합성곱 신경망(CNN)을 활용한 항공 시스템의 이상 탐지 모델 연구

임현재,김태림,송종규,김범수 항공우주시스템공학회 2023 항공우주시스템공학회지 Vol.17 No.4

Recently, Urban Aircraft Mobility (UAM) has been attracting attention as a transportation system of the future, and small drones also play a role in various industries. The failure of various types of aviation systems can lead to crashes, which can result in significant property damage or loss of life. In the defense industry, where aviation systems are widely used, the failure of aviation systems can lead to mission failure. Therefore, this study proposes an anomaly detection model using deep learning technology to detect anomalies in aviation systems to improve the reliability of development and production, and prevent accidents during operation. As training and evaluating data sets, current data from aviation systems in an extremely low-temperature environment was utilized, and a deep learning network was implemented using the convolutional neural network, which is a deep learning technique that is commonly used for image recognition. In an extremely low-temperature environment, various types of failure occurred in the system’s internal sensors and components, and singular points in current data were observed. As a result of training and evaluating the model using current data in the case of system failure and normal, it was confirmed that the abnormality was detected with a recall of 98 % or more. 최근 미래의 운송시스템으로 도심교통항공(Urban Aircraft Mobility)이 주목받고 있으며 소형 드론도 다양한 산업에서 역할을 하고 있다. 다양한 종류의 항공 시스템 고장은 추락으로 막대한 재산 및 인명 피해로 이어질 수 있다. 항공 시스템이 많이 활용되는 무기체계에서도 고장은 임무 실패의 결과를 유발한다. 본 논문에서는 항공 시스템의 이상(Anomaly)을 탐지하여 개발 및 생산 간 시스템의 신뢰도를 높이고 운용 중 사고를 예방할 수 있도록 딥러닝 기술을 활용한 이상 탐지 모델을 연구했다. 모델 훈련 및 평가 데이터로 극저온 환경에서 시스템의 전류 데이터를 활용하였으며 이미지 인식에 많이 활용되는 딥러닝 기법 합성곱 신경망(CNN; Convolutional Neural Network)을 활용하여 딥러닝 네트워크를 구현했다. 시험 대상 시스템은 극저온 환경에서 다양한 형태의 고장이 유발되었고 전륫값의 특이점이 나타났다. 시스템 정상 및 고장 데이터를 활용하여 모델을 훈련 시키고 평가한 결과 98% 이상의 재현율(Recall)로 이상 탐지하는 것을 확인했다.

택사와 옥수수 수염 추출액이 개구리 피부를 통한 수분 이동에 미치는 영향

임현재 충남대학교 의과대학 지역사회의학연구소 1981 충남의대잡지 Vol.8 No.2

The effects of Alisma orientale S. and Zea mays L. water extract on the water uptke and urine flow was investigated in the intact frog. 86 frogs were divided into two groups : immersion and injection. Immersion-group was submerged in 100ml of diluted solution containing Alisma orientale S. extract(0.3%, 1g%) and Zea mays L. exract (0.3g%, 1g%), and 100ml of distilled water as a control. Injection-group was injected Alisma orientale S. extract (20mg, 60mg) and Zea mays L. extract(40mg, 120mg) into the dorsal lymphatic sac, and submerged in 100ml of distilled water. Body weight, urine flow and urinary concentration of chloride and osmolarity were measured for 1 hour interval during 3 hours under continuous oxygen supply at room temperature. The results obtained summerized as follows : 1. Relationships between urine flow rate and body weight difference of pre-and-post urine emptying were statistically significant in all cases. 2. Urine flow rates were significantly increased(P<0.05) for first 1 hour in groups injected Alisma orientale S. extract and groups immersed in Zea mays L. extract. 3. Increase rates of body weight (rate of water uptake) were increased significantly (P<0.05) for first 1 hour in groups injected Alisma orientale S. extract and groups immersed in Zea mays L. extract. 4. Urinary osmotic concentrations were inversely related to urine flow rate in all groups, and the excretion rates of osmolar particles were significantly increased(P<0.05) when urine flow level were high. 5. Urinary chloride concentration was increased for first 1 hour in the groups immersed in Alisma orientale S. extract. In the above results, it suggest that increased urine formation was due to the increased water uptake across the frog skin in both groups injected Alisma orientale S, extract and immersed Zea mays L. extract, and their mechanism of action was different.

일본의 인지증(치매) 고령자 부양가족의 서비스 이용과 심리적 변화과정

임현재 한국통합사례관리학회 2012 한국케어매니지먼트연구 Vol.7 No.-

본 연구는 일본의 인지증(치매) 고령자를 부양하고 있는 가족의 심리적 변화과정을 중심으로, 케어를 담당하는 가족의 주관적 관점에서 부양가족의 심리적 변화과정에 영향을 미치는 서비스의 검토를 목적으로 하였다. 그 결과 대부분의 가족부양자는 데이케어나 단기 입소 서비스와 같이 일시적으로 케어로부터 해방(解放) 가능한 레스파이트(respite) 서비스의 이용률이 높은 것으로 나타났다. 그리고 가족부양자의 심리적 변화과정은 불규칙적이나 <당황·부정>과 <혼란·노여움·거부>의 초기의 심리적 단계에서 현재는 <포기·납득>과 <수용>의 심리적 단계로 변화해 있었다. 또한 데이케어를 중심으로 한 레스파이트 서비스는 부양가족의 심리적 변화에 영향을 주고 있었다. 인지증(치매) 고령자의 부양가족은 데이케어를 중심으로 단기 입소 서비스와 방문 케어 서비스를 이용함으로써 심리적 변화에 균형이 잡히고 가족부양자의 이상적인 심리적 단계라고 할 수 있는 <수용>의 단계로 원활하게 이행(移行) 한다고 할 수 있다. This research is to clarify the use service and psychological change of family caregiver, focusing on the psychological change which arises in care process of family caregiver supporting elderly with dementia in Japan, to examine the use of service which has influenced the psychological change from subjective viewpoint. As a result, it shows that the percentage of the use of respite service, which enables to be released from care temporary like day care and short stay, was high in most of family caregiver. The psychological change of family caregiver although it is irregular, from the initial <embarrassment, negative> and <rejection, anger, confusion> to <divisible> and <acceptance> of current psychological step has changed. Also, use of service of respite had influenced the psychological change. In short, the psychological change of family caregiver supporting elderly with dementia is taking balance by using service as day care, shortstay and visit care, and changing to ideal psychological step of family caregiver so called <acceptance>.

Waist- to-Height Ratio as a Simple and Useful Indicator for Non-alcoholic Fatty Liver Disease in Korean Women

임현재,서민석,이혜리,심재용,강희택,이용제 대한비만학회 2016 The Korean journal of obesity Vol.25 No.1

Background: It has not been determined which obesity index might be most appropriate to predict nonalcoholic fatty liver disease in Asian populations. This study aimed to evaluate the usefulness of the waist-to-height ratio in assessing patients with nonalcoholic fatty liver disease and to identify the optimal cut-off values useful for predicting nonalcoholic fatty liver disease. Methods: Receiver operating characteristic curve analyses were conducted in order to assess the accuracy of the waist circumference, body mass index, and waist-to-height ratio for detecting nonalcoholic fatty liver disease among 616 women aged 20 years or older. To evaluate the optimal value of anthropometric indices, the Youden J-index (sensitivity+specificity-1) was used. Results: The area under the ROC curve of waist-to-height ratio was highest among anthropometric obesity indices as follows: 0.776 (0.731-0.822) for waist circumference, 0.775 (0.728-0.822) for body mass index, and 0.792 (0.748-0.836) for waist-to-height ratio, respectively. Using a waist-to-height ration cut-off value of 0.49, the sensitivity and specificity for detecting nonalcoholic fatty liver disease were 72.3 % and 74.7%, respectively. Conclusion: These results demonstrated that the waist-to-height ratio may be a better obesity index for identifying individuals at risk for nonalcoholic fatty liver disease in Korean women.

[P2-09] 미생물에 의한 인삼 ginsenoside의 aglycon으로의 전환

임현재,김혜미,전영은,윤성복,한승준,강일준 한국식품영양과학회 2008 한국식품영양과학회 학술대회발표집 Vol.2008 No.10

Changes in the incidence of cardiopulmonary resuscitation before and after implementation of the Life-Sustaining Treatment Decisions Act

임현재,조현우,오승영,류호걸,이한나 대한중환자의학회 2022 Acute and Critical Care Vol.37 No.2

Background: The Life-Sustaining Treatment (LST) Decisions Act allows withholding and withdrawal of LST, including cardiopulmonary resuscitation (CPR). In the present study, the incidence of CPR before and after implementation of the Act was compared.Methods: This was a retrospective review involving hospitalized patients who underwent CPR at a single center between February 2016 and January 2020 (pre-implementation period, February 2016 to January 2018; post-implementation period, February 2018 to January 2020). The primary outcome was monthly incidence of CPR per 1,000 admissions. The secondary outcomes were duration of CPR, return of spontaneous circulation (ROSC) rate, 24-hour survival rate, and survival-to-discharge rate. The study outcomes were compared before and after implementation of the Act.Results: A total of 867 patients who underwent CPR was included in the analysis. The incidence of CPR per 1,000 admissions showed no significant difference before and after implementation of the Act (3.02±0.68 vs. 2.81±0.75, P=0.255). The ROSC rate (67.20±0.11 vs. 70.99±0.12, P=0.008) and survival to discharge rate (20.24±0.09 vs. 22.40±0.12, P=0.029) were higher after implementation of the Act than before implementation.Conclusions: The incidence of CPR did not significantly change for 2 years after implementation of the Act. Further studies are needed to assess the changes in trends in the decisions of CPR and other LSTs in real-world practice.

플러터(Flutter)를 활용한 유도탄 시험세트 소프트웨어 설계

임현재,박주현,홍재연,김주원 사단법인 한국국방기술학회 2025 한국국방기술학회 논문지 Vol.7 No.2

본 논문은 유도탄 시험세트 소프트웨어 개발에 플러터를 도입하는 방안을 제안한다. 기존 MFC, WPF와 같은 Windows 기반 GUI 애플리케이션 개발의 한계점을 분석하고, 플러터의 장점인 높은 생산성, 크로스플랫폼 지원, 모던한 디자인 등을 소개한다. 또한 성능 실험을 통해 플러터의 성능 및 한계점을 확인했다. UI 로드가 있는 상태에서 플러터 애플리케이션은 50Hz까지 안정적인 통신 성능을 발휘했다. 본 연구는 유도탄 시험세트 소프트웨어의 개발 방식 및 품질 발전에 기여할 것으로 예상된다. This paper proposes a plan to adopt Flutter for the development of guided missile test set software. We analyze the limitations of developing Windows-based GUI applications such as MFC and WPF, and introduce the advantages of Flutter, such as high productivity, cross-platform support, and modern design. Also we check the performance and limitations of Flutter through performance experiments. The Flutter application showed stable communication performance up to 50Hz under the load of the UI. This research is expected to contribute to the development paradigm and quality of Missile System Test Set software.

복합 임베디드 시스템 시계열 데이터를 활용한 딥러닝 이상 탐지 방법 비교 연구

임현재,한성재,박주성,안기성,박주현 항공우주시스템공학회 2024 항공우주시스템공학회지 Vol.18 No.5

비행체 같은 복합 임베디드 시스템은 고장이 발생하면 심각한 위험을 초래할 수 있다. 본 논문에서는 복합 임베디드 시스템에서 출력되는 시계열 데이터 셋과 LSTM, 1차원 CNN과 같은 딥러닝 알고리즘을 활용하여 이상 탐지 모델을 생성하고 추론 결과를 비교했다. 그 결과 1차원 CNN 모델이 좋은 성능을 보였다. 이전 연구(합성곱 신경망을 활용한 항공 시스템의 이상 탐지 모델 연구)에서 생성한 2차원 CNN 모델의 추론 성능을 비교한 결과 정확도와 재현율은 2차원 CNN 모델이 높았지만, 추론 속도는 1차원 CNN 모델이 빨랐다. 실시간 이상 탐지가 필요한 복합 임베디드 시스템의 이상 탐지 모델에는 1차원 CNN 모델이 적합한 것으로 판단된다. Complex embedded systems such as aircraft can lead to serious hazards when failures occur. This paper presents an anomaly detection model using deep learning techniques such as LSTM and 1D CNN on time-series datasets generated from complex embedded systems and compares inference results. Results showed that the 1D CNN model outperformed the LSTM model. Compared with the inference performance of a two-dimensional CNN model created in a previous study (Anomaly Detections Model of Aviation System by CNN), the two-dimensional CNN model had higher accuracy and recall. However, the 1-dimensional CNN model had faster inference speed. We can conclude that the 1D CNN model is more suitable than the LSTM model for anomaly detection in complex embedded systems that require real-time anomaly detection.

최소 전압 주입 방법을 이용한 직류단 션트 저항 인버터의 과변조 연구

임현재(Hyunjae-Lim),백현준(Hyunjun-Baek),하주형(Juhyung-Ha),임민철(Mincheol-Yim),이근호(Geunho-Lee) 한국자동차공학회 2019 한국자동차공학회 부문종합 학술대회 Vol.2019 No.5

인공지능을 활용한 항공 시스템의 이상 탐지 필요성과 방법

임현재(Hyun-Jae Im) 항공우주시스템공학회 2023 항공우주시스템공학회 학술대회 발표집 Vol.2023 No.10