• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.45 no.2
• /
• pp.273-282
• /
• 1996

This work is conceded with the estimation of output derivatives and their use for the design of robust controller for linear systems with system uncertainties due to modeling errors and disturbances. It is assumed that a nominal transfer function model and quantitative bounds for system uncertainties and known. The developed control schemes are shown to achieve regulation of the system output and ensures boundedness of the system states without imposing any structural conditions on system uncertainties and disturbances. Output derivative estimation is first conducted through restructuring of the plant in a specific parameterization. They are utilized for constructing robust nonlinear high-gain feedback controller of a SMC(Sliding Mode Control)type. The performances of the developed controller are evaluated and shown to be effective and useful through simulation study.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1265-1274
• /
• 2018

This paper proposes a virtual flux (VF) and positive-sequence power based control strategy to improve the performance of grid-interfaced three-phase voltage source converters against unbalanced and distorted grid conditions. By using a second-order generalized integrator (SOGI) based VF observer, the proposed strategy achieves an AC voltage sensorless and grid frequency adaptive control. Aiming to realize a balanced sinusoidal line current operation, the fundamental positive-sequence component based instantaneous power is utilized as the control variable. Moreover, the fundamental negative-sequence VF feedforward and the harmonic attenuation ability of a sequence component generator are employed to further enhance the unbalance regulation ability and the harmonic tolerance of line currents, respectively. Finally, the proposed scheme is completed by combining the foregoing two elements with a predictive direct power control (PDPC). In order to verify the feasibility and validity of the proposed SOGI-VFPDPC, the scenarios of unbalanced voltage dip, higher harmonic distortion and grid frequency deviation are investigated in simulation and experimental studies. The corresponding results demonstrate that the proposed strategy ensures a balanced sinusoidal line current operation with excellent steady-state and transient behaviors under general grid conditions.

• The Korean Journal of Physiology and Pharmacology
• /
• v.24 no.1
• /
• pp.89-99
• /
• 2020

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.