• The Journal of Korea Robotics Society
• /
• v.18 no.2
• /
• pp.143-154
• /
• 2023

In this study, a fast motion planning method for the swing motion of a 6x6 wheel-legged robot to traverse large obstacles and gaps is proposed. The motion planning method presented in the previous paper, which was based on trajectory optimization, took up to tens of seconds and was limited to two-dimensional, structured vertical obstacles and trenches. A deep neural network based on one-dimensional Convolutional Neural Network (CNN) is introduced to generate keyframes, which are then used to represent smooth reference commands for the six leg angles along the robot's path. The network is initially trained using the behavioral cloning method with a dataset gathered from previous simulation results of the trajectory optimization. Its performance is then improved through reinforcement learning, using a one-step REINFORCE algorithm. The trained model has increased the speed of motion planning by up to 820 times and improved the success rates of obstacle crossing under harsh conditions, such as low friction and high roughness.

• Structural Engineering and Mechanics
• /
• v.88 no.2
• /
• pp.179-188
• /
• 2023

The mechanical behaviours of biopolymer-treated soil depend on the formation of soil-biopolymer matrices. In this study, various biopolymers(e.g., xanthan gum (XG), locust bean gum (LBG), sodium alginate (SA), agar gum (AG), gellan gum (GE) and carrageenan kappa gum (KG) are selected to treat three types of natural soil at different concentrations (e.g., 1%, 2% and 3%) and curing time (e.g., 4-365 days), and reveal the reinforcement effect on natural soil by using unconfined compression tests. The results show that biopolymer-treated soil obtains the maximum unconfined compressive strength (UCS) at curing 14-28 days. Although the UCS of biopolymer-treated soil has a 20-30% reduction after curing 1-year compared to the maximum value, it is still significantly larger than untreated soil. In addition, the UCS increment ratio of biopolymer-treated soil decreases with the increase of biopolymer concentration, and there exists the optimum concentration of 1%, 2-3%, 2%, 1% and 2% for XG, SA, LBG, KG and AG, respectively. Meanwhile, the optimum initial moisture content can form uniformly biopolymer-soil matrices to obtain better reinforcement efficiency. Furthermore, the best performance in increasing soil strength is XG following SAand LBG, which are significantly better than AG, KG and GE.

• Advances in nano research
• /
• v.16 no.4
• /
• pp.395-412
• /
• 2024

This article develops a nonclassical size dependent nanoplate model to study the dynamic response of functionally graded carbon nanotube (FGCNT) nanoplates under a moving load. Both nonlocal and microstructure effects are incorporated through the nonlocal strain gradient elasticity theory. To investigate the effect of reinforcement orientation of CNT, four different configurations are studied and analysed. The FGM gradation thorough the thickness direction is simulated using the power law. In the context of the first order shear deformation theory, the dynamic equations of motion and the associated boundary conditions are derived by Hamilton's principle. An analytical solution of the dynamic equations of motion is derived based on the Navier methodology. The proposed model is verified and compared with the available results in the literature and good agreement is found. The numerical results show that the dynamic performance of FGCNT nanoplates could be governed by the reinforcement pattern and volume fraction in addition to the non-classical parameters and the moving load dimensionless parameter. Obtained results are reassuring in design and analysis of nanoplates reinforced with CNTs.

• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1472-1479
• /
• 2024

Cyber-Physical Energy Systems (CPESs) integrate cyber and hardware components to ensure a reliable and safe physical power production and supply. Renewable Energy Sources (RESs) add uncertainty to energy demand that can be dealt with flexible operation (e.g., load-following) of CPES; at the same time, scenarios that could result in severe consequences due to both component stochastic failures and aging of the cyber system of CPES (commonly overlooked) must be accounted for Operation & Maintenance (O&M) planning. In this paper, we make use of Deep Reinforcement Learning (DRL) to search for the optimal O&M strategy that, not only considers the actual system hardware components health conditions and their Remaining Useful Life (RUL), but also the possible accident scenarios caused by the failures and the aging of the hardware and the cyber components, respectively. The novelty of the work lies in embedding the cyber aging model into the CPES model of production planning and failure process; this model is used to help the RL agent, trained with Proximal Policy Optimization (PPO) and Imitation Learning (IL), finding the proper rejuvenation timing for the cyber system accounting for the uncertainty of the cyber system aging process. An application is provided, with regards to the Advanced Lead-cooled Fast Reactor European Demonstrator (ALFRED).

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.116-123
• /
• 2024

In the manufacturing industry, dispatching systems play a crucial role in enhancing production efficiency and optimizing production volume. However, in dynamic production environments, conventional static dispatching methods struggle to adapt to various environmental conditions and constraints, leading to problems such as reduced production volume, delays, and resource wastage. Therefore, there is a need for dynamic dispatching methods that can quickly adapt to changes in the environment. In this study, we aim to develop an agent-based model that considers dynamic situations through interaction between agents. Additionally, we intend to utilize the Q-learning algorithm, which possesses the characteristics of temporal difference (TD) learning, to automatically update and adapt to dynamic situations. This means that Q-learning can effectively consider dynamic environments by sensitively responding to changes in the state space and selecting optimal dispatching rules accordingly. The state space includes information such as inventory and work-in-process levels, order fulfilment status, and machine status, which are used to select the optimal dispatching rules. Furthermore, we aim to minimize total tardiness and the number of setup changes using reinforcement learning. Finally, we will develop a dynamic dispatching system using Q-learning and compare its performance with conventional static dispatching methods.

• Advances in nano research
• /
• v.17 no.3
• /
• pp.257-274
• /
• 2024

This work aims to study and analyse the dynamic size dependent behvior of functionally graded carbon nanotubes (FGCNTs) nanoplates embedded in elastic media and subjected to moving point load. The non-classical effect is incorporated into the governing equations using the nonlocal strain gradient theory (NSGT). Four different reinforcement configurations of the carbon nanotubes (CNTs) are considered to show the effect of reinforcement configuration on the dynamic behvior of the FGCNTs nanoplates. The material characteristics of the functionally graded materials are assumed to be continuously distributed throughout the thickness direction according to the power law. The Hamiltonian principle is exploited to derive the dynamic governing equations of motion and the associated boundary conditions in the framework of the first order shear deformation plate theory. The Navier analytical approach is adopted to solve the governing equations of motion. The obtained solution is checked by comparing the obtained results with the available results in the literature and the comparison shows good agreement. Numerical results are obtained and discussed. Obtained results showed the significant impact of the elastic foundation parameters, the non-classical material parameters, the CNT configurations, and the volume fractions on the free and forced vibration behaviors of the FGCNT nanoplate embedded in two parameters elastic foundation and subjected to moving load.

• Proceedings of the Korean Geotechical Society Conference
• /
• 1999.10a
• /
• pp.185-192
• /
• 1999

This paper presents the results of a parametric study on the behavior of tunnel face reinforced with horizontal pipes. A three-dimensional finite element model was adopted in this study to capture the three-dimensional nature of tunnel face behavior under various boundary conditions. A parametric study was peformed on a wide range of boundary conditions with emphasis on the effect of reinforcing layouts on the deformation behavior of tunnel face. The results of analysis such as tunnel face deformation behavior under various conditions were thoroughly analyzed, and a database for the behavior of tunnel face under different reinforcing conditions was established for future development of a semi-empirical design/analysis method for the tunnel face reinforcing technique. The results indicated that there exits an optimum reinforcing layout for a given tunnel condition, which must be selected with due consideration of tunnel geometry and ground condition.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1999.10a
• /
• pp.701-704
• /
• 1999

The purpose of this study was to investigate the effectiveness of the flexural rehabilitation of the pre-loaded reinforced concrete beams strengthened by the steel plate. Main test parameters were the existence and the magnitude of the pre-loading at the flexural of rehabilitation and the tensile reinforcement ratio of the specimens. Seven beam specimens were tested to investigate the effectiveness of the rehabilitation method. Test results showed that the ultimate load capacities of the pre-loaded specimens were higher than not-pre-loaded specimens at the rehabilitation. The cause of the pharameter was analyzed if is suggested that the bond failure between the concrete and the strengthening steel plate occured prior to the yielding of the tension reinforcement. The member flexural stiffnesses, were similar regardless of the load conditions at retrofit and failure modes showed brittle aspect caused by rip-off failure.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2000.04a
• /
• pp.218-223
• /
• 2000

A both mixing process of electro-magnetic stirring and mechanical process technique were used to fabricate particulate metal matrix composites(PMMCs) for variation of particle size. The PMMCs were tested for their tensile test for with and without heat treatment with T6. PMMCs fabrication processing conditions for both electrical and mechanical process are also suggested. In order to thixoforming of PMMCs, fabricated billet are reheated by using the optimal coil designed as a function of length between PMMC billet and coil surface, coil diameter and billet length. The effect of reinforcement distribution on billet temperature variation are investigated with calculated solid fraction theory proposed as a function of matrix alloy and volume fraction of reinforcement.

• Proceedings of the Korean Geotechical Society Conference
• /
• 1999.10a
• /
• pp.473-480
• /
• 1999

This paper presents the results of a parametric study on the behavior of mechanically stabilized earth retaining wall. It has been recognized that the currently available design guidelines, which is base on the limit equilibrium approach, cannot properly account the interaction effect between the components, construction sequence, and foundation settlement which may impose a significant influence on the wall behavior. A parametric study using finite element analysis was performed to investigate the behavior of MSE wall under different construction conditions and the applicability of the current design approach. In the parametric analysis, the effects of the construction sequence, the surcharge, and the foundation stiffness were studied and a detailed finite element modeling for various components of the system were employed. The results, such as wall displacement and earth pressure distributions, reinforcement forces, vertical stress distribution were then thoroughly analyzed to investigate the effect of construction details on the wall behavior.