Search Results

Heterogeneous physical systems can often be considered as highly complex, consisting of a large number of subsystems and components, along with the associated interactions and hierarchies amongst them. The simulation of a large-scale, complex system can be computationally expensive and the dynamic interactions may be highly nonlinear. One approach to address these challenges is to increase the computing power or resort to a distributed computing environment. An alternative to improve the simulation computational performance and efficiency is to reduce CPU required time through the application of surrogate models. Surrogate modeling techniques for dynamic simulation models can be developed based on Recurrent Neural Networks (RNN).This study will present a method to improve the overall speed of a multi-physics time-domain simulation of a complex naval system using a surrogate modeling technique.

The objective of this paper is to develop a generalized loss management model to account for the usage of thermodynamic work potential in vehicles of any type. The key to accomplishing this is creation of a differential representation for vehicle loss as a function of operating condition. This differential model is then integrated through time to obtain an analytical estimate for total usage (and loss) of work potential consumed by each loss mechanism present during vehicle operation. The end result of this analysis is a better understanding of how the work potential initially present in the fuel, batteries, etc. is partitioned amongst all losses relevant to the vehicle's operation. The loss partitioning estimated from this loss management model can be used in conjunction with cost accounting systems to gain a better understanding of underlying drivers on vehicle manufacturing and operating costs.

System effectiveness has become the prime metric for the evaluation of military aircraft. As such, it is the designer's goal to maximize system effectiveness. Industry documents indicate that all future military aircraft will incorporate signature reduction as an attempt to improve system effectiveness. Today's operating environments demand low observable aircraft which are able to reliably eliminate valuable, time critical targets. Thus, it is desirable to be able to evaluate the signatures of a vehicle, as well as the influence of signatures on the systems effectiveness of a vehicle. Previous studies have shown that shaping of the vehicle is one of the most important contributors to radar cross section and must be considered from the very beginning of the design process. This research strives to meet these needs by developing a parametric geometry radar cross section prediction tool.

A design process is formulated and implemented for the taxonomy selection and system-level optimization of an Efficient Multi-Mach Aircraft Current Technology Concept and an Advanced Concept. Concept space exploration of taxonomy alternatives is performed with multi-objective genetic algorithms and a Powell’s method scheme for vehicle optimization in a multidisciplinary modeling and simulation environment. A dynamic sensitivity visualization analysis tool is generated for the Advanced Concept with response surface equations.

The enabling of advanced design methods in an internet-capable framework will be discussed in this paper. The resulting framework represents the next generation of design and analysis capability in which engineering decision- making can be done by geographically distributed team members. A new internet technology called the lean-server approach is introduced as a mechanism for granting Web browser access to frameworks and domain analyses. This approach has the underpinnings required to support these next generation frameworks - collaboratories. A historical perspective of design frameworks is discussed to provide an understanding of the design functionality that is expected from framework implementations to insure design technology advancement. Two research areas were identified as being important to the development of collaboratories: design portals and collaborative methods.

The design of complex systems, such as commercial aircraft, has drastically changed since the middle 1970's. Budgetary and airline requirements have forced many aerospace companies to reduce the amount of time and monetary investments in future revolutionary concepts and design methods. The current NASA administration has noticed this shift in aviation focus and responded with the “Three Pillars for Success” program. This program is a roadmap for the development of research, innovative ideas, and technology implementation goals for the next 20 years. As a response to this program, the Aerospace Systems Design Laboratory at Georgia Tech is developing methods whereby forecasting techniques will aid in the proper assessment of future vehicle concepts. This method is called Technology Impact Forecasting (TIF). This method is applied to a medium-range, intra-continental, commercial transport concept.

This paper critically evaluates the use of Neural Networks (NNs) as metamodels for design applications. The specifics of implementing a NN approach are researched and discussed, including the type and architecture appropriate for design-related tasks, the processes of collecting training and validation data, and training the network, resulting in a sound process, which is described. This approach is then contrasted to the Response Surface Methodology (RSM). As illustrative problems, two equations to be approximated and a real-world problem from a Stability and Controls scenario, where it is desirable to predict the static longitudinal stability for a High Speed Civil Transport (HSCT) at takeoff, are presented. This research examines Response Surface Equations (RSEs) as Taylor series approximations, and explains their high performance as a proven approach to approximate functions that are known to be quadratic or near quadratic in nature.