http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
블록들의 유사성을 고려한 Adaptive Block-Wise RPLS
윤경우,이영학,한종훈 한국화학공학회 2003 Korean Chemical Engineering Research(HWAHAK KONGHA Vol.41 No.5
부분 최소 자승법 (partial least squares, PLS)은 복잡한 공분산 구조를 가진 상관성이 큰 변수들이 포함된 공정을 모델링 하는데 효과적으로 사용되어 왔다. 하지만 공정 변화에 적응하는 온라인 모델을 만들기 위해 부분 최소 자승법을 이용하는 것은 많은 제약이 따른다. 그래서 많이 이용되고 있는 방법이 recursive PLS(RPLS)이다. RPLS는 새로운 데이터 샘플이 들어오면 모델을 갱신 한다. 하지만 데이터의 상당한 양이 모이거나 공정의 큰 변화가 있을 때까지 모델을 갱신하지 않을 수도 있다. 따라서 이 논문에서는 공정 변화에 잘 적응하고 모델 갱신을 블록 단위로 하는 block-wise RPLS의 온라인 적응 방법인 moving window 방법을 이용하였다. Block-wise RPLS는 희귀 모델을 만들고 검증하는데 필요한 계산 부하를 감소시키는 효과가 있다 본 연구에서는 모델을 구성하는 윈도우내에 있는 블록들 사이의 상관관계를 나타내는 지표를 정의한 다음, 그 지표를 통해 forgetting factor를 정하는 방법을 제시하였다. 이 방법은 공정에 대한 일반적인 지식이 필요하지 않고 단지 데이터 블록의 상관관계만으로 적응 모델의 예측력을 높일 수 있는 방법이다. 따라서 다른 방법들이 나타내지 못한 상관성 정보를 포함시켰다. 이 방법을 기반으로 정유 공장의 가열로의 NO_(x)(nitrogen oxides) 배출 농도를 예측하는 모델을 forgetting factor를 이용해 구성했고, 이 forgetting factor에 의해 예측력이 최소 에러 값의 95%이상 근접함을 확인하였다. Partial least squares (PLS) regression has been effectively used as one of the data-driven empirical modeling to deal with a large number of variables in the complicated covariance structure. But Partial least squarer (PLS) regression has limitations in the online model update. Therefore, recursive PLS has ken used for the online adaptation of the model. The RPLS algorithm is implemented as soon as some new samples are available. It may be desirable not to update the model until significant amount of data are collected and the process has gone through significant changes. In this paper. we used the block-wise recursive partial least squares (RPLS) algorithms with a moving window for the adaptation of the model according to the process shins. The block-wise RPLS algorithm has been used to reduce computational load in PLS regression and its cross-validation. In this work we defined the index to represent correlation between blocks and proposed how to determine the forgetting factors through it. This proposed method is to improve the prediction power of the adaptive modeling considering correlation between blocks without general process knowledge. Therefore, this proposed method included correlation information that different methods did not express. The method was tested to process heaters in the oil refining wmpany. We costructed the prediction model of the effluent NO, composition with forgetting factors and showed that the prediction power approximated more than 95% of the minimum error.
한경호,조윤석 단국대학교 1998 論文集 Vol.32 No.-
ITU G.721, G.723 standard ADPCM algorithm is implemented by using Tl's fixed point DSP start kit (DSK). The ADPCM can be implemented on a various rates, such as 16Kbps, 32Kbps and 40Kbps. The ADPCM is a sample based compression technique and its complexity is not so high as the other speech compression techniques such as the other speech compression techniques such as CELP, VSELP and GSM, etc. and can be implementd on a low cost DSP kit. ADPCM is widely applicable to most of the low cost speech compression application such as the tapeless answering machine, simultaneous voice and fax modem, digital phone, etc. TMS320C50 DSK system has a TMS320C50 DSP is a low cost fixed point DSP chip and an AIC (analog interface chip) which operates as a single chip A/D and D/A converter with 14 bit resolution. ADPCM C code is compiled by TI TMS320C50 C-compiler and implemented on the DSK on-chip memory. Speech signal input is converted into 14 bit linear PCM data and encoded into ADPCM data and the data is sent to PC through RS232C. The ADPCM data on PC is received by the DSK through RS232C and then decoded to generate the 14 bit linear PCM data and converted into the speech signal. The DSK system has audio in/out jack and we can input and output the speech signal. The simulation is performed on Pentium PC with C-code executable files. The result of speech is compared with the original speech. It is performed encoding on DSK and decoding on PC and the result is audible but not so good as that of PC simulation.
윤경한(kyong-Han Yun),이용상(Yong-Sang Lee),김영철(Young-Chol Kim),변윤섭(Yeun-Sub Byun) 대한전기학회 2010 대한전기학회 학술대회 논문집 Vol.2010 No.7
본 논문은 바이모달 트램의 위치추정을 위한 초기위치 탐색 알고리즘에 대한 내용을 다루었다. 바이모달 트램의 위치추정알고리즘은 도로에 매설된 자석표지의 인식을 기반으로 하며, 자석표지 감지결과를 이용하여 차량의 추정위치를 계산한다. 차량의 초기위치 탐색 알고리즘은 자석표지의 극성정보를 이용하며, 일정 수량의 자석표지 극성정보를 수집하여 데이터베이스와 비교함으로써 감지된 자석의 위치좌표를 확인하고, 이 좌표정보를 이용하여 차량의 추정위치를 계산한다. 설계된 알고리즘은 주행시험 데이터를 이용하여 시뮬레이션을 통해 성능을 확인하였다.
윤경한(kyong-Han Yun),김영철(Young-Chol Kim),이용상(Yong-Sang Lee),정해봉(Hae-Bong Jung),변윤섭(Yeun-Sub Byun) 대한전기학회 2009 대한전기학회 학술대회 논문집 Vol.2009 No.7
본 논문은 한국 철도기술연구원에서 바이모달 트램의 안내/추진/도킹 제어기 검증을 위해 제작중인 LabVIEW, PXI기반의 실시간 시뮬레이터를 위한 시뮬레이션 프로그램 개발에 대한 내용을 다루었다. 시뮬레이터 프로그램은 바이모달 트램의 비선형 동적모델, 위치추정필터, 안내/추진/도킹 제어기로 구성되며, 비선형 동적모델에는 실제 차량에서 제어기나 관측기에 사용되는 각종 측정값들을 모사하기 위한 센서계측시스템이 포함된다.
윤경한(Kyong-Han Yun),이용상(Yong-Sang Lee),민경득(Kyung-Deuk Min),김영철(Young-Chol Kim),변윤섭(Yeun-Sub Byun) 대한전기학회 2011 전기학회논문지 Vol.60 No.10
This paper is concerned with the robustness evaluations of the guidance controller for a bimodal tram which is being developed by the Korea Railroad Research Institute (KRRI). The bimodal tram is an all-wheel steered multiple-articulated vehicle as a new kind of transportation vehicle. This vehicle has to be equipped with an automatic guidance system. In [1], such a controller has been recently proposed. However, since the performance is affected by weight change of the vehicle due to number of the passenger, model parameter uncertainties depending on the state of friction and the elasticity of the tire, and a typhoon, the controller designed must be examined with these conditions. As expected, because the vehicle dynamics is highly nonlinear, for the sake of investigating the robustness of the controller we compose two simulation ways based on the vehicle models which are implemented by the ADAMS and the MATLAB/LabVIEW toolboxes. Different uncertainties and a typhoon disturbance have been considered for the simulation conditions. Simulation results are shown.
자기적 안내제어시스템을 이용하는 굴절차량의 위치 및 방위각 추정
윤경한(Kyong-Han Yun),김영철(Young-Chol Kim),민경득(Kyung-Deuk Min),변윤섭(Yeun-Sub Byun) 대한전기학회 2011 전기학회논문지 Vol.60 No.10
For automated guidance control of a magnetically guided-all wheel steered vehicle, it is necessary to have information about position and orientation of the vehicle, and deviations from the reference path in real time. The magnet reference system considered here consists of three magnetic sensors mounted on the vehicle and magnetic markers, which are non-equidistantly buried in the road. This paper presents an observer to estimate such position and orientation at the center of gravity of the vehicle. This algorithm is based on the simple kinematic model of vehicle and uses the data of wheel velocity, steering angle, and the discrete measurements of marker positions. Since this algorithm requires the exact values of initial states, we have also proposed an algorithm of determining the initial position and orientation from the 16 successive magnet pole data, which are given by the magnetic measurement system(MMS). The proposed algorithm is capable of continuing to estimate for the case that the magnetic sensor fail to measure up to three successive magnets. It is shown through experimental data that the proposed algorithm works well within permissible error range.