• 한국산학기술학회논문지
• /
• 제22권2호
• /
• pp.714-720
• /
• 2021

머신 러닝 및 신호처리에 활용되고 있는 정보이론적 학습법(ITL, information theoretic learning)은 커널 사이즈(σ) 설정이 매우 민감한 어려움을 지닌다. ITL의 성능지표중 하나인 코렌트로피 함수를 최대화하는 성능지표에 대해, 기울기에 존재하는 1/σ2를 제거한 뒤 남은 커널 사이즈에 대해 적응적으로 조절하는 방법들이 연구되었다. 이 논문에서는, 1/σ2의 커널 사이즈가 실제 시스템의 민감성이나 불안정에 큰 역할을 하고 있으며 남은 부분에 존재하는 커널 사이즈에 대한 최적해는 오차의 절대값 근방에 수렴함에 따라 오히려 수렴 후 가중치 갱신을 멈추게 하는 부작용이 나타남을 밝혔다. 이에 적응적 커널 사이즈 조절 대신 적절한 상수를 선택하는 것이 보다 효과적이라는 것을 제안하였고, 실험결과에서 동일한 수렴 속도에 약 2dB 향상된 정상상태 MSE를 보였다. 제안한 방식을 더욱 열악한 다경로 채널환경에 적용하여 실험한 결과 4dB 이상의 성능향상을 보여 제안한 방식은 열악한 상황일수록 더욱 향상된 성능을 보임을 알 수 있다.

• 유기물자원화
• /
• 제30권4호
• /
• pp.131-139
• /
• 2022

본 연구는 니카라과 마나과 La Chureca 매립장의 온실가스 감축 사업을 하고자 매립장 반입현황조사, 폐기물 성상조사 및 온실가스 배출량 산정을 하였다. 반입량과 성상조사를 바탕으로 한 IPCC 모형을 통한 온실가스 배출량 산정 결과 2006년부터 2043년까지 연평균 290,147 ton-CO₂/year였으며, IPCC 모형의 불확도를 고려하여 보수적으로 산정한 배출량은 217,610 ton-CO₂/year로 나타났다. La Chureca 매립장에서 온실가스 포집 가능량을 모형 불확도를 고려하고 포집 효율을 보수적으로 산정했음에도 CDM에 등록된 다른 중미 사례의 중간값과 평균값을 상회했으며 오차 요인에 대한 민감도 분석에도 결과가 크게 다르지 않았다. 본 연구는 La Chureca 매립장 온실가스 감축 사업의 타당성을 평가하고 이행 방안을 도출하기 위한 온실가스 배출량에 대한 기초자료로 활용이 가능할 것으로 판단된다.

• 대기
• /
• 제31권1호
• /
• pp.85-100
• /
• 2021

The necessity of accurate high-resolution meteorological forecasts becomes increasing in socio-economical applications and disaster risk management. The Korea Meteorological Administration Post-Processing (KMAPP) system has been operated to provide high-resolution meteorological forecasts of 100 m over the South Korea region. This study evaluates and improves the KMAPP performance in simulating wind speeds over complex terrain areas using the ICE-POP 2018 field campaign measurements. The mountainous measurements give a unique opportunity to evaluate the operational wind speed forecasts over the complex terrain area. The one-month wintertime forecasts revealed that the operational Local Data Assimilation and Prediction System (LDAPS) has systematic errors over the complex mountainous area, especially in deep valley areas, due to the orographic smoothing effect. The KMAPP reproduced the orographic height variation over the complex terrain area but failed to reduce the wind speed forecast errors of the LDAPS model. It even showed unreasonable values (~0.1 m s-1) for deep valley sites due to topographic overcorrection. The model's static parameters have been revised and applied to the KMAPP-Wind system, developed newly in this study, to represent the local topographic characteristics better over the region. Besides, sensitivity tests were conducted to investigate the effects of the model's physical correction methods. The KMAPP-Wind system showed better performance in predicting near-surface wind speed during the ICE-POP period than the original KMAPP version, reducing the forecast error by 21.2%. It suggests that a realistic representation of the topographic parameters is a prerequisite for the physical downscaling of near-ground wind speed over complex terrain areas.

• 한국산학기술학회논문지
• /
• 제22권3호
• /
• pp.627-632
• /
• 2021

적응 신호 처리 및 머신 러닝 등에 활용되고 있는 정보 이론적 학습법(ITL, information theoretic learning)은 커널 사이즈(��) 설정이 성능에 큰 영향을 미친다. ITL 기반의 학습법의 하나인 코렌트로피 알고리듬은 충격성 잡음에 강인성과 채널 왜곡 보상 특성을 함께 지니고 있으나 커널 사이즈 선택에 매우 민감하거나 불안정한 특성도 지니고 있다. 이에, 이 논문에서는 기울기 분모에 나타나는 커널 사이즈의 세제곱이 미치는 민감성을 고려하고, 커널 사이즈의 미세 변동에 대한 오차 전력 변화율을 이용하여 커널 사이즈를 적응적으로 갱신하는 방법을 제안하여 코렌트로피 알고리듬에 적용하였다. 제안된 적응 커널 사이즈 추정 방법을 다중 경로 채널과 충격성 잡음 환경에 대해 실험하였다. 제안한 방식은 고정 커널사이즈의 기존 알고리듬에 비해 2배 빠른 수렴 속도를 나타냈고 초기 커널 사이즈 2.0 에서 6.0 에 대해 모두 적절히 수렴하는 능력을 보였다. 이에 초기 커널 사이즈 선택에 큰 여유도를 가지고 성능을 향상시킬 수 있음을 입증하였다.

• 지질공학
• /
• 제32권3호
• /
• pp.389-399
• /
• 2022

Observations on the influence of the fluid infiltration on the breakdown pressure during laboratory hydraulic fracturing tests, along with an analysis of the applicability of the breakdown pressure prediction for cylindrical samples using Quasi-static and Linear Elastic Fracture Mechanics approaches were carried out. These approaches consider fluid infiltration through the so-called radius of fluid infiltration or crack radius, a parameter that is not a material property. Two sets of tests under pressurization rate controlled and injection rate controlled tests were used to evaluate the applicability of these methods. The difficulty of the estimation of the radius of fluid infiltration was solved by back calculating this parameter from an initial set of tests, and later, the obtained relationships were used to predict breakdown pressures for a second set of tests. The results showed better predictions for the injection rate than for the pressurization rate tests, with average errors of 3.4% and 18.6%, respectively. The larger error was attributed to differences in the testing conditions for the pressurization rate tests, which had different applied vertical pressures. On the other hand, for the tests carried out under constant injection rate, the Linear Elastic Fracture Mechanics solution reported lower errors compared to the Quasi-static solution, with values of 3% and 3.8%, respectively. Moreover, a sensitivity analysis illustrated the influence of the radius of fluid penetration or crack radius and the tensile strength on the breakdown pressure, suggesting a need for a careful estimation of these values. Then, the calculation of breakdown pressure considering fluid infiltration in cylindrical samples under triaxial conditions is possible, although larger data sets are desirable to validate and derive better relations.

• Geomechanics and Engineering
• /
• 제28권6호
• /
• pp.599-611
• /
• 2022

Tendon reinforced cemented soil is applied extensively in foundation stabilisation and improvement, especially in areas with soft clay. To solve the deterioration problem led by steel corrosion, the glass fiber-reinforced polymer (GFRP) tendon is introduced to substitute the traditional steel tendon. The interface bond strength between the cemented soil matrix and GFRP tendon demonstrates the outstanding mechanical property of this composite. However, the lack of research between the influence factors and bond strength hinders the application. To evaluate these factors, back propagation neural network (BPNN) is applied to predict the relationship between them and bond strength. Since adjusting BPNN parameters is time-consuming and laborious, the particle swarm optimisation (PSO) algorithm is proposed. This study evaluated the influence of water content, cement content, curing time, and slip distance on the bond performance of GFRP tendon-reinforced cemented soils (GTRCS). The results showed that the ultimate and residual bond strengths were both in positive proportion to cement content and negative to water content. The sample cured for 28 days with 30% water content and 50% cement content had the largest ultimate strength (3879.40 kPa). The PSO-BPNN model was tuned with 3 neurons in the input layer, 10 in the hidden layer, and 1 in the output layer. It showed outstanding performance on a large database comprising 405 testing results. Its higher correlation coefficient (0.908) and lower root-mean-square error (239.11 kPa) were obtained compared to multiple linear regression (MLR) and logistic regression (LR). In addition, a sensitivity analysis was applied to acquire the ranking of the input variables. The results illustrated that the cement content performed the strongest influence on bond strength, followed by the water content and slip displacement.

• Annals of Clinical Neurophysiology
• /
• 제24권1호
• /
• pp.7-16
• /
• 2022

Background: There is inadequate information on the validation of diabetic microvascular complications in the Korean National Health Insurance Service data set. We aimed to validate the diagnostic algorithms regarding the nephropathy, neuropathy, and retinopathy of diabetes. Methods: From various secondary and tertiary medical centers, we selected 6,493 patients aged ≥ 40 years who were diagnosed with diabetic microvascular complications more than once based on codes in the 10th version of the International Classification of Diseases (ICD-10). During 2019 and 2020, we randomly selected the diagnoses of 200 patients, 100 from each of two hospitals. The positive predictive value (PPV), negative predictive value, error rate, sensitivity, and specificity were determined for each diabetic microvascular complication according to the ICD-10 codes, laboratory findings, diagnostic studies, and treatment procedure codes. Results: Among the 200 patients who visited the hospital more than once and had the diagnostic codes of diabetic microvascular complications, 142, 110, and 154 patients were confirmed to have the gold standard of diabetic nephropathy (PPV, 71.0%), diabetic neuropathy (PPV, 55.0%), and diabetic retinopathy (PPV, 77.0%), respectively. The PPV and specificity of diabetic nephropathy (PPV, 71.0-81.4%; specificity, 10.3-53.4%), diabetic neuropathy (PPV, 55.0-81.3%; specificity, 66.7-76.7%) and diabetic retinopathy (PPV, 77.0-96.6%; specificity, 2.2-89.1%) increased after combining them with the laboratory findings, diagnostic studies, and treatment procedures codes. These change trends were observed similarly for both hospitals. Conclusions: Defining diabetic microvascular complications using ICD-10 codes and their related examination codes may be a feasible method for studying diabetic complications.

• 대한핵의학회지
• /
• 제22권1호
• /
• pp.39-45
• /
• 1988

To evaluate diagnostic accuracy of liver scintigraphy we analysed liver scans of 143 normal and 258 patients with various liver diseases. Three ROC curves for SOL, liver cirrhosis and diffuse liver disease were fitted using rating methods and areas under the ROC curves and their standard errors were calculated by the trapezoidal rule and the variance of the Wilcoxon statistic suggested by McNeil. We compared these results with that of National Institute of Radiological Science in Japan. 1) The sensitivity of liver scintigraphy was 74.2% in SOL, 71.8% in liver cirrhosis and 34.0% in diffuse liver disease. The specificity was 96.0% in SOL, 94.2% in liver cirrhosis and 87.6% in diffuse liver diasease. 2) ROC curves of SOL and liver cirrhosis approached the upper left-hand corner closer than that of diffuse liver disease. Area (${\pm}$ standard error). under the ROC curve was $0.868{\pm}0.024$ in SOL and $0.867{\pm}0.028$ in liver cirrhosis. These were significantly higher than $0.658{\pm}0.043$ in diffuse liver disease. 3) There was no interobserver difference in terms of ROC curves. But low sensitivty and high specificity of authors' SOL diagnosis suggested we used more strict decision threshold.

• 한국인터넷방송통신학회논문지
• /
• 제23권2호
• /
• pp.113-118
• /
• 2023

실내에서의 위치 추정을 위한 WiFi fingerprint 방식은 기존의 인프라를 이용하며 절대 좌표를 추정하는 장점이 있어 많은 연구가 진행되고 있다. 기존의 연구에서는 주로 위치 추정 알고리즘에 대한 연구에 집중되었지만 정확도를 개선하는 것이 한계에 도달했다. 그러나 스마트폰과 같은 무선랜 수신기에서 전파의 수신 감도보다 작은 신호는 측정이 불가하므로 이 값들을 처리하는 방법에 따라서 위치 추정 오차가 달라진다. 본 논문에서는 측정된 무선랜 공유기의 수신 신호 데이터를 다양한 방식으로 사전 처리하여 기존의 알고리즘에 적용함으로써 위치 추정 정확도를 높이는 방법을 제안하였고, 크게 향상된 정확도를 얻을 수 있었다. 또한 사전 처리된 데이터를 KNN 방식과 CNN 방식에 적용하여 그 성능을 비교하였다.

• Steel and Composite Structures
• /
• 제48권2호
• /
• pp.179-190
• /
• 2023

Many recent attempts have sought accurate prediction of pile pullout resistance (P_ul) using classical machine learning models. This study offers an improved methodology for this objective. Adaptive neuro-fuzzy inference system (ANFIS), as a popular predictor, is trained by a capable metaheuristic strategy, namely equilibrium optimizer (EO) to predict the P_ul. The used data is collected from laboratory investigations in previous literature. First, two optimal configurations of EO-ANFIS are selected after sensitivity analysis. They are next evaluated and compared with classical ANFIS and two neural-based models using well-accepted accuracy indicators. The results of all five models were in good agreement with laboratory P_uls (all correlations > 0.99). However, it was shown that both EO-ANFISs not only outperform neural benchmarks but also enjoy a higher accuracy compared to the classical version. Therefore, utilizing the EO is recommended for optimizing this predictive tool. Furthermore, a comparison between the selected EO-ANFISs, where one employs a larger population, revealed that the model with the population size of 75 is more efficient than 300. In this relation, root mean square error and the optimization time for the EO-ANFIS (75) were 19.6272 and 1715.8 seconds, respectively, while these values were 23.4038 and 9298.7 seconds for EO-ANFIS (300).