Configuration Optimization of Airborne Missile Launcher Pod using Multi-Objective Decision Analysis Process Combined with Numerical Optimization
- 주제(키워드) Airborne Missile Launcher Pod (AMLP) , Central Composite Design (CCD) method , Computational Fluid Dynamics (CFD)
- 발행기관 아주대학교
- 지도교수 Euiwhan Kim
- 발행년도 2019
- 학위수여년월 2019. 8
- 학위명 석사
- 학과 및 전공 일반대학원 시스템공학과
- 실제URI http://www.dcollection.net/handler/ajou/000000029022
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
The purpose of this study is to modify and optimize the configuration of the Airborne Missile Launcher Pod (AMLP) to extend its usage from ground or helicopter mounting to the fixed-wing aircraft where the aerodynamic drag becomes important. Many engineering problems need a multi-objective optimization approach to generate an optimal strategy for decision making. The objectives considered in this study are the aerodynamic drag coefficient, the mass, and the cost of the AMLP. The methodology used to find the optimum strategy was the Multi-Objective Decision Analysis (MODA) process which is combined with statistical techniques to generate mathematical regression models of configuration variables. These models were used in the optimization of the system. The drag coefficients were obtained from Computational Fluid Dynamics (CFD) simulation where Central Composite Design (CCD) method was used to decrease the expensive CFD computation. The model for mass (or volume) was generated from the linear regression analysis. Then, the weighting Min-Max method in combination with the constrained method was used to determine the optimal values for the modified AMLP. Weighting combination of objectives was generated to demonstrate the Pareto set for the best solutions. The configuration obtained from this study can be provided to stakeholders for objective evaluations and help to make the decision to choose the best solution.
more목차
List of Tables iv
List of Figures v
List of Abbreviation vi
1 Introduction 1
1.1 2.75-Inch Guided Rocket 1
1.2 Airborne Missile Launcher Pod (AMLP) 2
1.3 Previous Works & Motivation 4
1.4 Purpose of Study 5
1.5 Methodology 6
2 Methods Used for Optimization 7
2.1 MODA Steps Combined with Numerical Optimization 7
2.2 Regression Model Method 15
2.2.1 First Order Regression Model 16
2.2.2 Second Order Regression Model 18
2.2.3 Analysis of Significance in Regression Model 19
2.3 Design of Experiment (DOE) Method 23
2.3.1 Full Factorial Design 24
2.3.2 Central Composite Design (CCD) 25
2.4 CFD Simulation for Aerodynamic Analysis 28
2.5 Aerodynamic Drag Coefficient 29
3 Configuration Optimization 31
3.1 Problem Identification 31
3.2 Objectives Identification 32
3.3 Generate Alternatives 34
3.4 Develop Objective Evaluation 35
3.4.1 Central Composite Design (CCD) Method 35
3.4.2 Linear Regression Method 38
3.5 Develop Value Function & Weighting 39
3.6 Develop Constraints 41
3.7 Deterministic Result 41
4 Analysis of Result 46
4.1 Sensitivity Analysis 46
4.2 Develop Recommendation 54
4.3 ANOVA & Regression Analysis 56
5 Conclusion & Future Work 65
5.1 Conclusion 65
5.2 Future Work. 66
Bibliography 67
