고정한 이나눔 2017-02 24404 https://aurora.ajou.ac.kr/handle/2018.oak/11179 학위논문(석사)--아주대학교 일반대학원 :산업공학과,2017. 2 1. Introduction 1 2. Review of the State-of-the-Art Research 4 3. Development of Direct-write AM Systems 7 3.1. Powder materials 7 3.2. Nozzle sizes 8 3.3. Agitation system 9 3.4. Overall experimental setup 10 4. Plan of the Experiments 11 4.1. Expression of the uniformity in powder deposition 11 4.2. Screening experiments: selection of variables and their ranges 13 4.3. Model selection using a 2³ design with center point replicates 15 4.4. Design of the experiments for an RSM 18 5. Results and Discussion 20 5.1. Experiment results and additional statistical treatment 20 5.2. Statistical analysis: response surface model 21 5.3. Identification of the optimal condition and validation 27 6. Conclusions 31 References 32 Abstract (Korean) 34 Appendix 35 eng The Graduate School, Ajou University 아주대학교 논문은 저작권에 의해 보호받습니다. 차세대 직접 분말 분사 방식 적층 조형 시스템을 위한 통계 분석 기반 최적 분말 분사 조건 결정 Nanum Lee Thesis 아주대학교 일반대학원 Nanum Lee 일반대학원 산업공학과 2017. 2 Master http://dcoll.ajou.ac.kr:9080/dcollection/jsp/common/DcLoOrgPer.jsp?sItemId=000000024404 Direct-write Powder 3D-Printing Response Surface Methodology Functionally Graded Materials The direct-write powder-based additive manufacturing system possesses great potential to produce functionally graded materials (FGM) that require complex material property change within a product. The stable powder deposition is critical for successfully developing a new direct-write powder additive manufacturing system using bio-plastic powders. This thesis focuses on identifying the optimal process conditions of the stable powder deposition for the new additive manufacturing system. This study uses statistical analysis to uncover the relationship of the process parameters with the powder deposition consistency. The statistical analysis considers main process conditions such as the dispensing nozzle diameter, powder size range, powder type, and agitation frequency. The linearity of the cumulative deposition over time is selected as the criteria to compare the deposition consistency. The study identifies the most suitable statistical model and parameter combinations by examining the curvature of the linear model using replicated center points. The statistical analysis also evaluated the lack-of-fitness of the suggested quadratic models. In addition, this study characterizes an appropriate method to handle exceptional cases of no powder flow data by replacing this missing data value with some values in the dataset. Through these statistical analysis, this study presents a response surface model that best describes the optimal parameter combinations of stable linear powder deposition and verifies the optimal conditions. This study will help develop new additive manufacturing systems and create superior FGM structure.