This paper describes a program which learns good strategies for two-poison, deterministic, zero-sum board games of perfect information. The program learns by simply playing the game against either a human or computer opponent. The results of the program's teaming of a lot of games are reported. The program consists of search kernel and a move generator module. Only the move generator is modified to reflect the rules of the game to be played. The kernel uses a temporal difference procedure combined with a backpropagation neural network to team good evaluation functions for the game being played. Central to the performance of the program is the search procedure. This is a the capture tree search used in most successful janggi playing programs. It is based on the idea of using search to correct errors in evaluations of positions. This procedure is described, analyzed, tested, and implemented in the game-teaming program. Both the test results and the performance of the program confirm the results of the analysis which indicate that search improves game playing performance for sufficiently accurate evaluation functions.

Disaster mitigation, especially as a concept similar to damage mitigation caused by heavy rainfalls and flood is closely related to long-term development plan. This plan of an harbor area where is located in lower region is established and carried under consideration of disaster mitigation concept such as internal drainage and so on. Flood damage is somewhat predictable in accordance with height, stream and rainfall characteristics of the region. Therefore it is necessary to establish national and urban plan under consideration of this fact. But this consideration of existing regulation and institution is insufficient and improvement of regulation and institution is needed. This consideration of disaster mitigation fields is regulated declaratively and inclusively in national plan which is established in broad region, and specifically and detailedly in urban plan which is established in narrow region. The program to improve regulation and institution is proposed in order to consider disaster mitigation fields as a level of this plan.

In order to study the long-term change of heavy metal concentrations in the Kumho river water, water analysis was conducted at l3sites surrounding the Kumho river system for l8times from September 1993 to August 1999. Analytical items for the study of water quality are Cu, Zn, Cd, Cr, Fe, Mn and Pb. The six year term studied in this work was divided into Part Ⅰ and Part Ⅱ, which covers the period from September 1993 to August 1996 and the period from September 1996 to August 1999, respectively. The mean concentrations of Cu, Zn, Cd, Cr, Fe, Mn and Pb in the unit of ppm for the PartⅠ period showed 0.032, 0.025, 0.006, 0.006, 0.050, 0.053 and 0.019, respectively. The mean concentrations of Cu, Zn, Cd, Cr, Fe, Mn and Pb in the unit of ppm for the Part Ⅱ period showed 0.001, 0.001, 0.001, 0.004, 0.020, 0.002 and 0.002, respectively. The heavy metal concentrations in the Kumho river water for the second period were found to be decreased by 1/32, 1/25, 1/6, 1/1.5, 1/2.5, 1/26.5 and 1/9.5 for Cu, Zn, Cd, Cr, Fe, Mn and Pb, respectively. The present results clearly indicate that the water quality in the Kumho river is improving in terms of heavy metal contaminations.

본 연구는 서울지점의 연강수량 기록 중 1900년을 전후한 장기 건조기와 같이 연속적으로 발생하는 과우해에 대해 주목하고, 이와같이 연속된 과유해의 재현 가능성을 살펴보는 것을 목적으로 한다. 과우해의 기준으로는 과우해의 발생에 독립성이 보장되도록 포아송 분포를 따르는 수준으로 결정하였으며, 평균-0.5표준편차의 절단수준에서는 대략 99%의 유의수준에서, 평균-0.75표준편차 및 그 이하의 절단수준에 대해서는 95%의 유의수준에서 포아송 분포를 만족시키

유역사면 토양 내에서 지하수의 거동을 모의할 수 있는 TOPMODEL을 이용하여 월별 유출량을 추정하였다. 한국수자원공사에서 운영중인 사연댐 유역(124.5 )을 분석대상유역으로 선정하였으며, 국립지리원에서 작성한 1/5,000 수치지도와 지리정보시스템 도구인 Arc/Info를 이용하여 지형인자 등을 분석하였다. TOPMODEL을 이용한 월별 유출량에 대한 모의결과를 관측치와 비교할 때 재현성이 우수한 것으로 판단되었다. 격자 크기에 따른 영향을 검토하

In an attempt to test experimental condition of preparing grape powder, grapes having less commercial value was used and tried. With drying method, spray and freeze drying were satisfactory to produce power. Moisture content and odor retention were better by the latter method. Three grape strains stored for 40 days contained more odors than those stored for 5 days. Maltose 90% plus dextrin 10% was suitable for drying support. To increase odror sense, citric acid and vitamin C can be added up to 0.1 and 0.2%, respectively. Considering these conditions, grape complex powder prapared from grape powder 20% comprising drying support, glucose 79.7%, citric acid 0.1%, vitamin C 0.2% with freeze drying was the best by overall evaluation including sensory test. When campbell and neomuscut were mixed by 15:5 or 10:10, sensory evaluation was also ameliorated.

본 연구의 목적은 융설모형을 국내에 도입하고 이를 강우-유출모형과 결합하여 우리나라 북부 산간지방의 장기 유출해석을 위한 융설의 영향을 고찰하는데 있다. 선택된 모형은 미국 국립기상청의 온도지수 융설모형이며, 이를 유연면적이 1,059.6 인 내린천 유역에 적용.검토하였다. 모형에 의해서 계산된 강설의 지역적 분포에 대한 시간적 변동성은 인제 측후소에서 관측된 강설량의 변동성과 잘 일치하는 것으로 나타났다. 또한 모형에 의해서 계산된 토양수분 및 하천유

유통을 위한 포도 보존 중 품질에 미치는 영향을 구명하여 보존기간 연장에 효과적인 적정 포장재 제작에 대한 기초자료를 얻고자 포도 캠벨얼리를 공시하여 스치로폴 상자, 마닐라상자 등 6처리를 두어 시험한 결과, 중량감소는 유통일수가 경과함에 따라 많았으며, 처리 간에는 15일 보존시 스치로폴상자에 비해 스치로폴상자+(스치로폴접시 + BioPE 필름밀봉), 스티로폴상자 + (스치로폴접시 + 랩 밀봉)처리가 적었으며, 비정상과는 밀봉구에서 많았으나,

Chemical, physical md microbial analysis of salted baechu(Chinese cabbage) and its exudate were conducted during storage in large plastic bags(LDPE, HDPE, PVC-box) at 0. Salinity was slightly lowered for 2 weeks storage and maintained that level thereafter in salted haechu, but in its exudate increased rapidly for 2 weeks storage and decreased thereafter. In all treatment pH decreased to 5.0 in salted baechu and to 4.5 in its exudate for 6 weeks storage. Reducing sugar content of salted baechu was 1.5∼2.0 fold higher than that of its exudate, and decresed slowly in both. Total viable cells increased rapidly for 2 weeks but maintained that level thereafter and lactic acid bacteria increased continuously until 6 weeks. Lightness of exudate decreased rapidly in PVC-box, but decreased slowly in LDPE. Compression force of salted baechu increased continuously untill 8 weeks except for PVC-box.

댐 퇴사량 예측방법은 수리학적 메커니즘을 이용하는 방법과 실측자료를 설계치로 적용하는 경험적 방법으로 구분할 수 있다. 전자는 물리적 해석방법으로 단기간에 이루어지는 현상을 규명하는데 적용할 수 있으며, 후자는 댐사수량 결정등 장기적인 설계변수 추정에 용이하게 이용할 수 있다. 이중 경험적 자료인 배사비 곡선식을 이용하면 월단위의 저수지 운영이 가능하며, 저류량에 대응하는 퇴사량을 정보변수로 추적하면 퇴사관리의 최적운영이 가능하다. 이와 같은 접근방법은

교량지점에서의 총세굴깊이는 세가지 성분 즉, 장기하상변동, 단면축소세굴, 그리고 국부세굴로 구성된다. 그러므로 장기하상변도의 분석은 교량지점에서의 총세굴깊이를 산정하는데 있어 중요하다고 할 수 있다. 본 연구에서는 CHARIMA모형과 HEC-6모형을 사용하여 남한강교 지점에서의 장기하상변동을 분석하였다. 분석결과, 5년 동안 정상류로 물이 흐를 경우 CHARIMA모형은 약 45cm의 하상고의 상승을 보여준 반면 HEC-6모형은 약 5cm정도 하상고가 저