The Computer Algebra Systems in Mathematics Education : A Review of Recent Research

Kwon, Oh-Nam,Shin, Eun-Ju,Shon, Bo-Kyung 이화여자대학교 교과교육연구소 2001 East west education Vol.18 No.-

This study examines the potential role for Computer Algebra Systems(CAS) in such an educational context, particularly their potential to get students more engaged in the learning of mathematics as well as to reduce the "strangle hold" of prerequisite algebraic skills, in courses such as functions, introductory calculus. What is the present status of and near-future outlook for computer algebra in Korean secondary schools? The answer that arises from this review of available fact is: bright with promise, but still dim. Several recent developments have set the stage for an acceleration in the dissemination rate for CAS to high schools. However, this review of some past and present conditions indicates that many years will pass before computer algebra is impacting high school mathematics as much as hand-held calculators are doing now.

확장된 칼만필터를 이용한 순환신경망의 훈련 알고리즘

권오신 국립7개대학공동논문집간행위원회 2005 공업기술연구 Vol.5 No.-

Recurrent Neural network trained with gradient based learning algorithm has a drawback of slow convergence rates and long training sequences required for satisfactory performance. In a high speed communications, fast convergence speed and short training symbols are very important conditions. This paper presents decision feedback equalizers using a recurrent neural network trained with Extended Kalman Filter (EKF). The main features of the proposed EKF algorithm are fast convergence rates and good performance using short training symbol. Through experiments of channel equalization, performance of the recurrent neural networks with the EKF is evaluated.

침지식 한외 여과막을 이용한 무산소/혐기 교대운전형 반응조의 질소·인처리에 관한 연구

신대윤,김동임,권오상 조선대학교 환경공해연구소 2004 環境公害硏究 Vol.18 No.-

Nowadays, Water is deteriorating with several reasons such as increase of wastewater which is flowed into river and limitation of water resource. In this study, reuse water system was configurated using submerged membrane bioreactor(SMBR) system consisted of anoxic, selector, and aerobic tank with membrane to concentrate on plans for securing of water resource by reusing wastewater treatment effluen. This study was examined the feasibility of applying a Intermittently Internal Recycle Membrane Bioreactor to household wastewater treatment. The bioreactor was intermittently recycle to alternate anaerobic/anoxic condition while find was entered the early stage of anoxic period and membrane filtration occurred during aeration period. With about 5,000㎎/L of MLSS , 94.4% of TCOD_(Mn), 49.6% of T-N and 56.8% T-P, With about lO.OOO㎎/L of MLSS 94.0% of TCOD_(Mn), 51.3% T-N and 49.7% of T-P and With about lO.OOO㎎/L of MLSS (anoxic 1hr, anaerobic 2hr) 96.6% of TCOD_(Mn), 67.2% of T-N and 53.0% of T-P could be removed. Therefore, this study confirmed the possibility of the domestic water treatment using membrane bioreactor process with Intermittently internal recycle.

Chlorella vulgaris의 세포막에서 일어나는 Glycine 흡수과정에 관한 연구

權五溶,辛昌男,吳炅煥,李錦洙 충남대학교 자연과학연구소 1981 忠南科學硏究誌 Vol.8 No.1

In order to datemine that glycine uptake in the cell membrane of Chlorella vulgaris is depend on the specificity of substrate and energy required, it has been investigated on the reaction velocity of glycine and certain carrier. Some effects on the transmural fluxes of glycine across the cell membrane were also detemined. 1. The time course of glycine uptake was a linear function of time to early 10 min, and increased in the steady state slowly. 2. Glycine influx had shown apparent affinity in the range of 0.001 mM to 10 mM solution of glycine and followed Michaelis-Menten Kinetics(Km-0.16 mM, Vmax=0.31 nM2.0x10^7 cells min). 3. Glycine influx with 1 mA IAA was reduced about 50 percentage, suggesting that glycine uptake process is energy-dependent. 4. Glycine uptake was inhibited by glutamic acid, alanine, phenylalanine, proline, and methionine with the exception of histidine. It was suggested that the inhibition was competeted with some amino acids described above.

地域社會開發事業의 意義와 그展望에 對한 小考

崔錦紅,李順愛,吳信姬,池亭順 梨花女子大學校 師範大學 社會生活科 1961 綠友硏究論集 Vol.- No.3

鷄龍山 陸松群落 및 리기다松群落의 林床植生에 關하여

權五溶,辛昌男,吳炅煥 충남대학교 자연과학연구소 1979 學術硏究誌 Vol.6 No.2

A study was carried out to ascertain the standing crops, species composition, and relationships between the standing crops and soil nutrients in the Pine forest and the Pitch pine forest of Gyeryong Mt. Species compositions of the Pine forest and the Pitch pine forest were 36 and 20 species, and dominant species of them was Festuca ovina . Total live standing crops of ground vegetation, 215.28-631.67g/㎡ and 188.62-446.13g/㎡, were highest in September, and net productivities were 13.02-169.95g/㎡ and 12.62-109.58g/㎡, respectively. Total live standing crops were significantly correlated with the water content and the organic matter of soil in each forest, and comparatively with the total nitrogen and the exchangeable phosphorus in the Pitch pine forest.

Pinus koraiensis와 Pinus parviflora의 樹高에 따른 氣孔數 變異에 關하여

權五容,辛昌男,吳炅煥,方在旭 충남대학교 자연과학연구소 1980 學術硏究誌 Vol.7 No.2

Stomata1 distribution in relation to the height of Pinus parviflora and Pinus koraiensis was studied 1. Stomata1 bands per needle of P parviflora and P koraiensis were 8.60±0.62 - 11.47±0.74, 14.07±1.03 - 16.60±0.99 and the number of stomata per band were 382.0±8.80 404.27±26.16, 1,326. 67±87.78 - 2,286.58±214.57, respectively 2. The number of stomatal band was significantly different between 3, 4m and 5, 6m in P parviflora and between 5, 6m and 7m in P. koraiensis Consequently, it is suggested that stomatal frequency gradient associated with increasing height above the ground is correlated with the greater degree of exposure to light.

淸溪山 落葉松林의 林床草本 群叢에 對한 種相關에 관하여

權五溶,辛昌男,鄭會台 충남대학교 자연과학연구소 1983 忠南科學硏究誌 Vol.10 No.2

The study of the herbaceous species association in the larch forest of Mt. Chungkyeu was carried out from April to October, 1980. 1. The herbaceous ground association was consist of 43 species, Oplismenus undulatifolius, Festuca ovina, Oalamagrostis arundinaceae var. brachytricha and Carex pediformis var. rhizina were dominant species. 2. The major herbaceous species, of which important value was above 10, were Oplismenus undulatifolius(28.2%), Festuca ovina(26.7%), Miscanthus purpurascens(26.5%), Calamagrostis arundinaceae var. brachytricha(21.2%), Carex nanella(17.5%), Carex lanceolata(15.4%), Carex pediformis var. rhizina(14.1%), Carex siderosticta(12.7%), Potentilla fragarioides(11.7%) and Artemisia keiskeana(10.5%). 3. Classifying the herbaceous ground association by partial ordination, it was divided into two kinds of association ; Carex-Oplismenus association and Calamagrostis-Oplismenus association. 4. The results of the species association analysis was the same as the ordination. The coefficients of species exhibited the positive value in the same association, on the contrary, it was negative value in the different association.

CFRP 판재의 섬유방향에 따른 절삭성 평가

신봉철,하석재,강민형,권오양,조명우 한국공작기계학회 2008 한국공작기계학회 추계학술대회논문집 Vol.2008 No.-

In recent, the CFRP(Carbon Fiber Reinforced Plastics) composite has been widely used because it has light-weight and high mechanical strength. Therefore, CFRP composite is applied to variable industrial areas such as aerospace, automobiles, sporting goods, ships and so on. Generally, CFRP composite has been used in the industrial field by mechanical cutting processes. In this research, the CFRP composite specimen are laminated by 64plies with orientation angles such as 0°, 45°, 90° and [0°/±45°/90°]4S. The results are analyzed with cutting force and cutting surface.

언센티드 칼만필터를 이용한 순환신경망 학습 및 채널 등화에의 응용

권오신 국립7개대학공동논문집간행위원회 2003 공업기술연구 Vol.3 No.-

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. This paper presents decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizer utilizing unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental result is presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.