• 소성∙가공
• /
• 제30권1호
• /
• pp.27-34
• /
• 2021

In this study, response surface method (RSM) was used in modeling and multi-objective optimization of the parameters of AA5052-H32 in incremental sheet forming (ISF). The goals of optimization were the maximum forming angle, minimum thickness reduction, and minimum surface roughness, with varying values in response to changes in production process parameters, such as tool diameter, tool spindle speed, step depth, and tool feed rate. A Box-Behnken experimental design (BBD) was used to develop an RSM model for modeling the variations in the forming angle, thickness reduction, and surface roughness in response to variations in process parameters. Subsequently, the RSM model was used as the fitness function for multi-objective optimization of the ISF process based on experimental design. The results showed that RSM can be effectively used to control the forming angle, thickness reduction, and surface roughness.

• 운동영양학회지
• /
• 제14권2호
• /
• pp.81-86
• /
• 2010

The purpose of this study was to identify the factors associated with child's Body Mass Index (BMI) and physical activity. The participants (n = 133) were Korean juveniles (3^rd and 4^th graders) and their parents. They completed a questionnaire packet including the SPARK (Sports, Play, and Active Recreation for Kids) survey and the parent equivalent survey. Correlation, multiple linear regression and binary logistic regression analyses were applied to identify the association between child's BMI and 10 factors of SPARK as predict or variables. 25.6% of the participants were classified as overweight (21.1%) or obesity (4.5%). 3 parental factors including mother's BMI and frequency of mother's and father's physical activity were identified as significant predictors of children's BMI. The 10 variables accounted for 28% of the variance (p<.01) in the linear regression model. These results provide insight into parental factors which are related to a child's BMI and physical activity. Parental role modeling which refers to parents' efforts to model an active lifestyle for children plays an important role.

• 디지털융복합연구
• /
• 제18권11호
• /
• pp.259-266
• /
• 2020

과거 인공지능 분야에서는 지식 기반의 전문가 시스템 및 머신러닝 알고리즘들을 금융 분야에 적용하는 연구가 꾸준하게 수행되어 왔다. 특히 주식에 대한 지식 기반의 시스템 트레이딩은 이제 보편화되었고, 최근에는 대용량 데이터에 기반한 딥러닝 기술을 주가 예측에 적용하기 시작했다. 이중 LSTM은 시계열 데이터에 대한 검증된 모델로서 주가 예측에도 적용되고 있다. 본 논문에서는 주가 예측 모델로서 LSTM을 적용할 때 성능향상을 위해 고려해야 할 복잡한 매개변수 설정과 적용 함수들에 대해 적합한 조합 방법을 제안하도록 한다. 크게 가중치와 바이어스에 대한 초기화 대상과 설정 방법, 과적합을 피하기 위한 정규화 적용 대상과 설정 방법, 활성화 함수 적용 방법, 최적화 알고리즘 선택 등을 제시한다. 이 때 나스닥 상장사들에 대한 대용량 데이터를 바탕으로 각각의 방법들을 적용하여 정확도를 비교하면서 평가한다. 이를 통해 주가 예측을 위한 LSTM 적용 시 최적의 모델링 방법을 실증적인 형태로 제안하여 현실적인 시사점을 갖도록 한다. 향후에는 입력 데이터의 포맷과 길이, 하이퍼파라미터들에 대한 성능평가를 추가 수행하여 주요 설정 항목들의 조합에 대한 일반화 연구를 수행하고자 한다.

• 한국산림과학회지
• /
• 제107권2호
• /
• pp.184-192
• /
• 2018

본 연구는 중부지역 일본잎갈나무의 수간중량 추정식을 개발하기 위하여 지역별 개체목의 지위와 직경을 고려하여 표준지를 선정한 뒤, 총 55본의 표본목을 선정하였다. 표본목의 실측자료를 이용하여 수피외생중량, 수피외건중량, 수피내 건중량 및 이용중량 등 네종류의 중량을 추정하기 위해 11개의 모형을 비교 분석하였다. 중량을 추정하기 위해 흉고직경을 변수로 이용하는 경우, 흉고직경과 수고를 이용하는 경우, 재적을 이용하는 경우의 3개 변수에 따라 모형을 구분하였다. 최적의 모형은 적합지수와 추정의 표준오차 및 잔차의 분포를 이용하여 모형의 이행능력을 비교하여 선정하였다. 그 결과, 1변수인 흉고직경을 이용하는 식은 $W=a+bD+c^D^2$이며 설명력은 90~92%를 나타냈고, 흉고직경과 수고를 이용한 2변수식은 $W=aD^bH^C$으로 97~98%의 설명력으로 2변수 모형이 1변수 모형 보다 높은 적합도를 보였다. 또한 전체수간에 대한 재적과 이용재적에 대한 식인 W = aV로 중량추정설명력이 98~99%으로 높게 나타났고 SEE도 7.7~17.5, CV(%)도 8.0~10.0으로 우세한 적합도를 보였다. 본 연구는 개체목의 중량정보 제공 및 임분단위의 중량 바이오매스추정식의 기초연구로 활용될 것으로 판단된다.

• 구강회복응용과학지
• /
• 제30권1호
• /
• pp.16-22
• /
• 2014

목적: 본 연구의 목적은 동일한 모형에서 제작된 지르코니아 single 코핑과 3-unit 구조물의 변연 및 내면 적합도를 평가하는 것이다. 연구 재료 및 방법: 상악 우측 제2소구치가 상실되고 상악 제1소구치와 제2대구치가 지대치로 형성된 에폭시 모형(epoxy model)을 제작한 후 레이저 스캐너를 이용하여 스캔 하였다. CAD 소프트웨어를 이용하여 지르코니아 구조물을 디자인하고 CAM을 사용하여 지르코니아 single 코핑과 3-unit 구조물을 제작 하였다. 지르코니아 single 코핑은 상악 제1소구치, 제2대구치에서 각각 10개씩 제작 되었고 지르코니아 3-unit 구조물도 10개가 제작 되었다. 적합도 측정을 위하여 제작된 실리콘 복제(silicone replica)를 4등분하고 각 면의 측정 위치(P1, P2, P3, P4, P5)에서 디지털 현미경을 사용하여 적합도를 측정하였다. 각 위치에서 측정된 불일치 값들은 변연 불일치(MG, P1), 축 불일치(AG, P2와 P3의 평균), 교합 불일치(OG, P4와 P5의 평균)의 세 범주로 분류하였다. 결과: Single 코핑에서 변연 불일치의 범위는 18.47 - 40.54 ${\mu}m$, 축 불일치의 범위는 39.73 - 73.61 ${\mu}m$, 교합 불일치의 범위는 116.90 - 134.69 ${\mu}m$이었다. 3-unit 구조물에서 변연 불일치의 범위는 45.95 - 87.44 ${\mu}m$이고, 축 불일치의 범위는 23.78 - 57.00 ${\mu}m$이고, 교합 불일치는 99.89 - 131.06 ${\mu}m$이었다. 결론: 본 연구의 결과에 따르면 single 코핑에 비해 3-unit 구조물의 변연 불일치가 증가하였지만 임상학적 수용 범위(- 120 ${\mu}m$) 내에 있었으며 3-unit 구조물의 불일치 증가를 방지하기 위해서 더욱 균질한 지르코니아 블록 사용과 소결공정들을 조절해야 할 것이다.

• 대한간호학회지
• /
• 제39권4호
• /
• pp.518-527
• /
• 2009

Purpose: The purpose of this study was to identify factors influencing quality of life in kidney transplant recipients and to understand the concrete pathway of influence and the power of each variable, so that integrated prediction model to promote the quality of life of kidney transplant recipients could be developed. Methods: The sample was composed of 218 patients in follow-up care after a kidney transplant in one of 4 university hospitals in the Honam area. A structured questionnaire was used and the collected data were analyzed for fitness, using the LISREL program. Results: This model was concise and extensive in predicting the quality of life of kidney transplant recipients. Conclusion: The research verified the factors influencing quality of life for kidney transplant recipients and it verified that direct factors such as perception of health state, compliance, self-efficacy, stress and indirect factors such as self-efficacy and social support can be important factors to predict the quality of life for recipients. Moreover, those variables represent 87% of variance in explaining quality of life in a prediction model so that the variables can be utilized to predict quality of life for kidney transplant recipients.

• 제어로봇시스템학회논문지
• /
• 제10권3호
• /
• pp.280-285
• /
• 2004

This paper is concerned with the modeling and identification of sequencing batch reactor (SBR) via genetic algorithm based polynomial neural network (GA-based PNN). The model describes a biological SBR used in the wastewater treatment process fur nitrogen removal. A conventional polynomial neural network (PNN) is applied to construct a predictive model of SBR process fur nitrogen removal before. But the performances of PNN depend strongly on the number of input variables available to the model, the number of input variables and type (order) of the polynomials to each node. They must be fixed by the designer in advance before the architecture is constructed. So the trial and error method must go with heavy computation burden and low efficiency. To alleviate these problems, we propose GA-based PNN. The order of the polynomial, the number of input variables, and the optimum input variables are encoded as a chromosome and fitness of each chromosome is computed. Simulation results have shown that the complex SBR process can be modeled reasonably well by the present scheme with a much simpler structure compared with the conventional PNN model.

• 대한간호학회지
• /
• 제41권3호
• /
• pp.364-373
• /
• 2011

Purpose: In this study the fitness of a path model for the relationship among biological risk disposition, sociocultural risk factors, self-control, parent-adolescent communication, and risk behavior in adolescents was examined. Methods: The participants were 387 adolescents. The data were analyzed with the PASW 18.0 and AMOS 18.0 programs. Results: Sociocultural risk factors, self-control, and parent-adolescent communication showed a direct effect on risk behavior for adolescents, while biological risk disposition and sociocultural risk factor showed an indirect effect on risk behavior for adolescents. The modified path model of adolescents' risk behavior was showed a good fit with the model ($X^2$/df=2.37, GFI=.95, AGFI=.92, RMSEA=.06 [.05

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권2호
• /
• pp.702-723
• /
• 2020

The application of machine learning (ML) in intrusion detection has attracted much attention with the rapid growth of information security threat. As an efficient multi-label classifier, kernel extreme learning machine (KELM) has been gradually used in intrusion detection system. However, the performance of KELM heavily relies on the kernel selection. In this paper, a novel multiple kernel extreme learning machine (MKELM) model combining the ReliefF with nature-inspired methods is proposed for intrusion detection. The MKELM is designed to estimate whether the attack is carried out and the ReliefF is used as a preprocessor of MKELM to select appropriate features. In addition, the nature-inspired methods whose fitness functions are defined based on the kernel alignment are employed to build the optimal composite kernel in the MKELM. The KDD99, NSL and Kyoto datasets are used to evaluate the performance of the model. The experimental results indicate that the optimal composite kernel function can be determined by using any heuristic optimization method, including PSO, GA, GWO, BA and DE. Since the filter-based feature selection method is combined with the multiple kernel learning approach independent of the classifier, the proposed model can have a good performance while saving a lot of training time.

• International Journal of Ocean System Engineering
• /
• 제2권4호
• /
• pp.233-238
• /
• 2012

In this study, we carried out a precedent investigation for an ultra-large container ship, which is expected to be a higher value-added vessel. We studied a preliminary optimized design technique for estimating the principal dimensions of an ultra-large container ship. Above all, we have developed optimized dimension estimation models to reduce the building costs and weight, using previous container ships in shipbuilding yards. We also applied a generalized estimation model to estimate the shipping service costs. A Genetic Algorithm, which utilized the RFR (required freight rate) of a container ship as a fitness value, was used in the optimization technique. We could handle uncertainties in the shipping service environment using a Monte-Carlo simulation. We used several processes to verify the estimated dimensions of an ultra-large container ship. We roughly determined the general arrangement of an ultra-large container ship up to 1500 TEU, the capacity check of loading containers, the weight estimation, and so on. Through these processes, we evaluated the possibility for the practical application of the preliminary design model.