Scott Andrew Burton

Efficient Supersonic Air Vehicle Preliminary Conceptual Multi-Disciplinary Design Optimization Results

The Air Force Research Lab (AFRL) is investigating concepts for the next generation Efficient Supersonic Air Vehicle (ESAV). Research efforts are currently on going in the form of contracted efforts with the major airframers and internal research that is collaborative with academic partners. One of the goals of the research efforts is to investigate new design methods that can be applied to this class of vehicle. Accurate performance and especially weight estimation early in the design cycle are a priority. Simultaneous design of major sub-systems is another. With advance technologies coming from adaptive versatile engine technology for the propulsion system, active aero elastic wing, flutter suppression, gust load alleviation and maneuver load control, tailless supersonic for the airframe / control system, and directed energy weapon systems stressing the power and thermal sub-systems, the system design problem for this platform is complex beyond any existing aircraft flying today. This places pressure on the design methods to accurately predict the performance and weight of concepts accurately early in the design cycle. This paper presents initial efforts that are focused on airframe and stability control technology impacts on weight and performance in the conceptual design phase.

TURBINE BLADE PROBABILISTIC ANALYSIS USING SEMI-ANALYTICAL SENSITIVITIES

Over the years engineering design practices have matured to a point where significant performance gains are not likely obtainable with further enhancement of existing deterministic analyses. Future design improvements will require that engineers address the underlying uncertainty present in the parameters of engineering models. This paper investigates the application and implementation of first-order reliability and finite element semi-analytical sensitivities to the probabilistic analysis of an aircraft engine turbine blade. State-of-theart computer aided design (CAD) techniques are employed to automate and coordinate finite element meshing for shape sensitivity calculations. Random variables are used to model uncertainty in turbine blade shape parameters. Two reliability techniques are examined: the mean-value first-order (MVFO) and the Hasofer-Lind Rackwitz-Fiessler first-order (HLRF) reliability methods. The HLRF requires accurate gradient information which is accomplished via a semi-analytical finite element technique. Computational costs of the probabilistic analyses and a discussion of implementation issues are presented.

Efficient Supersonic Air Vehicle Analysis and Optimization Implementation using SORCER

The Air Force Research Lab’s Multidisciplinary Science and Technology Center is currently investigating conceptual design processes and computing frameworks that could significantly impact the design of the next generation efficient supersonic air vehicle (ESAV). To make the technological advancements required of a new ESAV, the conceptual design process must accommodate both lowand high-fidelity multidisciplinary engineering analyses. These analyses may be coupled and computationally expensive, which poses a challenge since a large number of configurations must be analyzed. In light of these observations, a design process described herein uses the SORCER (Service-Oriented Computing Environment) software to combine propulsion, structures, aerodynamics, performance, and aeroelasticity in a multidisciplinary analysis (MDA) of an ESAV. The SORCER engineering software provides the MDA automation and tight integration to grid computing resources necessary to achieve the volume of analyses required for conceptual design. Details of the SORCER implementation are illustrated through ESAV design studies using a gradient-based optimization method. A discussion of preliminary optimization results and SORCER grid computing integration is provided.

Multistage Low Pressure Turbine Airfoil Shape Optimization using the C 3 Process

3 process uses commercial CAD and meshing software in conjunction with a 3D multistage computational fluid dynamics (CFD) code to analyze and optimize airfoil shapes. The C 3 process takes advantage of a simple gradient-based optimization method adapted for parallel computing to shape airfoils for a three stage LPT. The paper reviews the CFD- based performance results of baseline and optimized designs.

Object Models for Distributed Multidisciplinary Analysis and Optimization (MAO) Environments that Promotes CAE Interoperability

The next generation Best in Class Multidisciplinary Analysis and Design Environments such as FIPER, GLOBUS, and SORCER, use object oriented languages as their “glue” to combine existing monolithic applications. With the emergence of object oriented languages such as C++ and Java, it is appropriate to apply object oriented techniques to the development of application wrappers and the exposure of data in the computing environment. Such an approach greatly enhances the reuse of wrappers, communication between wrapped applications, and the ability to replace existing applications with others (i.e., plug and play). Properly designed object models promote a level of interoperability between analysis and design applications that perform similar functions. The current effort presents requirements for a Distributed MAO Environment and several object models for engineering data, analysis and design methods.

Multidisciplinary Analysis with SORCER using Domain-Specific Objects

The Air Force Research Lab’s (AFRL) Multidisciplinary Science and Technology Center (MSTC) is investigating conceptual design processes and computing frameworks that could significantly improve air vehicle designs. The AFRL-developed computing framework SORCER (Service-Oriented Computing Environment) has been used to implement multidisciplinary analyses (MDAs) of several air vehicles. One of the challenges of using the flexible service-based, network-centric SORCER framework is the complexity faced by users when developing integrated MDA models. To help alleviate this concern, Java-based domain-specific objects (DSOs) are proposed to simplify the implementation of SORCER service operations and integrated MDA models. The design and use of several DSOs are investigated in this paper and illustrated with examples.

TURBINE BLADE RELIABILITY-BASED OPTIMIZATION USING A VARIABLE- COMPLEXITY METHOD

This paper investigates the application and implementation of an aircraft engine turbine blade reliability-based optimization (RBO). The turbine blade is designed to be minimum volume while satisfying component reliability-based constraints on displacement and stress. Design variables consist of computer aided design (CAD) shape parameters. Uncertainty is introduced via random variable models of material and load parameters. A sequential qaudratic programming (SQP) technique is used in conjunction with first-order reliability theory to design the blade. State-of-the-art CAD techniques are employed to automate and coordinate the necessary finite element analyses required by the optimization and reliability algorithms. Three RBO constraint evaluation techniques are examined: the mean-value (MV) firstorder reliability method, the Hasofer-Lind RackwitzFiessler first-order reliability method (FORM), and a variable-complexity (VC) approach. Computational costs of the different RBO strategies and a discussion of implementation issues are presented.

Application of Approximate Optimization to Turbine Blade Design In a Network -Centric Environment

A trust region based approximate optimization strategy has been implemented within a network centric environment. The strategy sequentially builds response surface approximations of the objective and constra ints based on current and previous design data. These approximations are optimized subject to move limits that are updated depending on the trust region ratio (the trust region ratio measures the goodness of approximations. The strategy has been applied to turbine blade design and optimization. The results indicate that the strategy has good convergence properties and is able to converge quickly compared to traditional optimization.