 센서, 이 세 가지 센서를 활용해 의자에서 사용자의 존재 여부를 감지하고, 사람과 물건을 구별하며, 사용자가 의자에서 일어나는 시점을 파악합니다. 특히, 고객이 의자에 앉게 되면 시스템이 자동으로 시간을 체크하여 실시간으로 이용 시간을 측정하게 됩니다. 이렇게 수집된 정보는 웹 기반의 사용자 인터페이스를 통해 제공되어, 이용 시간 관리가 보다 편리해집니다.

• 정보처리학회논문지A
• /
• 제14A권1호
• /
• pp.73-76
• /
• 2007

단위원의 내부로부터 Jordan 영역으로의 등각사상을 구하는 것은 일반적으로 비선형방정식인 Theodorsen 방정식을 푸는 것으로 귀결된다. 저자는 이 비선형 방정식의 수치적 해법 중 가장 효율적인 방법으로 알려진 Wegmann의 해법에 저주파 필터를 적용하여 개선하고 새로운 산법의 수렴성을 이론적으로 증명한 바 있다[1, 2]. 또한 이 해법에 있어 참값을 모르더라도 오차평가가 가능한 방법을 제안하였다[3]. 본 논문에서는 참값을 모르더라도 오차평가가 가능한 연구결과를 이용하여 주어진 문제영역과 허용오차에 따라 자동으로 수치등각사상이 결정되는 알고리즘을 제안한다. 이 알고리즘에서는 지금까지 경험에 의존했었던 표본수와 저주파 필터 파라메터가 주어진 문제영역에 따라 자동으로 결정된다. 이것은 문제의 난이도가 문제영역의 변형에 의존한다는 전제로 문제영역의 모양을 결정하는 함수를 Fourier 급수로 전개, 분석하여 얻을 수 있다. 수치실험을 통해 그 유효성을 입증한다.

• 융합신호처리학회논문지
• /
• 제20권3호
• /
• pp.118-124
• /
• 2019

최근 산업현장에서는 생산성, 품질, 고객 만족도를 향상시키기 위해 정보통신기술(ICT)을 기반으로 한 스마트 팩토리 체제로 전환하고 있다. 스마트 팩토리를 실현함에 있어 가장 중요한 장치는 무인 반송차(AGV)이며, AGV의 도입이 증가하고 있다. 일반적으로 AGV은 범용 PLC를 이용하여 개발하고 있으나, 범용 PLC로 개발된 AGV의 가격은 고가이며, 부피 또한 크다. 한편, 산업현장에서는 작업장의 공간적 제약 때문에 소형화, 용이한 재구성 등이 가능한 저가의 AGV를 요구하고 있다. 따라서 본 연구에서는 이러한 문제점을 해결하기 위해 PLC 내장형 AGV 제어기의 설계법을 제안하고, 그 성능을 평가하였다. 그 결과, 우수한 속도제어와 주행 정밀도(속도제어 오차=0.021[m/s], 주행자세제어의 평균오차 = 2.1[mm])를 보였다. 이와 같이 제안된 AGV 제어기를 산업현장에 적용한다면, 저비용으로 소형화와 재구성 등이 가능할 것이다.

• International Journal of Railway
• /
• 제5권4호
• /
• pp.144-147
• /
• 2012

New transport technologies have been rapidly developed and various transport systems are ready to introduce. However, there no exists effective decision making tools for pre-feasibility studies on transport systems. The pre-feasibility studies should provide reliable solutions about whether a certain transport system is feasible or not by cost-benefit analysis and effectiveness test of enhancing regional economy. This study introduces the smart transport investment/information system (SMARTIS) for supporting decision making with ease, fast and reliable methods. The SMARTIS incorporates travel demand estimates into the process of feasibility studies with automatic data processing methods to give reliable solutions quickly. The SMARTIS is expected to be informative for transit agencies, planners, and operators.