• 한국컴퓨터정보학회논문지
• /
• 제22권3호
• /
• pp.139-145
• /
• 2017

The purpose of this study is to establish a predictive causal model of codependency that is a main issue of family problem on the base of Triple ABC-X model which is a kind of family stress model. For the purpose of this study, we reviewed the concept and characteristics of codependency, affecting factors of codependency, and then reviewed the basic concept and logic of Triple ABC-X Model as theoretical viewpoint for the purpose of establishing a predictive causal model of codependency. We established it through examining main variables of codependency from Triple ABC-X Model. Main ingredients of the predictive causal model include boundary ambiguity, internal working model, internal and external locus of control, self-regard, social support, individual maladjustment etc. We established a predictive model of codependency basing on logic inferences among the variables. This study is expected to be used basic data to introduce some implications and for hereafter research.

• 자동차안전학회지
• /
• 제5권2호
• /
• pp.32-38
• /
• 2013

This paper presents collision avoidance using model predictive control algorithm. A model predictive control algorithm determines lateral tire force and yaw moment and steering angle input and differential braking input is determined from lateral tire force and yaw moment. A constraint for model predictive control is designed for obstacle avoidance. A objective function is designed to minimize lateral tire force and yaw moment input and to follow changed lane after collision avoidance. The performance of proposed algorithm has been investigated via computer simulation conducted to vehicle dynamic software CARSIM and Matlab/Simulink.

• Journal of Power Electronics
• /
• 제19권2호
• /
• pp.497-508
• /
• 2019

The predictive speed control (PSC) strategy can realize the simultaneous control of speed and current by using one cost function. As a model-based control method, the performance of the PSC is vulnerable to model mismatches such as load torque disturbances and parameter uncertainties. To solve this problem, this paper presents a robust predictive speed control (RPSC) strategy for surface-mounted permanent magnet synchronous motor (SPMSM) drives. The proposed RPSC uses extended state observers (ESOs) to estimate the lumped disturbances caused by load torque changes and parameter mismatches. The observer-based prediction model is then compensated by using the estimated disturbances. The introduction of ESOs can achieve robustness against predictive model uncertainties. In addition, a modified cost function is designed to further suppress load torque disturbances. The performance of the proposed RPSC scheme has been corroborated by experimental results under the condition of load torque changes and parameter mismatches.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제26권1호
• /
• pp.23-48
• /
• 2022

In this paper, we provide predictive models for the market price of fruits, and analyze the performance of each fruit price predictive model. The data used to create the predictive models are fruit price data, weather data, and Korea composite stock price index (KOSPI) data. We collect these data through Open-API for 10 years period from year 2011 to year 2020. Six types of fruit price predictive models are constructed using the LSTM algorithm, a special form of deep learning RNN algorithm, and the performance is measured using the root mean square error. For each model, the data from year 2011 to year 2018 are trained to predict the fruit price in year 2019, and the data from year 2011 to year 2019 are trained to predict the fruit price in year 2020. By comparing the fruit price predictive models of year 2019 and those models of year 2020, the model with excellent efficiency is identified and the best model to provide the service is selected. The model we made will be available in other countries and regions as well.

• Nuclear Engineering and Technology
• /
• 제33권1호
• /
• pp.102-110
• /
• 2001

In this work, the model predictive control method was applied to a linear model and a nonlinear model of steam generators. The parameters of a linear model for steam generators are very different according to the power levels. The model predictive controller was designed for the linear steam generator model at a fixed power level. The proposed controller at the fixed power level showed good performance for any other power levels by designed changing only the input-weighting factor. As the input-weighting factor usually increases, its relative stability does so. The steam generator has some nonlinear characteristics. Therefore, the proposed algorithm has been implemented for a nonlinear model of the nuclear steam generator to verify its real performance and also, showed good performance.

• KIEAE Journal
• /
• 제17권4호
• /
• pp.83-88
• /
• 2017

Purpose: The objective of this study is to develop a predictive model for calculating the amount of cooling load for the different setback temperatures during the setback period. An artificial neural network (ANN) is applied as a predictive model. The predictive model is designed to be employed in the control algorithm, in which the amount of cooling load for the different setback temperature is compared and works as a determinant for finding the most energy-efficient optimal setback temperature. Method: Three major steps were conducted for proposing the ANN-based predictive model - i) initial model development, ii) model optimization, and iii) performance evaluation. Result:The proposed model proved its prediction accuracy with the lower coefficient of variation of the root mean square errors (CVRMSEs) of the simulated results (Mi) and the predicted results (Si) under generally accepted levels. In conclusion, the ANN model presented its applicability to the thermal control algorithm for setting up the most energy-efficient setback temperature.

• Communications for Statistical Applications and Methods
• /
• 제23권4호
• /
• pp.343-353
• /
• 2016

In doubly-censored data, an originating event time and a terminating event time are interval-censored. In certain analyses of such data, a researcher might be interested in the elapsed time between the originating and terminating events as well as regression modeling with risk factors. Therefore, in this study, we introduce a model evaluation method to measure the predictive ability of a model based on negative predictive values. We use a semiparametric estimate of the predictive accuracy to provide a simple and flexible method for model evaluation of doubly-censored survival outcomes. Additionally, we used simulation studies and tested data from a prostate cancer trial to illustrate the practical advantages of our approach. We believe that this method could be widely used to build prediction models or nomograms.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1000-1003
• /
• 1996

Nonlinear unified predictive control(UPC) algorithm was applied to the temperature control of a batch polymerization reactor for polymethylmethacrylate(PMMA). Before the polymerization reaction is initiated, the parameters of the process model are determined by the recursive least squares(RLS) method. During the reaction, nonlinearities due to generation of heat of reaction and variation of heat transfer coefficients are predicted through the nonlinear model developed. These nonlinearities are added to the process output from the linear process model. And then, the predicted process output is used to calculate the control output sequence. The performance of nonlinear control algorithm was verified by simulation and compared with that of the linear unified predictive control algorithm. In the experiment of a batch PMMA polymerization, nonlinear unified predictive control was implemented to regulate the temperature of the reactor, and the validity of the nonlinear model was verified through the experimental results. The performance of the nonlinear controller turned out to be superior to that of the linear controller for tracking abrupt changes in setpoint.

• 한국지능시스템학회논문지
• /
• 제16권4호
• /
• pp.416-422
• /
• 2006

소프트웨어 개발 인력 프로파일에 대한 현존하는 모든 통계적 모델들은 소프트웨어 사용과 개발 프로세스의 가정에 기반을 두고 있어 일반적으로 적용 가능한 추정과 예측 모델이 없는 실정이다. 본 논문은 예측필터를 적용하여 소프트웨어 개발 투입 인력 프로파일을 예측하였다. 먼저 소프트웨어 개발 인력분포를 살펴보고, 예측필터를 적용하기 위해 모델의 입력 -출력, 모수를 결정하는 방법을 제시하였다. 이어서 제안된 모델의 유용성은 실제 개발된 소프트웨어 프로젝트로부터 획득된 데이터 분석으로 경험적으로 검증되었다. 평균 상대오차와 Pred(0.25)에 기반하여 제안된 예측필터는 잘 알려진 통계적 추정 모델들과 비교되었다. 검증 결과 예측필터는 단순한 구조를 갖고 있으면서도 소프트웨어 인력분포를 적절히 표현하는 결과를 보였다.

• 한국정보통신학회논문지
• /
• 제25권6호
• /
• pp.749-756
• /
• 2021

제조 생산공정에는 다양한 센서를 통해 실시간으로 양질의 데이터가 데이터베이스에 축적되고 있다. 이와 함께 통계적으로 접근하기 까다로운 데이터에 대해서 높은 수준의 정확도로 예측모델을 구축할 수 있는 머신러닝이 보급되면서 '4차 산업화 시대'를 맞이하고 있다. 본 논문에서는 이러한 제조업계의 흐름에 따라 업계의 주요 관심사인 제품의 품질특성을 예측하는 머신러닝 모델의 성능을 향상하는 방법을 제시한다. 머신러닝 모델의 성능을 향상하는데 일반적으로 사용되는 샘플 크기의 증가, Hyper-Parameter의 최적화 및 적절한 알고리즘 선택의 효과를 검증한다. 그리고, 새로운 성능향상 방법을 제시하고, 그 효과를 검증해본다. 논문에서 제시한 방법을 통해서 제조업에서는 더욱 향상된 성능의 예측모델을 구축, 품질예측과 관리에 크게 이바지할 수 있을 것이다.