http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
비구조적 불확실성이 존재하는 DC모터에 대한 외란관측기 기반 제어기의 강인성에 대한 연구
조남훈,Jo, Nam-Hoon 대한전기학회 2017 전기학회논문지 Vol.66 No.1
In this paper, we study the robustness of disturbance observer based controller for DC motor in the presence of unmodeled dynamics. It is well known that the robustness property usually becomes weaker as the control gain becomes larger. On the contrary to this expectation, it is shown that the phase margin of DOB controller remains quite a large value even though the time constant of Q-filter becomes smaller. The computer simulation results show that DOB controller is able to stabilize the motor system even in the presence of unmodeled dynamics. On the contrary, the unity-feedback system fails to maintain stability when a high gain feedback is employed for the purpose of achieving better disturbance attenuation performance.
비최소위상 외란관측기 제어기의 도립진자 적용에 관한 연구
조남훈(Nam Hoon Jo) 제어로봇시스템학회 2018 제어·로봇·시스템학회 논문지 Vol.24 No.6
Due to powerful ability of disturbance attenuation and uncertainty compensation, the disturbance observer (DOB) controller has been widely employed in control industries. But, it is well known that it is not easy to design conventional DOB controllers for non-minimum phase systems. In this paper, we design a non-minimum DOB controller to achieve high performance control for inverted pendulum system. It is shown via simulation studies that disturbance attenuation performance can be substantially improved by the application of non-minimum phase DOB controller. Since the derived DOB controller may have very high order, it is suggested that model reduction technique can be used for the design of DOB controller. It is shown that the order of DOB controller can be decreased without degrading the system performance significantly.
조기학습정지를 이용한 원전 SG세관 결함크기 예측 신경회로망의 성능 향상
趙南熏(Nam-Hoon Jo) 대한전기학회 2008 전기학회논문지 Vol.57 No.11
In this paper, we consider a performance improvement of neural network for predicting defect size of steam generator tube using early stopping. Usually, neural network is trained until MSE becomes less than a prescribed error goal. The smaller the error goal, the greater the prediction performance for the trained data. However, as the error goal is decreased, an over-fitting is likely to start during supervised training of a neural network, which usually deteriorates the generalization performance. We propose that, for the prediction of an axisymmetric defect size, early stopping can be used to avoid the over-fitting. Through various experiments on the axisymmetric defect samples, we found that the difference between the prediction error of neural network based on early stopping and that of ideal neural network is reasonably small. This indicates that the error goal used for neural network training for the prediction of defect size can be efficiently selected by early stopping.
신경회로망을 이용한 원전SG 세관 결함패턴 분류성능 향상기법
趙南熏(Nam-Hoon Jo),李享範(Hwang-Beom Lee),韓基元(Ki-Won Han),宋城鎭(Sung-Jin Song) 대한전기학회 2007 전기학회논문지 Vol.56 No.7
In this paper, we study the classification of defects at steam generator tube in nuclear power plant using eddy current testing (ECT). We consider 4 defect patterns of SG tube: Ⅰ-In type, Ⅰ-Out type, Ⅴ-In type, and Ⅴ-Out type. Through numerical analysis program based on finite element modeling, 400 ECT signals are generated by varying width and depth of each defect type. In order to improve the classification performance, we propose new feature extraction technique. After extracting new features from the generated ECT signals, multi-layer perceptron is used to classify the defect patterns. Through the computer simulation study, it is shown that the proposed method achieves 100% classification success rate while the previous method yields 91% success rate.
趙南熏(Nam-Hoon Jo) 대한전기학회 2006 전기학회논문지A Vol.55 No.11
In this paper, we study the problem of model selection for Support Vector Machine(SVM) predictor for short-term load forecasting. The model selection amounts to tuning SVM parameters, such as the cost coefficient C and kernel parameters and so on, in order to maximize the prediction performance of SVM. We propose that Cross-Validation method can be used as a model selection algorithm for SVM-based load forecasting technique. Through the various experiments on several data sets, we found that the difference between the prediction error of SVM using Cross-Validation and that of ideal SVM is less than 5%. This shows that SVM parameters for load forecasting can be efficiently tuned by using Cross-Validation.
조남훈(Nam-Hoon Jo) 대한전기학회 2021 전기학회논문지 Vol.70 No.12
In this paper, we study the robust stability of closed-loop system with noise reduction-disturbance observer (NR-DOB) when the relative degree of plant is greater than that of nominal model. Compared with the case where the relative degree of plant is the same as that of nominal model, it is shown that the nominal model for NR-DOB controller should be carefully chosen in order to achieve the robust stability. Computer simulation results to confirm the proposed condition are also included.
결함 형태 분류 과정이 필요없는 SG 세관 결함 크기 추정 시스템의 성능 평가
조남훈(Nam-Hoon Jo) 한국비파괴검사학회 2010 한국비파괴검사학회지 Vol.30 No.1
본 논문에서는 원전SG세관 결함 크기 추정을 위한 새로운 구조의 추정시스템에 대한 연구를 수행한다. 기존의 연구에서는 결함 크기를 추정하기 위하여 각각의 결함 형태별로 결함크기추정시스템을 설계하였다. 이와 같은 경우, 추정시스템의 구조가 복잡해지고 결함 크기 추정 이전에 수행하는 결함형태분류기의 정확성이 떨어질 경우 결함 크기 추정 성능도 결과적으로 악화될 수밖에 없다. 이에 본 논문에서는 결함 형태분류 과정을 필요로 하지 않는 결함크기추정시스템의 성능을 분석하고 이를 향상시키기 위한 방안을 연구하였다. 기존의 추정시스템은 각각의 결함 형태별로 특화된 추정기를 사용하기 때문에 추정 성능이 훨씬 뛰어날 것으로 예상되었지만, 실험 결과 두 추정시스템의 성능 차이는 그리 크지 않다는 것을 알 수 있었다. 따라서 결함형태분류기의 정확성이 완벽하지 않을 경우, 본 논문에서 제안한 구조의 추정기가 효과적으로 사용될 수 있을 것으로 기대된다. In this paper, we study a new estimation system for the prediction of steam generator tube defects. In the previous research works, defect size estimators were independently designed for each defect types in order to estimate the defect size. As a result, the structure of estimation system is rather complex and the estimation performance gets worse if the classification performance is degraded for some reason. This paper studies a new estimation system that does not require the classification of defect types. Although the previous works are expected to achieve much better estimation performance than the proposed system since it uses the estimator specialized in each defect, the performance difference is not so large. Therefore, it is expected that the proposed estimator can be effectively used for the case where the defect type classification is imperfect.
상태궤환 적분제어기법을 이용한 HIV 감염 환자에 대한 약물 치료기법
조남훈(Nam-Hoon Jo) 대한전기학회 2015 전기학회논문지 Vol.64 No.10
In this paper, a drug treatment protocol is proposed for an HIV infection model that explicitly includes the concentration of healthy T cells, infected T cells, and HIV. Since real parameters of HIV infection model differ from patient to patient, most drug treatment protocols are not able to achieve the treatment goal in the presence of modelling errors. Recently, based on the nonlinear robust control theory, a robust treatment protocol has been proposed that deals with parameter uncertainties. Although the developed scheme is inherently complex, it cannot be applied to the case where all parameters are unknown. In this paper, we propose a new drug treatment protocol that is much simpler than the previous one but can achieve the treatment goal even when all model parameters are unknown. The simulation results verify that the substantial improvement in the performance can be achieved by the proposed scheme.