Nhu-Van Nguyen

Possibility-Based Multidisciplinary Optimisation For Electric-Powered Unmanned Aerial Vehicle Design

This paper describes a possibility-based multidisciplinary optimisation for electric-powered unmanned aerial vehicles (UAVs) design. An in-house integrated UAV (iUAV) analysis program that uses an electric-powered motor was developed and validated by a Predator A configuration for aerodynamics, weight, and performance parameters. An electric-powered propulsion system was proposed to replace a piston engine and fuel with an electric motor, power controllers, and battery from an eco-system point of view. Moreover, an in-house Possibility-Based Design Optimisation (iPBDO) solver was researched and developed to effectively handle uncertainty variables and parameters and to further shift constraints into a feasible design space. A sensitivity analysis was performed to reduce the dimensions of design variables and the computational load during the iPBDO process. Maximising the electric-powered UAV endurance while solving the iPBDO yields more conservative, but more reliable, optimal UAV configuration results than the traditional deterministic optimisation approach. A high fidelity analysis was used to demonstrate the effectiveness of the process by verifying the accuracy of the optimal electric-powered UAV configuration at two possibility index values and a baseline.

Subsonic Airfoil and Flap Hybrid Optimization Using Multi-Fidelity Aerodynamic Analysis

A method for airfoil and slotted ap design optimization using hybrid optimization approach and multidelity aerodynamic analysis was developed. The procedure is based on global optimization using medium delity aerodynamic solvers to localize the optimum design search region and local gradient-based optimization using combination of medium delity solver, high delity CFD solver and surrogate models. The geometry parameterization of airfoil is performed using CST method. Flap parameterization method was developed using Piecewise Bezier curves to capture speci c geometry of slotted ap. Designed airfoil and ap show good aerodynamic performance for given ight conditions of VLA category aircraft: low cruise drag and high lift at landing speed using slotted ap and without it. The aerodynamic analysis method shows good agreement with experimental results due to combination of strong points of each solver.

Multidisciplinary General Aviation Aircraft Design Optimizations Incorporating Airworthiness Constraints

In this paper, aircraft design is explained as a Multidisciplinary Design Optimization (MDO) which incorporates airworthiness requirements. In a preliminary sizing phase, preliminary aircraft configurations are defined from a user-requirements analysis and conventional conceptual design process. Conceptual design and analysis modules are developed and integrated into a multidisciplinary optimization to improve the aircraft conceptual design process. Multidisciplinary feasibility (MDF) is established by implementing a multidisciplinary analysis which couples a preliminary sizing module with a conceptual design and analysis module. During an aircraft application certification phase, design constraints are established by a Design-Certification Related Table (DCRT) and selected by investigating aircraft safety requirements including Korea Airworthiness Standards (KAS) and Federal Aviation Regulations (FAR). By carefully selecting design variables, a multidisciplinary design optimization is performed. The case study is general aviation aircraft design to demonstrate the feasibility and effectiveness of rapid aircraft conceptual design. Minimization of take-off gross weight of general aviation aircraft is performed and the design result shows the feasibility and effectiveness of the aircraft conceptual design process. I. Introduction ecently in Korea, the aircraft industry has been trying to replace flight aircraft with general aviation aircraft for personal air vehicles and training aircraft for airline pilots. For this reason, airworthiness has gained greater importance recently. The logical and systematic aircraft design process has been developed to rapidly obtain an optimum configuration at an early stage of aircraft design. Generally, aircraft design optimization based on semiempirical equations has been well-established since the beginning of aircraft design. These methods, based on Jan Roskam 5 , Raymer 6 methods, GASP 7 , and ACSYNT 8 , have been used extensively and efficiently to rapidly acquire the results of analysis in MDO frameworks. However this aircraft design process does not consider airworthiness requirements at the conceptual design stage. Airworthiness is necessary during the entire aircraft design process to ensure hazard identification and risk management. A certification of airworthiness shall be issued by the contracting state on the basis of satisfactory evidence that the aircraft complies with the design aspects of appropriate airworthiness requirements In this paper, the proposed approach is an aircraft design process incorporating airworthiness requirements. The scope of application for airworthiness requirements is set and design constraints are generated by analyzing the relationship between design variables and certification requirements. Analysis modules are well-developed and integrated into the multidisciplinary optimization environment to refine the aircraft configuration during synthesis of the conceptual design. Eventually, the Multidisciplinary Feasibility technique is applied to resolve the coupling variable in the conceptual aircraft design synthesis process. The case study is the design of a regional jet aircraft to demonstrate the feasibility and effectiveness of rapid aircraft configuration generation

Multidisciplinary Unmanned Combat Air Vehicle-UCAV Design Optimization Using Variable Complexity Modeling

To improve design results while avoiding expensive computational cost, a variable complexity modeling (VCM) framework is studied and presented. The logical and systematical approach for aircraft conceptual design synthesis process is established. The code development and validation for each phase have been done for several types of aircraft. Moreover, the phase II parameterization and integration procedure are developed which allows the linking of design variables to to tal take-off gross weight (TOGW). The lift and drag are evaluated by using the high fidelity model in which lift coefficient is integrated in phase II by VCM approach and drag coefficient is used as an offline design value to the design process at cruise condition. The case study is applied to design UCAV and minimize the TOGW of UCAV.

Development of Repetitively Enhanced Neural Networks (RENN) for Efficient Missile Design and Optimization

An improved approach for design optimization of air intercept missile is developed and presented. A Bayesian learning technique is mapped into Back-propagation neural networks (BPNN) to establish an accurate and effective system approximation, namely an enhanced neural network module. Then, the surrogate models are generated and sent to a hybrid optimizer in which a tentative optimum result is obtained and updated into the training data to refine the response surfaces. This process, which is called Repetitively Enhanced Neural Networks (RENN), is executed repeatedly to refine the response surface until the convergent optimum solution is obtained. A numerical example and a two-member frame design are presented and discuss to demonstrate the accuracy and feasibility of RENN. Eventually, this RENN approach is applied to re-design the air intercept missile-AIM

Multidisciplinary Regional Jet Aircraft Design Optimization Using Advanced Variable Complexity Techniques

To improve design results while avoiding expensive computational cost, an advanced variable complexity modeling (AVCM) framework is studied and presented. AVCM is introduced by using Neural Network as a scaling method to low-fidelity model in the optimization loop in which trust region management strategy is implemented to ensure the matching of low-fidelity models to high-fidelity results. The logical and systematical approach for aircraft design synthesis program-ADSP is developed and validated by three regional jet aircrafts. Then, AVCM is integrated successfully under ADSP in multidisciplinary RJA design optimization to improve the accuracy of design method without any noticeable increase in design turnaround time Minimization of Take-off Gross Weight-TOGW is performed and the design result shows the feasibility and effectiveness of the present AVCM framework under MDO problem.

Stability Analysis of Full Geometry Aircraft through CFD and Response Surface Method

Aircraft stability derivatives have traditionally been analyzed by wind tunnel tests or low fidelity models. The experimental method should be the most accurate but it requires a long lead time. On the other hand, the low fidelity approach is the fastest method but it has an accuracy issue. To overcome these difficulties, a high fidelity numerical approach using computational fluid dynamics analysis combined with response surface method is proposed to estimate static stability characteristics of a low speed aircraft in this paper. A set of aerodynamic coefficients and static stability derivatives of the wing-body-tail model at low Mach number of 0.16 have been computed using Reynold’s Averaged Navier-Stokes CFD analyses and compared with the available wind tunnel data. A set of aerodynamic coefficients with respect to a minimal number of angles of attack and angles of sideslip have been calculated using the CFD code at first. Then the complete set of aerodynamic coefficients and static derivatives, which cover a whole angle-of-attack and angle-of-sideslip range, have been generated using the response surface method. The predicted results demonstrate the utility of the present approach by showing good agreement with the experiment. Finally, in order to predict stability derivatives of subsonic aircraft in the operating condition, proposed techniques are applied to Mach number 0.5.

The repetitive optimization design strategy using neural network and hybrid algorithm

In this paper, a Bayesian learning technique, mapped into feed-forward artificial neural networks, is considered as a system approximation, which, for training, highly non-linear and implicit complex functions. This process is integrated with a hybrid algorithm (HA) in the proposed design optimization strategy. The combination of the back-propagation Levenberg-Marquardt (BPLM) algorithm and the Bayesian learning technique shows good and accurate generalization, which creates the meta-model, considered as the fitness and constraints function in the hybrid algorithm. Here, a genetic algorithm (GA), hybridized with a local gradient-based method, performs the effective and robust evolutionary search and reduces the computation cost. D-optimality is used to select the appropriate points in the design space, to obtain the significant responses. A numerical example, the design of a two-member frame and air intercept missile-AIM design optimization problem are presented to demonstrate the accuracy and feasibility of the process.