• 문화기술의 융합
• /
• 제8권6호
• /
• pp.741-747
• /
• 2022

피로는 개인의 능력을 저하되게 하여 업무 수행을 어렵게 하며, 피로가 누적되면 집중력이 저하되어 안전사고를 초래할 가능성이 증가하게 된다. 피로에 대한 자각은 주관적이나, 실제 현장에서는 피로의 수준을 정량적으로 측정할 필요가 있다. 기존 연구에서 피로 수준은 다원적 피로 척도와 같은 주관적 평가에, 생체신호 분석 등의 객관적지표를 추가하여 전문가의 판단으로 측정하는 방식이 제안되었으나, 이러한 방법은 일상생활에서 실시간으로 피로도를 평가하기 어렵다. 본 논문은 현장에서 녹음한 음성 데이터를 이용하여 실시간으로 작업자의 피로 수준을 판정하는 피로도 분류 모델에 관한 연구이다. 현장에서 수집한 음성 데이터를 이용하여 로지스틱 분류, 서포트 벡터 머신, 랜덤 포레스트 등의 기계학습 모델을 학습시킨다. 성능을 평가한 결과, 정확도가 0.677 ~ 0.758로 우수한 성능을 보여주었고, 이 중에서 로지스틱 분류가 가장 우수한 성능을 나타냈다. 실험 결과로부터 음성신호를 이용하여 피로도를 분류하는 것이 가능하다는 것을 알 수 있다.

• 말소리와 음성과학
• /
• 제8권4호
• /
• pp.97-101
• /
• 2016

This paper presents a CART(Classification and Regression Tree)-based model to diagnose mental fatigue using speech technology. The parameters used in the model are the significant speech parameters highly correlated to the fatigue and questionnaire responses obtained before and after imposing the fatigue. It is shown from the experiments that the proposed model achieves classification accuracies of 96.67% and 98.33% using the speech parameters and questionnaire responses, respectively. This implies that the proposed model can be used as a tool to diagnose the mental fatigue, and that speech technology is useful to diagnose the fatigue.

• 한국컴퓨터정보학회논문지
• /
• 제27권6호
• /
• pp.43-52
• /
• 2022

사람이 느끼는 피로는 다양한 생체신호로부터 측정이 가능한 것으로 알려져 있으며, 기존 연구는 질병과 관련된 심각한 피로수준을 산정하는데 주된 목적을 두고 있다. 본 연구에서는 피실험자의 영상을 이용하여 딥러닝 기반의 영상 분석 기술을 적용, 피로 여부를 판단하기 위한 모델을 제안한다. 특히 화상 분석에서 통상적으로 사용되는 객체 인식, 요소 추출과 함께 영상 데이터의 시계열적 특성을 고려하여 방법론을 교차한 3개 분석모델을 제시했다. 다양한 피로상황에서 수집된 정면 얼굴 영상 데이터를 이용하여 제시된 모델을 실험하였으며, CNN 모델의 경우 0.67의 정확도로 피로 상태를 분류할 수 있어 영상 분석 기반의 피로 상태 분류가 유의미하다고 판단된다. 또한 모델별 학습 및 검증 절차 분석을 통해 영상 데이터 특성에 따른 모델 적용방안을 제시했다.

• The Korean Journal of Pain
• /
• 제32권2호
• /
• pp.120-128
• /
• 2019

Background: We aimed to explore the American College of Rheumatology (ACR) 1990 and 2011 fibromyalgia (FM) classification criteria's items and the components of Fibromyalgia Impact Questionnaire (FIQ) to identify features best discriminating FM features. Finally, we developed a combined FM diagnostic (C-FM) model using the FM's key features. Methods: The means and frequency on tender points (TPs), ACR 2011 components and FIQ items were calculated in the FM and non-FM (osteoarthritis [OA] and non-OA) patients. Then, two-step multiple logistic regression analysis was performed to order these variables according to their maximal statistical contribution in predicting group membership. Partial correlations assessed their unique contribution, and two-group discriminant analysis provided a classification table. Using receiver operator characteristic analyses, we determined the sensitivity and specificity of the final model. Results: A total of 172 patients with FM, 75 with OA and 21 with periarthritis or regional pain syndromes were enrolled. Two steps multiple logistic regression analysis identified 8 key features of FM which accounted for 64.8% of variance associated with FM group membership: lateral epicondyle TP with variance percentages (36.9%), neck pain (14.5%), fatigue (4.7%), insomnia (3%), upper back pain (2.2%), shoulder pain (1.5%), gluteal TP (1.2%), and FIQ fatigue (0.9%). The C-FM model demonstrated a 91.4% correct classification rate, 91.9% for sensitivity and 91.7% for specificity. Conclusions: The C-FM model can accurately detect FM patients among other pain disorders. Re-inclusion of TPs along with saving of FM main symptoms in the C-FM model is a unique feature of this model.

• Journal of Advanced Marine Engineering and Technology
• /
• 제39권9호
• /
• pp.903-910
• /
• 2015

본 연구의 목적은 유탄성 효과를 고려한 완전통계 피로해석 프로그램을 구축하기 위한 기초 연구로 강체 선체운동을 바탕으로 한 완전통계 피로해석 프로그램을 구축하는 것이다. 프로그램의 신뢰성을 확보하기 위해 두 가지 선종에 대해 선급 피로해석 결과와 비교하였고, 결과가 일치함을 알 수 있었다. 향후 유탄성 선체운동 결과를 반영하고 광대역 피로손상 모델을 도입하면 유탄성 효과를 고려한 완전통계 피로해석 프로그램을 개발할 수 있을 것으로 사료된다.

• 한국공작기계학회논문집
• /
• 제10권4호
• /
• pp.14-21
• /
• 2001

To estimate crack growth rate and cycle ratio uniquely, many investigators have developed various kinds of mechanical parameters and theories. But, thes have produced local solution space through single parameter. Neural Networks can perform patten classification using several input and output parameters. Fatigue damage model by neural networks was used to recognize the relation between da/dN/N/N(sub)f, and half-value breadth ratio B/Bo, fractal dimension D(sub)f, and fracture mechanical parameters in 2024-T3 aluminium alloy. Learned neural networks has ability to predict both crack growth rate da/dN and cycly ratio /N/N(sub)f within engineering estimated mean error(5%).

• Journal of Advanced Marine Engineering and Technology
• /
• 제21권5호
• /
• pp.542-548
• /
• 1997

In this study an evaluation method of fatigue strength of membrane type LNG tank is presented with FEM analysis and experimental approach of seam and raised edge welds. The study contains the following : l)FEM analysis of test specimens 2)Fatigue tests of seam and raised edge welds 3)Estimation of cumulative damage factor of the welds on the basis of safe life design concept complying with the rules of classification society 4)Review of the effect of mean stress on the fatigue strength 5)Modelling of fatigue life of the welds which is changeable by weld heights With the results obtained in this study, a model ${\Delta}{\delta}/h^2=0.13553\;{N_{f}}^{-0.3151}$ for seam and raised edge welds having a given weld height is proposed to be useful for designers and inspectors.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제6권4호
• /
• pp.1096-1110
• /
• 2014

Whipping/springing research started in the 50'ies. In the 60'ies inland water vessels design rules became stricter due to whipping/springing. The research during the 70-90'ies may be regarded as academic. In 2000 a large ore carrier was strengthened due to severe cracking from North Atlantic operation, and whipping/springing contributed to half of the fatigue damage. Measurement campaigns on blunt and slender vessels were initiated. A few blunt ships were designed to account for whipping/springing. Based on the measurements, the focus shifted from fatigue to extreme loading. In 2005 model tests of a 4,400 TEU container vessel included extreme whipping scenarios. In 2007 the 4400 TEU vessel MSC Napoli broke in two under similar conditions. In 2009 model tests of an 8,600 TEU container vessel container vessel included extreme whipping scenarios. In 2013 the 8,100 TEU vessel MOL COMFORT broke in two under similar conditions. Several classification societies have published voluntary guidelines, which have been used to include whipping/springing in the design of several container vessels. This paper covers results from model tests and full scale measurements used as background for the DNV Legacy guideline. Uncertainties are discussed and recommendations are given in order to obtain useful data. Whipping/springing is no longer academic.

• 드라이브 ㆍ 컨트롤
• /
• 제20권4호
• /
• pp.54-63
• /
• 2023

High and intermittent loading cycles induce fatigue damage to transmission components, resulting in premature gearbox failure. To identify gearbox defects, numerous vibration-based diagnostics techniques, using several artificial intelligence (AI) algorithms, have recently been presented. In this paper, an optimized deep belief network (DBN) model for gearbox problem diagnosis was designed based on time-frequency visual pattern identification. To optimize the hyperparameters of the model, a particle swarm optimization (PSO) approach was integrated into the DBN. The proposed model was tested on two gearbox datasets: a wind turbine gearbox and an experimental gearbox. The optimized DBN model demonstrated strong and robust performance in classification accuracy. In addition, the accuracy of the generated datasets was compared using traditional ML and DL algorithms. Furthermore, the proposed model was evaluated on different partitions of the dataset. The results showed that, even with a small amount of sample data, the optimized DBN model achieved high accuracy in diagnosis.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2000년도 춘계학술대회논문집 - 한국공작기계학회
• /
• pp.341-347
• /
• 2000

To estimate crack growth rate and cycle ratio uniquely, many investigators have developed various kinds of mechanical parameters and theories. But, these have produced local solution space through single parameter. Neural Networks can perform pattern classification using several input and output parameters. Fatigue damage model by neural networks was used to recognize the relation between da/dN N/Nf, and half-value breadth ratio B/BO0, fractal dimension Df and fracture mechanical parameters in 2024-T3 ability to predict both crack growth rate da/dN and cycle ratio N/Nf within engineering estimated mean error (5%).