• Architectural research
• /
• v.26 no.1
• /
• pp.1-12
• /
• 2024

This study introduces a deep reinforcement learning (DRL)-based methodology for optimizing apartment space arrangements, addressing the limitations of human capability in evaluating all potential spatial configurations. Leveraging computational power, the methodology facilitates the autonomous exploration and evaluation of innovative layout options, considering architectural principles, legal standards, and client re-quirements. Through comprehensive simulation tests across various apartment types, the research demonstrates the DRL approach's effec-tiveness in generating efficient spatial arrangements that align with current design trends and meet predefined performance objectives. The comparative analysis of AI-generated layouts with those designed by professionals validates the methodology's applicability and potential in enhancing architectural design practices by offering novel, optimized spatial configuration solutions.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.12 no.4
• /
• pp.179-186
• /
• 2008

Probing inside of concrete structures is one of the important steps in assessing condition of the structure. For the assessment, electromagnetic induction method and electromagnetic wave method are currently applied to the measurement of cover depth, and the detection of reinforcement embedded in concrete. To determine detection capability of locating reinforcement embedded in concrete, commercially available nondestructive testing (NDT) equipments have been tested. The equipments include electromagnetic wave system and electromagnetic induction system. In the tests, nine concrete specimens which have the dimensions of 1,000mm(length))${\times}$300mm(width) with thickness varying from 125mm to 150mm are used. The reinforcement are located at 45, 60, 100mm depth from the concrete surface. Horizontal reinforcement spacing has been set over 100mm. From the outcome, it is shown that error is increased as the diameter of reinforcement enlarge in case of using electromagnetic induction method. In case of using electromagnetic wave method, the detection of reinforcement embedded in deep is good in the view of reliability because of using the relative permittivity on the real cover depth.

• Proceedings of KOSOMES biannual meeting
• /
• 2017.04a
• /
• pp.275-275
• /
• 2017

• Smart Structures and Systems
• /
• v.13 no.6
• /
• pp.943-957
• /
• 2014

When subjected to fatigue loading, the main failure mode of partially prestressed concrete (PPC) structure is the fatigue fracture of tensile reinforcement. Therefore, monitoring and evaluation of the steel stresses/strains in the structure are essential issues for structural design and healthy assessment. The current study experimentally investigates the possibility of using fiber Bragg grating (FBG) sensors to measure the steel strains in PPC beams in the process of fatigue loading. Six full-scale post-tensioned PPC beams were exposed to fatigue loading. Within the beams, the FBG and resistance strain gauge (RSG) sensors were independently bonded onto the surface of tensile reinforcements. A good agreement was found between the recorded results from the two different sensors. Moreover, FBG sensors show relatively good resistance to fatigue loading compared with RSG sensors, indicating that FBG sensors possess the capability for long-term health monitoring of the tensile reinforcement in PPC structures. Apart from the above findings, it can also be found that during the fatigue loading, there is stress redistribution between prestressed and non-prestressed reinforcements, and the residual strain emerges in the non-prestressed reinforcement. This phenomenon can bring about an increase of the steel stress in the non-prestressed reinforcement.

• Journal of Korea Game Society
• /
• v.18 no.4
• /
• pp.99-106
• /
• 2018

Board games have many game characters and many state spaces. Therefore, games must be long learning. This paper used reinforcement learning algorithm. But, there is weakness with reinforcement learning. At the beginning of learning, reinforcement learning has the drawback of slow learning speed. Therefore, we tried to improve the learning speed by using the heuristic using the knowledge of the problem domain considering the game tree when there is the same best value during learning. In order to compare the existing character the improved one. I produced a board game. So I compete with one-sided attacking character. Improved character attacked the opponent's one considering the game tree. As a result of experiment, improved character's capability was improved on learning speed.

• Advances in Computational Design
• /
• v.3 no.1
• /
• pp.49-69
• /
• 2018

This paper shows an optimal design for reinforced concrete rectangular combined footings based on a criterion of minimum cost. The classical design method for reinforced concrete rectangular combined footings is: First, a dimension is proposed that should comply with the allowable stresses (Minimum stress should be equal or greater than zero, and maximum stress must be equal or less than the allowable capacity withstand by the soil); subsequently, the effective depth is obtained due to the maximum moment and this effective depth is checked against the bending shear and the punching shear until, it complies with these conditions, and then the steel reinforcement is obtained, but this is not guaranteed that obtained cost is a minimum cost. A numerical experimentation shows the model capability to estimate the minimum cost design of the materials used for a rectangular combined footing that supports two columns under an axial load and moments in two directions at each column in accordance to the building code requirements for structural concrete and commentary (ACI 318S-14). Numerical experimentation is developed by modifying the values of the rectangular combined footing to from "d" (Effective depth), "b" (Short dimension), "a" (Greater dimension), "${\rho}_{P1}$" (Ratio of reinforcement steel under column 1), "${\rho}_{P2}$" (Ratio of reinforcement steel under column 2), "${\rho}_{yLB}$" (Ratio of longitudinal reinforcement steel in the bottom), "${\rho}_{yLT}$" (Ratio of longitudinal reinforcement steel at the top). Results show that the optimal design is more economical and more precise with respect to the classical design. Therefore, the optimal design presented in this paper should be used to obtain the minimum cost design for reinforced concrete rectangular combined footings.

• International Journal of Concrete Structures and Materials
• /
• v.10 no.4
• /
• pp.513-525
• /
• 2016

The resistance-demand approach has emerged as an effective approach for determining the shear capacity of reinforced concrete beams. This approach is based on the fact that both the shear resistance and shear demand are correlated with flexural tensile strain from compatibility and equilibrium requirements. The basic shear strength, under a given loading is determined from the intersection of the demand and resistance curves. This paper verifies the applicability of resistance-demand procedure for predicting the shear capacity of high strength concrete beams without web reinforcement. A total of 18 beams were constructed and tested in four-point bending up to failure. The test variables included the longitudinal reinforcement ratio, the shear span to depth ratio, and the beam depth. The shear capacity of the beams was predicted using the proposed procedure and compared with the experimental values. The results of the comparison showed good prediction capability and can be useful to design practice.

• Journal of The Korean Digital Architecture Interior Association
• /
• v.10 no.2
• /
• pp.31-38
• /
• 2010

This study is to suggest a strategy and trend of products development by major construction companies to form a trend and to set the trend in housing market through investigation about brand image campaign for TV ad, individual design items, result of products development and special plans. It is the strategy for products development of construction companies surveyed to be divided into five kinds. First, the change from in the unit to outside. Second, centralization of design powers about products development through submit and the wining of a prize in the domestic & foreign design fair to publicize a excellent quality in a roundabout way. Third, reinforcement of service for support the habitability and design manual for secure uniform quality. Forth, investment in the environment-friendly housing and new recycling energy. Fifth, reinforcement of organizational power for improving production capability.

• Structural Engineering and Mechanics
• /
• v.38 no.4
• /
• pp.529-547
• /
• 2011

This paper presents the global-local finite cover method (GL-FCM) that is capable of analyzing structures involving local heterogeneities and propagating cracks. The suggested method is composed of two techniques. One of them is the FCM, which is one of the PU-based generalized finite element methods, for the analysis of local cohesive crack growth. The mechanical behavior evaluated in local heterogeneous structures by the FCM is transferred to the overall (global) structure by the so-called mortar method. The other is a method of mesh superposition for hierarchical modeling, which enables us to evaluate the average stiffness by the analysis of local heterogeneous structures not subjected to crack propagation. Several numerical experiments are conducted to validate the accuracy of the proposed method. The capability and applicability of the proposed method is demonstrated in an illustrative numerical example, in which we predict the mechanical deterioration of a reinforced concrete (RC) structure, whose local regions are subjected to propagating cracks induced by reinforcement corrosion.

• Transactions on Control, Automation and Systems Engineering
• /
• v.3 no.3
• /
• pp.170-175
• /
• 2001

Q-learning is a kind of reinforcement learning where the agent solves the given task based on rewards received from the environment. Most research done in the field of Q-learning has focused on discrete domains, although the environment with which the agent must interact is generally continuous. Thus we need to devise some methods that enable Q-learning to be applicable to the continuous problem domain. In this paper, an extended fuzzy rule is proposed so that it can incorporate Q-learning. The interpolation technique, which is widely used in memory-based learning, is adopted to represent the appropriate Q value for current state and action pair in each extended fuzzy rule. The resulting structure based on the fuzzy inference system has the capability of solving the continuous state about the environment. The effectiveness of the proposed structure is shown through simulation on the cart-pole system.