• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.519-524
• /
• 2009

Recently, reinforcement learning methods have drawn much interests in the area of machine learning. Dominant approaches in researches for the reinforcement learning include the value-function approach, the policy search approach, and the actor-critic approach, among which pertinent to this paper are algorithms studied for problems with continuous states and continuous actions along the line of the actor-critic strategy. In particular, this paper focuses on presenting a method combining the so-called ACFRL(actor-critic fuzzy reinforcement learning), which is an actor-critic type reinforcement learning based on fuzzy theory, together with the RLS-NAC which is based on the RLS filters and natural actor-critic methods. The presented method is applied to a control problem for crawling robots, and some results are reported from comparison of learning performance.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.315-323
• /
• 2001

The concept of vertical partitioning has been discussed so far in an objective of improving the performance of query execution and system throughput. It can be applied to the areas where the match between data and queries affects performance, which includes partitioning of individual files in centralized environments, data distribution in distributed databases, dividing data among different levels of memory hierarchies, and so on. In general, a vertical partitioning algorithm should support n-ary partitioning as well as a globally optimal solution for the generation of all meaningful fragments. Most previous methods, however, have some limitations to support both of them efficiently. Because the vertical partitioning problem basically includes the fuzziness property, the proper management is required for the fuzziness problem. In this paper we propose an efficient vertical $\alpha$-partitioning algorithm which is based on the fuzzy theory. The method can not only generate all meaningful fragments but also support n-ary partitioning without any complex mathematical computations.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.4
• /
• pp.115-124
• /
• 2014

Since it costs a lot to build a home network system of new service provider, the entry barrier is quite high. If it would be possible to build a home network system with high performance at a low building cost, it would be quite attractive. If a home network would be built in P2P-SIP(Peer to peer - Session Initiation Protocol) structure, it is possible to decrease server maintenance cost and keep high usability by deconcentrating the traffic concentrated on the server. It is also possible to efficiently manage the location information of terminal, if quorum system based on fuzzy logic would be applied. In this paper we propose the P2P-SIP structure applied of quorum system based on fuzzy logic by which a home network system can be built at low cost. It was possible to know by comparing the structure with existing one that the structure has very good performance.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.4
• /
• pp.359-371
• /
• 2001

Project scheduling under resource constraint conditions have contained to many uncertain factors and it is perfonned by human experts. The expert identifies the activities of the project, decides the precedent relationships between these activities, and then construct the schedule using expected activity's duration. At this time, most of the scheduling methods concentrate on one of scheduling factor between activity duration and cost. And the activity duration, which is the most important factor in scheduling, is decided by heuristic of expert. Therefore it may cause uncertainty of activity duration decision and the use of this activity duration may increase the uncertainty of constructed schedule. This paper proposes Fuzzy-MOEH scheduling algorithm, which is the aggregation of the fuzzy number for deciding activity duration and applies the cost function for solving the problems of previous scheduling methods. This paper also analyze the utility and property of Fuzzy-MEOH algorithm through the comparison between Fuzzy-MEOH algorithm and existing MEOH algorithm.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.501-501
• /
• 2023

하천에서의 여가활동에 대한 수요가 증가함에 따라 각종 친수활동에 대한 안전도 평가가 사고예방을 위해 중요해지고 있다. 친수 활동의 안전은 수리 및 수질 인자에 크게 영향을 받지만 기존 친수지수는 수질 인자에만 집중되어 개발되어왔다. 하지만, 세일링, 패들링, 저동력보트 등 입수형 친수활동의 경우, 다양한 수리 현상에 큰 영향을 받기 때문에 유속, 흐름 방향, 수심 및 수면 폭 등의 수리인자를 친수지수에 반영할 필요가 있다. 또한, 친수활동에 위험이 되는 수리적 조건은 유량 조건과 하천의 평면적 공간에 따라 상이하게 발생하기에 이를 공간적으로 평가하는 것 역시 필요한 실정이다. 본 연구에서는 수리학적 요소를 기반으로 하천 친수 활동에 대한 안전도를 평가하기 위해 공간적으로 친수활동의 안정성을 평가할 수 있는 SRRI (Spatial River Recreation Index)를 제안하였다. SRRI의 개발을 위해 1단계에서는 다양한 유량 조건에서 EFDC 동수역학모형을 이용하여 수리 인자들의 공간적 분포를 재현한 후, 2단계에서는 퍼지합성법 (FSE)를 적용하여 수리인자의 모든 소속도와 가중치를 종합하여 하천 지점별 하천친수지수를 산정하였다. 개발한 SRRI를 낙동강-금호강 합류부에 적용한 결과, 유량 및 지형 조건에 따라 각 수리인자가 친수활동 안전성에 미치는 영향이 공간적으로 매우 상이하게 나타났다. 유향(흐름 방향)은 합류지점 부근에서 친수활동의 위험성을 크게 증가시키는 반면, 사행구간에서는 수심이 중요한 요인으로 나타났다. 고유량 조건에서는 유속이 세일링 및 패들링에서 가장 큰 영향을 미치는 요소로 작용하였다. 특히 세일링은 유량 변화에 민감하여 고유량시에는 주흐름부와 합류부 부근을 제외하고 일부 공간에서만 안전하게 이용이 가능한 것으로 나타났다. 반면 무동력 및 저동력보트는 유량 변화에 덜 민감하여 고유량 조건에서도 부분적으로 허용될 수 있었지만 사행구간의 고수심부에서는 위험 등급으로 권고되었다. 이러한 결과를 바탕으로 SRRI는 다양한 수리학적 조건을 기반으로 공간적 안전정보를 제공함으로써 많은 이용자들이 하천에서 보다 안전한 친수활동을 즐기는 데에 기여할 수 있을 것으로 판단된다.

• 한국사회정책
• /
• v.20 no.3
• /
• pp.47-76
• /
• 2013

The discussion of "new risks" in the field of social policy started to gain attention in the late 1990s and it is commonly argued that new risks are provoked by deindustrialization and/or globalization being more concentrated among the young, women and low skilled individuals. This study commences its inquiry with a conceptualization of labour market risk in an attempt to critically rethink the argument of new risk. A reevaluation of the concept is followed by an empirical investigation on the different types of risks and their changes by different degree. Eight-teen countries are selected in order to provide a comparative account to understand new risk. These are comparatively analyzed using the fuzzy-set ideal type approach to discover different types of social risks and to measure degrees of changes in relation to social risk. In sum, this paper aims to answer: what is new risk? and how do the characteristic of labour market risks differ in different post-industrial countries? The findings suggest that the types of risk are diverse and the speed or the directions of shift are also diverse.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.3039-3041
• /
• 1999

본 연구에서는 과수원에서 사용되는 농약살포용 스피드 스프레이어기(SS기)의 자율주행을 위한 시스템을 개발하였다. 과거의 SS기의 자율주행을 실현한 시스템은 대부분 고가의 장비를 사용하거나 실용성이 결여된 점에 비추어 본 연구에서는 실용가능성에 역점을 두어서 저가의 실용 가능한 시스템을 구성하고자 하였다. 개발된 자율주행 시스템의 특징으로는 1) 하드웨어 시스템을 고성능 프로세서를 이용하여 중앙집중식으로 개발 하였으며 2) 초음파센서의 보정 및 보강을 통해서 오차를 최소화하는 알고리즘을 개발하였으며 3) 현장에서 사용시 진동등의 noise에 강한 robust 시스템을 구성하였으며 4) 자율주행을 제어하기 위해서 퍼지알고리즘을 도입한 지능제어를 실현하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.31-31
• /
• 2017

북한강은 한강의 제 1 지류로 지리적 이점과 풍부한 유량으로 많은 댐들이 건설 되었다. 북한강 본류에는 6개의 크고 작은 댐이 있는데 이들 중 4개가 홍수조절 능력이 없는 발전용댐이다. 이 댐들은 발전효율 향상과 범람으로 인한 저수지 붕괴에 대비하기 위하여 수위를 일정하게 유지하는 방식으로 운영되고 있다. 그러나 최근에 발생하는 집중호우와 게릴라성 폭우는 이러한 목표를 달성하기 힘들게 하고 있다. 지금까지 댐 운영을 위한 유입량 예측에 관련된 연구는 많이 있었지만 대부분 예측 단위가 1시간 이상이었다. 본 연구의 대상이 된 북한강 발전용댐은 저수용량이 작아 적은 양의 유입에도 수위가 급변하는 특징이 있기 때문에 1시간 단위로 유입량을 예측하는 모형은 실제 운영에 적용하기 어렵다. 본 연구에서는 댐의 운영자가 수문 개도 의사결정에 참고할 수 있는 유입량 자료를 생성하기 위하여 10분 단위 유입량을 예측하였다. 또한 댐들이 직렬로 배치된 유역의 특징을 고려하여 상류에 댐이 있는 경우에는 상류 댐의 방류량을 고려하였다. 본 연구는 뉴로 퍼지 기법을 사용하였으며 2004~2016년까지 발생한 호우 사상을 이용하여 모형을 구성하고 검증하였다.

• 한국사회정책
• /
• v.25 no.1
• /
• pp.345-375
• /
• 2018

This study examines labour market and unemployment protection policies as a configuration in 12 OECD countries in order to investigate how countries from different regime conform to or diverse from previous welfare state regime discussion, and to examine its relationship with poverty and inequality. In analyzing the combination of the unemployment insurance, the unemployment assistance, and active labour market policy, firstly, fuzzy scores of unemployment insurance was calculated by analyzing the strictness of eligibility, duration of benefit and the generosity of income replacement rate. For unemployment assistance, the ratio of public assistance expenditure to the GDP in each country and the ratio of unemployment benefit level to the average wage in each country have been considered. As for the active labour market policy, the total expenditure per GDP of this policy was converted into fuzzy points and analyzed. As a result, 5 types in 2005 and 6 types in 2010 were generated. Specifically, 'assistance type(iAp)', 'insurance type (Iap)', 'comprehensive safety net type (IAP)', 'weak safety net type(iap)' were analyzed. This paper suggested policy implication for South Korean case, which consistently had high score for weak safety net type(iap).

• Korean Journal of Construction Engineering and Management
• /
• v.13 no.2
• /
• pp.70-79
• /
• 2012

The Korean Government recently has been focused on strengthening competitiveness of order and stimulating construction market in the international construction industry. It has planned to extend the ODPs (overseas development projects) in order to diversifying the international construction market of which is domestic construction companies, placing too much emphasis on plant projects of the Middle East. However, literature review of risk analysis in the ODPs shows that the number of case study is several. Therefore, Authors asserted the necessity of risk analysis in the ODPs. The purpose of this study is to suggest a methodology that find KRFs (key risk factors) in the ODPs and analyze them, using AHP and Fuzzy theory. As a result, the 37 KRFs are selected and explained characteristics of them. A future direction of this study is to suggest a risk management model in the ODPs and prove feasibility of it.