• Nuclear Engineering and Technology
• /
• 제50권7호
• /
• pp.1154-1159
• /
• 2018

Precise prediction of the radiation interaction position in scintillators plays an important role in medical and industrial imaging systems. In this research, the incident position of the gamma rays was predicted precisely in a plastic rod scintillator by using attenuation technique and multilayer perceptron (MLP) neural network, for the first time. Also, this procedure was performed using nonlinear regression (NLR) method. The experimental setup is comprised of a plastic rod scintillator (BC400) coupled with two PMTs at two sides, a $^{60}Co$ gamma source and two counters that record count rates. Using two proposed techniques (ANN and NLR), the radiation interaction position was predicted in a plastic rod scintillator with a mean relative error percentage less than 4.6% and 14.6%, respectively. The mean absolute error was measured less than 2.5 and 5.5. The correlation coefficient was calculated 0.998 and 0.984, respectively. Also, the ANN technique was confirmed by leave-one-out (LOO) method with 1% error. These results presented the superiority of the ANN method in comparison with NLR and the other methods. The technique and set up used are simpler and faster than other the previous position sensitive detectors. Thus, the time, cost and shielding and electronics requirements are minimized and optimized.

• 한국소음진동공학회논문집
• /
• 제18권4호
• /
• pp.448-457
• /
• 2008

A principal components analysis (PCA) of the median-plane head-related impulse responses (HRIRs) in the CIPIC HRTF database reveals that the individual HRIRs in the median plane can be adequately reconstructed by a linear combination of 12 orthonormal basis functions. These basis functions can be used to model arbitrary median-plane HRIRs, which are not included in the process to obtain the basis functions. Memory size can be reduced up to 5-fold depending on the number of HRIRs to be modeled. To clarify whether these basis functions can be used to model other set of arbitrary median plane HRIRs, a numerical error analysis for modeling and a series of subjective listening tests were carried out using the measured and modeled HRIRs. The results showed that the set of individual HRIRs in the median plane, which were measured in our lab using different measurement conditions, techniques, and source positions, can be modeled with reasonable accuracy. All subjects, involved in the subjective listening test, reported not only the accurate vertical perception but also the front-back discrimination with the modeled HRIRs based on 12 basis functions.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.147-147
• /
• 2018

In this article, we use an open source software library: TensorFlow, developed for the purposes of conducting very complex machine learning and deep neural network applications. However, the system is general enough to be applicable in a wide variety of other domains as well. The proposed model based on a deep neural network model, LSTM (Long Short-Term Memory) to predict the river water level at Okcheon Station of the Guem River without utilization of rainfall - forecast information. For LSTM modeling, the input data is hourly water level data for 15 years from 2002 to 2016 at 4 stations includes 3 upstream stations (Sutong, Hotan, and Songcheon) and the forecasting-target station (Okcheon). The data are subdivided into three purposes: a training data set, a testing data set and a validation data set. The model was formulated to predict Okcheon Station water level for many cases from 3 hours to 12 hours of lead time. Although the model does not require many input data such as climate, geography, land-use for rainfall-runoff simulation, the prediction is very stable and reliable up to 9 hours of lead time with the Nash - Sutcliffe efficiency (NSE) is higher than 0.90 and the root mean square error (RMSE) is lower than 12cm. The result indicated that the method is able to produce the river water level time series and be applicable to the practical flood forecasting instead of hydrologic modeling approaches.

• 산경연구논집
• /
• 제11권1호
• /
• pp.39-47
• /
• 2020

Purpose: This study investigated the effects of valuation errors on the capital market through the earnings forecasting errors of financial analysts. As a follow-up to Jensen (2005)'s study, which argued of agency cost of overvaluation, it was intended to analyze the effect of valuation errors on the earnings forecasting behavior of financial analysts. We hypothesized that if the manager tried to explain to the market that their firms are overvalued, the analysts' earnings forecasting errors would decrease. Research design, data and methodology: To this end, the analysis period was set from 2011 to 2018 of KOSPI and KOSDAQ-listed markets. For overvaluation, the study methodology of Rhodes-Kropf, Robinson, and Viswanathan (2005) was measured. The earnings forecasting errors of the financial analyst was measured by the accuracy and bias. Results: Empirical analysis shows that the accuracy and bias of analysts' forecasting errors decrease as overvaluation increase. Second, the negative relationship showed no difference, depending on the size of the auditor. Third, the results have not changed sensitively according to the listed market. Conclusions: Our results indicated that the valuation error lowered the financial analyst earnings forecasting errors. Considering that the greater overvaluation, the higher the compensation and reputation of the manager, it can be interpreted that an active explanation of the market can promote the accuracy of the financial analyst's earnings forecasts. This study has the following contributions when compared to prior research. First, the impact of valuation errors on the capital market was analyzed for the domestic capital market. Second, while there has been no research between valuation error and earnings forecasting by financial analysts, the results of the study suggested that valuation errors reduce financial analyst's earnings forecasting errors. Third, valuation error induced lower the earnings forecasting error of the financial analyst. The greater the valuation error, the greater the management's effort to explain the market more actively. Considering that the greater the error in valuation, the higher the compensation and reputation of the manager, it can be interpreted that an active explanation of the market can promote the accuracy of the financial analyst's earnings forecasts.

• 대한기계학회논문집A
• /
• 제41권2호
• /
• pp.121-127
• /
• 2017

자기베어링이 장착된 고속 회전기계의 상용화 관점에서 볼 때, 반경방향 자기베어링의 기계적중심과 자기적 중심 간의 중심 오차를 관리하고 대응하는 것이 매우 중요하다. 중심 오차를 측정하는 기존의 방법은 부상 제어기에서 제어 명령의 불균형을 최소화하는 자기적 중심 위치를 실험적으로 찾는 것인데, 이는 조립 단계에서 사용할 수 없다. 본 논문에서는 회전축의 위치를 측정하는 변위센서와 베어링 코일에 전류를 공급하는 전원만으로 중심 오차를 추정할 수 있는 새로운 방법을 제시한다. 회전축의 위치와 코일 전류에 따른 자기력 모델을 기반으로 당기기 시험에서의 접촉각과 중심 오차 간의 관계를 정립하고, 시험을 통해 측정한 접촉각과 모델 기반 접촉력 각도 간의 차이를 최소화함으로써 중심 오차를 추정한다. 유한요소해석을 이용하여 방법을 수치적으로 검증하고, 실험을 통해 추정 방법의 가능성을 확 인하였다.

• 한국방사선학회논문지
• /
• 제3권3호
• /
• pp.5-11
• /
• 2009

영상유도방사선치료(image guided radiation therapy: IGRT) 시 환자를 1차적으로 skin marker를 이용하여 위치시키고 2차적으로 OBI(on board imager)를 이용하여 해부학적 위치를 확인 후 couch를 움직여 set up을 보정하게 되는데, 이때 발생하는 오차에 대한 평가를 하려고 한다. 치료계획시 $0^{\circ}$와 $270^{\circ}$방향의 DRR(digital reconstructed radiography) 영상과 OBI로 촬영한 영상을 2차원-2차원 정합(2D-2D matching)으로 비교하여 치료계획시 환자의 셋업과 치료시 환자의 셋업의 오차를 비교하였다. Head&Neck 및 Spinal cord와 같은 주요장기 부위의 치료에서는 치료때 마다 OBI에 의하여 셋업시 확인하였으며, Chest 및 Abdomen&Pelvic 는 일주일에 2~3회 확인하였다. 그려나 보정 값은 모두 OIS(oncology information system)에 기록하여 160명의 환자를 대상으로 각각 Head&Neck, Chest 및 Abdomen&Pelvic으로 나누어 피부 지표를 이용한 셋업의 정확성을 평가하였다. Head&Neck 환자의 평균 셋업 오차는 각각 AP, SI, RL 방향에서 $0.2{\pm}0.2cm$, $-0.1{\pm}0.1cm$, $-0.2{\pm}0.0cm$ 로 나타났으며, Chest의 경우 $-0.5{\pm}0.1cm$, $0.3{\pm}0.3cm$, $0.4{\pm}0.2cm$ 로 나타났고 Abdomen의 경우 $0.4{\pm}0.4cm$, $-0.5{\pm}0.1cm$, $-0.4{\pm}0.1cm$로 나타났다. Pelvic 의 경우 $0.5{\pm}0.3cm$, $0.8{\pm}0.4cm$, $-0.3{\pm}0.2cm$ 나타났다. Head&Neck 같은 강체 (rigid body)는 셋업 오차가 Chest 및 Abdomen 부위에 비하여 상대적으로 작게 나타났다. Chest에서는 횡축 방향의 오차가 컸으며, Abdomen&Pelvic 에서는 AP 방향의 오차가 크게 나타났다. Chest에서 횡축오차가 크게 나타난 이유는 환자 셋업시 환자 몸의 휘어짐에 기인한 것이며, Abdomen에서의 AP방향의 오차가 큰 이유는 환자의 호흡으로 인해 앞뒤 위치의 변화 때문으로 사료된다. 환자 셋업 시스템에서는 systematic error는 나타나지 않았다. OBI는 해부학적 위치를 확인하기 때문에 병소가 피부에 위치해 있을 경우 피부마커로 셋업을 하는 것이 정확할 것으로 생각된다. 2차원-2차원 정합은 3차원-3차원 정합과 비교하여 rolling 오차를 찾아내지 못하나 환자의 피폭이 적다는 장점이 있으며 셋업 확인 시간이 짧기 때문에 실제 임상에서는 2차원-2차원 정합이 유용하였다.

• 한국콘텐츠학회논문지
• /
• 제11권7호
• /
• pp.234-238
• /
• 2011

온보드영상장치(OBI)를 사용하고 있는 콘빔CT(CBCT)를 이용하여 인체 팬텀 자세 및 위치와 모의치료시 인체 팬텀 자세 및 위치를 비교하여 CBCT의 3D 타깃 위치의 유용성을 평가하고자 한다. 실제방사선 치료와 동일한 과정으로 모의 치료계획을 하기 위해서 인체 팬텀(The Rando Phantom) 을 set up 한다. 기준점에 놓인 인체팬텀에서 CBCT를 이용하여 평행이동 및 회전이동 하였다. 이때 얻어진 영상들의 위치 차이에 대한 평균 및 편차를 인체 팬텀의 실제 이동 값과 비교하였다. 실험은 10회씩 반복하여 오차의 표준 편차를 구하였다. CBCT로 획득한 영상과 모의치료 시 획득한 CT영상을 비교하는 3D/3D 매칭에서 평균 setup의 residual error의 평균 및 표준편차는 lateral $0.2{\pm}-0.2$ mm, longitudinal $0.4{\pm}0.3$ mm, vertical $-0.4{\pm}0.1$ mm 로 각각 0~4 mm의 범위 이내로 나타났다. 모의실험 된 회전 내용은 $0.4{\pm}0.2$ mm, $0.3{\pm}0.3$ mm, 그리고 $0.3{\pm}0.4$ mm이다. 회전에 의한 error는 $0{\sim}0.6^{\circ}$ 범위이다. 인체 팬텀을 이용한 CBCT 3D/3D 매칭은 모의 치료 시와 환자 치료 시 정확한 정합을 함으로써 error를 최소화 하였다.

• 한국정밀공학회지
• /
• 제27권10호
• /
• pp.52-60
• /
• 2010

It is generally important to get a precise position information for autonomous unmanned vehicle(AUV) to run safely. For getting the position of AUV, the GPS has been using to navigation in a vehicle. Though it is useful to finding a position, it is difficult to precisely control a trajectory of the AUV due to large measuring error which may reach over 10 meters. Therefore to apply AUV it needs to compensate for the error. This paper proposes a method to more precisely localize AUV using three low-cost differential global positioning systems (DGPS). The distance errors between each DGPS are minimized as using the least square method (LSM) and the Kalman filter to eliminate a Gaussian white noise. The selected DGPS is cheaper and easier to set up than the RTK-GPS. It is also more precise than the general GPS. The proposed method can compensate the relatively position error according to stationary and moving distance of the AUV. For evaluating the algorithm by simulation, the DGPS signal with the Gaussian white noise to any points is generated by the AR model and compared with the measurement signal. It is confirmed that the proposed method can effectively compensate the position error as comparing with the measurement signal. The compensated position signal can be used to localize and control the AUV in the road.

• 한국가스학회지
• /
• 제18권4호
• /
• pp.68-75
• /
• 2014

독성가스시설에서 대형사고로 인한 손실을 감소시키기 위해서는 인적요소를 제어하고, 평가하는 것이 필요하다. 기존의 안전성 평가는 기계적 결함이나 시스템의 위험이 주로 강조되어 왔으며, 인적오류를 관리하는 평가방법이 부족하여 본 연구에서는 독성가스시설의 사고관리 상황에서 인적오류를 구성하는 수행영향인자 분류체계를 제안하고자 하였다. 이를 위해 주요 수행영향인자 분류체계의 유형을 수집하여 검토와 분석을 한 후 4가지(작업자 특성, 시스템, 직무특성 및 직무환경)로 나누고, 독성가스시설의 상황적 특성과 직무특성 및 작업환경을 고려하여 수행영향인자를 선택하였다. 이로 부터 인적오류의 평가에 사용하기 적절한 수행영향인자 체계를 구축하고, 구축한 수행영향인자 분류체계를 독성가스시설에서 발생하였던 사고 사례분석을 통해 적용성을 검토하였다.

• 한국자동차공학회논문집
• /
• 제16권1호
• /
• pp.52-63
• /
• 2008

The side collision reconstruction algorithm is developed using three dimensional car crash analysis. Medium size passenger car is modeled for finite element analysis. Total 24 side collision configurations, four different speed and six different angle, are set up for making side collision database. Deformation index and degree index are built up for each collision case. Deformation index is a kind of deformation estimate averaging displacement of side door of crashed car from finite element analysis result. Angle index is constructed measuring deformed angle of crashing car. There are two kinds of angle index, one is measured at driver's side and the other is measured at passenger's side. Also a collision analysis information in side of cars is used for giving a basis for scientific and practical reason in a reconstruction of the car accident. The analysis program, LS-DYNA3D is utilized for finite element analysis program for a collision analysis. Those database are used for side collision reconstruction. Side collision reconstruction algorithm is developed, and applied to find the collision conditions before the accident occurs. Three example collision cases are tried to check the effectiveness of the algorithm. Deformation index and angle index is extracted for the case from the analysis result. Deformation index is compared to the established database, and estimated collision speed and angle are introduced by interpolation function. Angle index is used to select a specific collision condition from the several available conditions. The collision condition found by reconstruction algorithm shows good match with original condition within 10% error for speed and angle. As a result, the calculation from the reconstruction of the situation is reproducing the situation well. The performance in this study can be used in many ways for practical field using deformation index and degree index. Other different collision situations may be set up for extending the scope of this study in the future.