http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Misganaw Abebe,박지우,김정,강범수 한국정밀공학회 2017 International Journal of Precision Engineering and Vol.18 No.2
Recently, a trend towards lightweight components in the automobile and aircraft industries has developed as one of the primary paths to improve the fuel consumption rate and reduce environmental pollution. Therefore, various sheet materials such as composite and fiber reinforced metal laminates, and metallic alloys are used to satisfy this trend. However, to ensure good impact performance of the automotive structural components and to reduce the production cost, metallic alloy sheets are thought to be the most attractive for sheet blank materials. In this study, the formability of the selected metallic alloys sheet materials such as AA3003-H14, AZ31B, DP600, is investigated on multi-point dieless forming (MDF). For verifying the formability, wrinkling and dimple occurrence, a limit index is newly proposed for convex and saddle target shapes based on: cushion thickness, curvature radius, bank thickness, and punch stroke. Numerical simulations are performed on ABAQUS commercial software with consideration of the springback behavior. An ordinary Kriging model prediction technique is applied to replace the computationally expensive finite element simulation to find the wrinkling and dimple occurrence limit. For forming limit index validation, one optimal case is conducted, and the simulation and experimental result show the product is wrinkling and dimpling free.
Multi-Objective Six-Sigma Approach for Robust Optimization of Multi-Point Dieless Forming Process
Misganaw Abebe,Junseok Yoon,강범수 한국정밀공학회 2020 International Journal of Precision Engineering and Vol.21 No.10
The significance of the stochastic nature of a multi-pointdieless forming (MDF) process for final product accuracy has been extensively recognized. It is an under-research area in terms of effective techniques and computational tools for easy industrial applications and implementation. It is also recognized that wrinkling and dimpling are the most MDF product defects. Deterministic optimization using the finite element (FE) analysis have been used to improve the formability of the MDF process; however, not considering the noises of the system could affect the variation of the final product quality. In this study, the uncertainty parameters are considered to avoid the random variation of the final product accuracy while using the same control process parameters. In addition, to avoiding computationally expensive FE analysis, the approximation model has been used for both product defects. A six-sigma robust optimization is applied to obtain both reliable and robust MDF products. Finally a numerical simulation with one experimental result of a given target shape shows that the wrinkling and dimpling defect is controlled using the obtained optimal process parameter setup.
Ship collision avoidance route planning using CRI-based A∗ algorithm
서찬희,노유정,Abebe Misganaw,Kang Young-Jin,박선영,권철현 대한조선학회 2023 International Journal of Naval Architecture and Oc Vol.15 No.-
This study presents a novel ship route planning algorithm that takes into account both operational economy and safety by integrating the A* algorithm with a collision avoidance algorithm that evaluates the Collision Risk Index (CRI) between the own ship and the target ship. The CRI-based A* algorithm defines a penalty zone, allowing the own ship to explore safe routes based on the International Regulations for Preventing Collisions at Sea 1972 (COLREGs) and performs an adaptive and effective node search on an extended local map grid according to various encounter situations. The proposed algorithm is validated through simulations of head-on, fine-broad crossing, converging crossing, and overtaking encounters, indicating an economical and safe optimum route compared to conventional ship domainbased route planning.
A Hybrid Model of ARIMA and LSTM for Ship Trajectory Prediction
Yoojeong Noh,Misganaw Abebe,Chanhee Seo 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.11
Every ship may have its own ways of sailing in the sea, however, to avoid the expected collision risk the trajectories of the encounter ship should be predicted to make a proper decision. In this study, we explore a hybrid of Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) models to estimate the surrounded ship trajectories using Automatic Identification System (AIS) data. A moving-average filtering was applied to decompose the data into linear and nonlinear components. ARIMA model was applied on the linear part of the data, and LSTM model was applied on the nonlinear part of the data. The accuracy of the model has been tested and validated under different situations with ARIMA, LSTM, and the previously suggested hybrid model. Finally, the model accuracy result shows that the proposed model can estimate the future trajectory well enough.
Ship collision avoidance route planning using CRI-based A<sup>*</sup> algorithm
Chanhee Seo,Yoojeong Noh,Misganaw Abebe,Young-Jin Kang,Sunyoung Park,Cheolhyeon Kwon The Society of Naval Architects of Korea 2023 International Journal of Naval Architecture and Oc Vol.15 No.-
This study presents a novel ship route planning algorithm that takes into account both operational economy and safety by integrating the A<sup>*</sup> algorithm with a collision avoidance algorithm that evaluates the Collision Risk Index (CRI) between the own ship and the target ship. The CRI-based A<sup>*</sup> algorithm defines a penalty zone, allowing the own ship to explore safe routes based on the International Regulations for Preventing Collisions at Sea 1972 (COLREGs) and performs an adaptive and effective node search on an extended local map grid according to various encounter situations. The proposed algorithm is validated through simulations of head-on, fine-broad crossing, converging crossing, and overtaking encounters, indicating an economical and safe optimum route compared to conventional ship domain-based route planning.
CRI와 COLREG를 이용한 선박 충돌 회피 알고리즘
서찬희(Chanhee Seo),노유정(Yoojeong Noh),미쓰가나 아베베 바예(Misganaw Abebe Baye) 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.11
In recent years, unmanned shipping of ships has continued to evolve and has come to be considered commercially viable. Path planning is an important process for unmanned operations. However, there are many variables at sea. And they should be considered when routing for economy and safety. This study aims to ensure the economics and safety of ships ’operation by adding collision avoidance algorithm to an improved A<SUP>*</SUP> algorithm, a ships’ route finding optimization algorithm considering the economics of operations. For this, the algorithm is constrained by applying the Collision Risk Index (CRI) and COLREGs rule to the path optimization algorithm. To validate the proposed algorithm, The collision avoidance algorithm was performed in various maritime traffic scenarios, and the results of using CRI were compared to those using other ship domain methods.