• Advances in Energy Research
• /
• 제6권2호
• /
• pp.91-101
• /
• 2019

This paper proposes an efficient data-driven approach to build models for predicting energy consumption in buildings. Data used in this research is collected by installing humidity and temperature sensors at different locations in a building. In addition to this, weather data from nearby weather station is also included in the dataset to study the impact of weather conditions on energy consumption. One of the main emphasize of this research is to make feature selection independent of domain knowledge. Therefore, to extract useful features from data, two different approaches are tested: one is feature selection through principal component analysis and second is relative importance-based feature selection in original domain. The regression model used in this research is gradient boosting regression and its optimal parameters are chosen through a two staged coarse-fine search approach. In order to evaluate the performance of model, different performance evaluation metrics like r2-score and root mean squared error are used. Results have shown that best performance is achieved, when relative importance-based feature selection is used with gradient boosting regressor. Results of proposed technique has also outperformed the results of support vector machines and neural network-based approaches tested on the same dataset.

• 한국콘텐츠학회논문지
• /
• 제21권11호
• /
• pp.135-144
• /
• 2021

최근 데이터 기반 의사결정 기술이 데이터 산업을 이끄는 핵심기술로 자리 잡고 있는바, 이를 위한 머신러닝 기술은 고품질의 학습데이터를 요구한다. 하지만 실세계 데이터는 다양한 이유에 의해 결측값이 포함되어 이로부터 생성된 학습된 모델의 성능을 떨어뜨린다. 이에 실세계에 존재하는 데이터로부터 고성능 학습 모델을 구축하기 위해서 학습데이터에 내재한 결측값을 자동 보간하는 기법이 활발히 연구되고 있다. 기존 머신러닝 기반 결측 데이터 보간 기법은 수치형 변수에만 적용되거나, 변수별로 개별적인 예측 모형을 만들기 때문에 매우 번거로운 작업을 수반하게 된다. 이에 본 논문은 수치형, 범주형 변수가 혼합된 데이터에 적용 가능한 데이터 보간 모델인 Denoising Self-Attention Network(DSAN)를 제안한다. DSAN은 셀프 어텐션과 디노이징 기법을 결합하여 견고한 특징 표현 벡터를 학습하고, 멀티태스크 러닝을 통해 다수개의 결측치 변수에 대한 보간 모델을 병렬적으로 생성할 수 있다. 제안 모델의 유효성을 검증하기 위해 다수개의 혼합형 학습 데이터에 대하여 임의로 결측 처리한 후 데이터 보간 실험을 수행한다. 원래 값과 보간 값 간의 오차와 보간된 데이터를 학습한 이진 분류 모델의 성능을 비교하여 제안 기법의 유효성을 입증한다.

• Nuclear Engineering and Technology
• /
• 제56권5호
• /
• pp.1887-1894
• /
• 2024

Fatigue strength of the structural materials of lead-cooled fast reactors (LFRs) and accelerator-driven systems (ADS) may be degraded in liquid metal (Lead or lead-bismuth eutectic (LBE)) environments. The fatigue crack growth (FCG) data of structural materials in liquid LBE are necessary for damage tolerance design, safety assessment and life management of key equipment. A novel monitoring system for fatigue crack length was designed on the compliance method and the monitor technology of crack opening displacement (COD) of CT specimens by the linear variable differential transformers (LVDT) system. It can be used to predict the crack length by monitoring the COD of CT specimens in harsh high-temperature liquid LBE using a LVDT system. The prediction accuracy of this system was verified by FCG experiments in room temperature air and liquid LBE at 150, 250 and 350 ℃. The first results obtained in the FCG test for T91 steel in liquid LBE at 350 ℃ are presented.

• 한국정보과학회논문지:정보통신
• /
• 제30권6호
• /
• pp.755-763
• /
• 2003

전파(propagation) 오류가 빈번한 무선 이동 네트워크에서는 전송 성능을 향상하기 위해 FEC(Forward Error Correction)알고리즘을 채택한다. 그러나 정적인 FEC방식은 연속적으로 변화하는 전파 오류율에 알맞은 정정 코드(check code)를 적용하지 못해 성능이 저하된다. 본 논문에서는 변화하는 무선 채널의 전파 오류율에 따라 FEC의 정정도를 알맞게 결정하는 링크 계층용 적응적 FEC기법인 FECA(FEC-level Adaptation)를 제안한다. FECA는 오류율이 높고, 오류율이 천천히 변화하는 무선 환경에 알맞은 알고리즘이다. 일례로 전파 간섭이 있는 환경에서 센서(sensor) 네트워크는 평균 오류율이 $10^{-6}$이상이며 오류율이 평균 수백 밀리초 이상 지속되는 것으로 관찰되었다. FECA는 분석적인 무선채널 시뮬레이션과 패킷 트레이스 기반(trace-driven) 시뮬레이션에서 정적 FEC 알고리즘에 비해 최대 15%이상 성능을 향상하였다.

• Wind and Structures
• /
• 제32권2호
• /
• pp.127-142
• /
• 2021

Wind load acting on a standalone structure is different from that acting on a similar structure which is surrounded by other structures in close proximity. The presence of other structures in the surrounding can change the wind flow regime around the principal structure and thus causing variation in wind loads compared to a standalone case. This variation on wind loads termed as interference effect depends on several factors like terrain category, geometry of the structure, orientation, wind incident angle, interfering distances etc., In the present study, a three building configuration is considered and the mean pressure coefficients on each face of principle building are determined in presence of two interfering buildings. Generally, wind loads on interfering buildings are determined from wind tunnel experiments. Computational fluid dynamic studies are being increasingly used to determine the wind loads recently. Whereas, wind tunnel tests are very expensive, the CFD simulation requires high computational cost and time. In this scenario, Artificial Neural Network (ANN) technique and Support Vector Regression (SVR) can be explored as alternative tools to study wind loads on structures. The present study uses these data-driven approaches to predict mean pressure coefficients on each face of principle building. Three typical arrangements of three building configuration viz. L shape, V shape and mirror of L shape arrangement are considered with varying interfering distances and wind incidence angles. Mean pressure coefficients (Cp mean) are predicted for 45 degrees wind incidence angle through ANN and SVR. Further, the critical faces of principal building, critical interfering distances and building arrangement which are more prone to wind loads are identified through this study. Among three types of building arrangements considered, a maximum of 3.9 times reduction in Cp mean values are noticed under Case B (V shape) building arrangement with 2.5B interfering distance. Effect of interfering distance and building arrangement on suction pressure on building faces has also been studied. Accordingly, Case C (mirror of L shape) building arrangement at a wind angle of 45º shows less suction pressure. Through this study, it was also observed that the increase of interfering distance may increase the suction pressure for all the cases of building configurations considered.

• 대한조선학회논문집
• /
• 제58권3호
• /
• pp.158-166
• /
• 2021

In this paper, a numerical analysis technique using a body force model is investigated to estimate the available net thrust of multi-pod-driven ice-breaking vessels under bollard pull and overload conditions. To employ the body force model in present flow simulations, drag and thrust components acting on the pod unit are calculated by using Propeller Open Water (POW) test data. The available net thrusts according to the direction of operation are evaluated in both bollard pull and overload conditions under deep water. The simulation results are compared with the model test data. The available net thrusts, calculated by the present analysis for ahead operating modes at 3~6 knots which are typical speeds of the target vessel in arctic field, are agreed well with the model test results. It is also found that the present result for astern operating mode appears approximately 6 % larger than the model test result. In addition, the available net thrusts are calculated under the both operating conditions accompanied by shallow water effects, and the main cause of the difference is studied. Based on the result of the present study, it is confirmed that the body force model can be applied to the performance evaluation of multi-pod propulsion system and the main engine selection in early design stage of the vessel.

• 구강회복응용과학지
• /
• 제34권3호
• /
• pp.196-207
• /
• 2018

임플란트의 장기적인 안정을 위해서는 이상적인 위치와 각도로 임플란트를 식립하는 것이 중요하며, 이를 위해서는 정확한 계획에 따른 정확한 수술과 보철이 중요하다. 본 증례에서는 치조골 흡수가 심하게 진행된 환자에서 CT data 및 진단왁스업을 스캔한 data를 이용하여 CT guided surgery를 시행하고, 단일구조 수복용 지르코니아와 CAD/CAM technique을 이용하여 전악 임플란트 고정성 보철로 수복하여 기능 및 심미적으로 만족할만한 결과를 얻었기에 이를 보고하고자 한다.

• 전기학회논문지P
• /
• 제57권3호
• /
• pp.338-342
• /
• 2008

Hybrid Electric Vehicle(HEV) is driven by an internal-combustion engine and an electric motor. It is a combination of an internal-combustion engine and several electrical equipments which use a high voltage battery, an electric motors, an inverter and others. But there is not any separate detailed enforcement regulations for high voltage electric appliances in the existing vehicle-related safety standards. So, test standards suggestion as well as test technique development need to be done for ensuring electrical safety, for an electric motor, a high voltage battery, a(n) inverter/converter and an electric power transmission units and other equipments to ensure the safety of high voltage electric appliances which is the HEV key electrical component. In this paper, We are to provide helpful data to support test technique development and test standard establishment for HEV design and electrical safety security by the following methods; by measuring the voltage, the electric current, and the frequency of HEV, by analyzing electrical characteristics of high voltage electric appliances, and by analyzing temperature characteristics of the electrical current among the analyzed electrical characteristics by thermal imagining cameras.

• 전자공학회논문지D
• /
• 제35D권12호
• /
• pp.1-11
• /
• 1998

고속펄스 전송을 위한 다중결합된 마이크로스트립 신호버스에서 제일 근접한 선로상에 유기되는 누화피크치를 최소화시키기 위해 선로자태를 합성시키는 방법을 제안하였다. 평행하게 배열된 데이터버스에서 구동선로로부터의 누화에너지는 가장 근접한 선로에 거의 집중되므로 반사파 제어를 위한 최적화기법을 통해 선로간의 평균간격을 늘리는 방법을 채택하였다. 입출력 파형예측을 위해 일반화된 S-행렬 기법을 적용하였으며 선로간의 평균간격을 늘리기 위해 합성된 다양한 형태의 비일정선로의 누화특성을 비교 분석하였다. 그 결과 펄스가 갖는 주파수 범위내에서 스펙트럼을 고르게 반사시키게 하는, dip을 갖는 쳬비셰프형 테이퍼가 입출력 파형의 보전성에 큰 영향을 주지 않는 범위내에서 누화피크치를 최소로 함을 볼 수 있었다.

• Smart Structures and Systems
• /
• 제11권1호
• /
• pp.19-34
• /
• 2013

The objective of this paper is to develop on-line system parameter estimation and damage detection technique from the response measurements through using the Recursive Covariance-Driven Stochastic Subspace identification (RSSI-COV) approach. To reduce the effect of noise on the results of identification, discussion on the pre-processing of data using recursive singular spectrum analysis (rSSA) is presented to remove the noise contaminant measurements so as to enhance the stability of data analysis. Through the application of rSSA-SSI-COV to the vibration measurement of bridge during scouring experiment, the ability of the proposed algorithm was proved to be robust to the noise perturbations and offers a very good online tracking capability. The accuracy and robustness offered by rSSA-SSI-COV provides a key to obtain the evidence of imminent bridge settlement and a very stable modal frequency tracking which makes it possible for early warning. The peak values of the identified $1^{st}$ mode shape slope ratio has shown to be a good indicator for damage location, meanwhile, the drastic movements of the peak of $2^{nd}$ mode slope ratio could be used as another feature to indicate imminent pier settlement.