• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.585-586
• /
• 2020

공정 데이터를 실시간으로 수집할 수 있는 스마트 팩토리의 장점을 활용하여, 일반적인 기계 학습 대신 강화 학습을 사용한다면 미리 요구되는 훈련 데이터 없이 행동 제어를 할 수 있다. 하지만, 현실 세계에서는 물리적 마모, 시간적 문제 등으로 인해 수천만 번 이상의 반복 학습이 불가능하다. 따라서, 본 논문에서는 시뮬레이터를 활용해 스마트 팩토리 분야에서 복잡한 환경 중 하나인 이송 설비에 초점을 둔 그리드 분류 시스템을 개발하고 협력적 다중 에이전트 기반의 강화 학습을 설계하여 효율적인 행동 제어가 가능함을 입증한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.662-665
• /
• 2024

IT 기술이 발전하며 실시간 전력 수요량 및 가격 등을 파악할 수 있는 스마트 그리드가 주목을 받고 있다. 스마트 그리드 환경에서는 에너지 저장 장치를 이용하여 소비자의 경제적 부담을 덜어낼 뿐만 아니라 에너지를 효율적으로 사용할 수 있다. 본 연구에서는 이러한 목표를 위해 과거 2시간 동안의 부하량 및 가격을 바탕으로 에너지 저장 장치의 충전 및 방전량을 결정하는 강화학습 알고리즘을 제안한다. 또한, 여러 강화학습 기법의 성능을 비교 분석한다.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.25 no.6
• /
• pp.294-298
• /
• 2015

According to the results of "Highway Bridges Long Life Repair Plan." The most serious damage to RC slabs is caused by fatigue deterioration, which results from the driving loads of large-sized vehicles, and aging of materials. In response to this, adhesion reinforcement using carbon fiber sheet is being adopted. In addition, carbon fiber strand sheet that holds the same material characteristics as CFS, but has superior workability, has been developed as a new reinforcement material. However, almost no studies have been conducted on CFSS in relation to fatigue resistance evaluation through fatigue tests under running wheel loads, with the exception of a few by some organizations. Therefore, in this study, specimens with front CFS adhesion reinforcement on the bottom surface of the RC slab and specimens with grid-type CFSS reinforcement were manufactured. Then, fatigue tests under running wheel loads were conducted, and thus fatigue resistance was evaluated using the specimens.

• International Journal of Highway Engineering
• /
• v.7 no.2 s.24
• /
• pp.1-12
• /
• 2005

This study was conducted to evaluate effect of geogrid and fabric, which are used underneath the overlaid asphalt pavement for retarding reflection cracking by simulated laboratory test. In this study, an interlayer at the interface between old concrete pavement surface and overlaid asphalt mixture, and polymer-modifier were used as an effort of retarding reflection crack initiation and for strengthening mixture. Five products were used in preparation of asphalt concrete beam specimen which was tack coated on top of jointed concrete block. Simulated Mode I and II fracture test were conducted under wheel loading and results were compared among those products. From the test results, several material and reinforcement combinations were observed to have a significant retardation effect against reflection cracking. The most effectively strengthened pavement against reflection cracking was found to be the LDPE-modifier asphalt mixture with a grid reinforcement at the bottom.

• Journal of Practical Engineering Education
• /
• v.14 no.1
• /
• pp.219-224
• /
• 2022

Q-Learning is a technique widely used as a basic algorithm for reinforcement learning. Q-Learning trains the agent in the direction of maximizing the reward through the greedy action that selects the largest value among the rewards of the actions that can be taken in the current state. In this paper, we studied a policy that can speed up agent training using Q-Learning in Frozen Lake 8×8 grid environment. In addition, the training results of the existing algorithm of Q-learning and the algorithm that gave the attribute 'direction' to agent movement were compared. As a result, it was analyzed that the Q-Learning policy proposed in this paper can significantly increase both the accuracy and training speed compared to the general algorithm.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.11 no.1
• /
• pp.11-22
• /
• 2009

Tunnel excavation with several sections and appropriate auxiliary measures such as face bolt and pre-grouting are widely used in case of weak and less rigid ground for the stability of a tunnel face during excavation. This papers first described the evaluation methods proposed in technical literature to maintain the tunnel face stable, and then studied by FEM analysis whether face reinforcement is need in what degree of ground deformation and strength features for the stability of a tunnel face when excavating by full excavation with sub-bench. Lastly, a three dimensional FEM analysis was performed to study how the tunnel face itself and the ground around the tunnel behave depending on different bolt layouts, length of bolts, number of bolts. There were relative differences in comparison of results on the stability of a tunnel face by a theoretical evaluation methods and FEM analysis, but the same in reinforced effect of face. It was found that the stability of a tunnel face can be obtained with face bolt installed longer than 1.0D (tunnel width), bolt density of about 1 bolt per every $1.5\;m^2$ (layout of grid type), and reinforcement area of $120^{\circ}$ arch area of upper section.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.3
• /
• pp.859-864
• /
• 2015

This paper analyzes the impact of Plug-in Electric vehicles(PEVs) on power demand and voltage change when PEVs are connected to the domestic distribution system. Specifically, it assesses PEVs charging load by charging method in accordance with PEVs penetration scenarios, its percentage of total load, and voltage range under load conditions. Concretely, we develop EMTDC modelling to perform a voltage distribution analysis when the PEVs charging system by their charging scenario was connected to the distribution system under the load condition. Furthermore we present evaluation algorithm to determine whether it is possible to adjust it such that it is in the allowed range by applying ULTC when the voltage change rate by PEVs charging scenario exceed its allowed range. Also, detailed analysis of the impact of PEVs on power distribution system was carried out by calculating existing electric power load and additional PEVs charge load by each scenario on new-town in Korea to estimate total load increases, and also by interpreting the subsequent voltage range for system circuits and demonstrating conditions for countermeasures. It was concluded that total loads including PEVs charging load on new-town distribution system in Korea by PEVs penetration scenario increase significantly, and the voltage range when considering ULTC, is allowable in terms of voltage tolerance range up to a PEVs penetration of 20% by scenario. Finally, we propose the charging capacity of PEVs that can delay the reinforcement of power distribution system while satisfying the permitted voltage change rate conditions when PEVs charging load is connected to the power distribution system by their charging penetration scenario.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.14 no.3
• /
• pp.169-176
• /
• 2011

The purpose of this study is to examine the ecologically suitable restoration characteristics in riparian slopes constructed by continuous fiber soil reinforced system (Geofiber system) which does not contain the concrete materials. The findings are as follows : (1) as the tested soil was not washed away by rainfalls and floods, Geofiber could replace the concrete wall and gravity stone net bag technique from the civil engineering structural point of view; (2) after one year of the construction, it was monitored that land cover ratio was 80-90%, which indirectly shows that vegetation is safely maintained; and (3) at the same time, 5-8 flora species were found in each test grid and more importantly dominant species have been moved from alien species to native herbaceous plants. From the above findings, Geofiber system is recommendable to restore the riparian slopes in terms of stability and natural landscape points. However, a long term monitering is needed considering flora succession process in a given environment as well as suitability tests should be carried out through the comparative investigations in other environments.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.990-993
• /
• 2019

인공지능(AI)은 최근 다양한 산업과 사회에서 패러다임을 바꾸고 있지만, 최첨단 AI 가 제조업에서는 즉각적인 성과를 보이지 못 하고 있다. 다시 말해, Industry 4.0 시점에서 기존의 접근 방법과 차별화되는 실용적인 방법론이 필요하다. 여기서 중요한 점은 '어떤' 데이터를 '어떻게' 활용하여 '어느' 부분에 적용할 것 인가이다. 제조업은 게임과 같이 가상의 캐릭터가 하나의 객체 단위로 구동되는 것이 아니라 수많은 하드웨어가 물리적으로 조합되어 연동한다. 따라서, 현실 세계에서는 물리적 마모, 고장 등으로 인해 엔지니어의 개입 없이 수천만 번 이상의 반복 학습이 불가능하다. 또, 제조업은 학습을 위한 방대한 양의 데이터를 수집하고 레이블링 하는 것이 매우 어렵다. 이 두 가지 한계를 극복할 수 있는 방법은 현실과 매우 유사한 환경을 시뮬레이션으로 재연 후 강화 학습을 사용하는 것이다. 제조 분야에서 아주 복잡한 환경 중 하나로 이송 설비가 있으며, 본 논문에서는 그리드 분류 시스템을 개발하고 강화 학습을 적용시킬 수 있는 환경을 설계한다.

• International Journal of Computer Science & Network Security
• /
• v.21 no.11
• /
• pp.345-353
• /
• 2021

Cloud computing represent a new era of computing that's forms through the combination of service-oriented architecture (SOA), Internet and grid computing with virtualization technology. Virtualization is a concept through which every cloud is enable to provide on-demand services to the users. Most IT service provider adopt cloud based services for their users to meet the high demand of computation, as it is most flexible, reliable and scalable technology. Energy based performance tradeoff become the main challenge in cloud computing, as its acceptance and popularity increases day by day. Cloud data centers required a huge amount of power supply to the virtualization of servers for maintain on- demand high computing. High power demand increase the energy cost of service providers as well as it also harm the environment through the emission of CO₂. An optimization of cloud computing based on energy-performance tradeoff is required to obtain the balance between energy saving and QoS (quality of services) policies of cloud. A study about power usage of resources in cloud data centers based on workload assign to them, says that an idle server consume near about 50% of its peak utilization power [1]. Therefore, more number of underutilized servers in any cloud data center is responsible to reduce the energy performance tradeoff. To handle this issue, a lots of research proposed as energy efficient algorithms for minimize the consumption of energy and also maintain the SLA (service level agreement) at a satisfactory level. VM (virtual machine) consolidation is one such technique that ensured about the balance of energy based SLA. In the scope of this paper, we explore reinforcement with fuzzy logic (RFL) for VM consolidation to achieve energy based SLA. In this proposed RFL based active VM consolidation, the primary objective is to manage physical server (PS) nodes in order to avoid over-utilized and under-utilized, and to optimize the placement of VMs. A dynamic threshold (based on RFL) is proposed for over-utilized PS detection. For over-utilized PS, a VM selection policy based on fuzzy logic is proposed, which selects VM for migration to maintain the balance of SLA. Additionally, it incorporate VM placement policy through categorization of non-overutilized servers as- balanced, under-utilized and critical. CloudSim toolkit is used to simulate the proposed work on real-world work load traces of CoMon Project define by PlanetLab. Simulation results shows that the proposed policies is most energy efficient compared to others in terms of reduction in both electricity usage and SLA violation.