New Approaches to Multidisciplinary Synthesis: An Aero-Structures-Control Application Using Statistical Techniques

An evolving aircraft synthesis simulation environment which offers improvements to existing methods at multiple levels of a design process is described in this paper. As design databases become obsolete due to the introduction of new technologies and classes of vehicles and as sophisticated analysis codes are often too computationally expensive for iterative applications, the design engineer may find a lack of usable information needed for decision making. Within the environment developed in this paper, rapid sensitivity analysis is possible through a unique representation of the relationship between fundamental design variables and system objectives. The combined use of the Design of Experiments and Response Surface techniques provides the ability to form this design relationship among system variables and target values, which is termed design-oriented in nature. The use of design-oriented analysis in a design space search with potentially conflicting objectives is addressed by the use of a Decision Support Problem Technique, which casts the search in terms finding ranged sets of acceptable designs as opposed to a traditional point design solution. Finally, a computing framework which facilitates the integration of the analysis with the design space search completes the simulation environment. The resulting product of this research is a case study demonstration: a supersonic transport, flexible wing design problem involving the aerodynamics, structures, and controls disciplines. This particular problem is particularly challenging because of the complex interactions involved in its solution and methods for overcoming these are highlighted in this paper.