Multidisciplinary Optimization in Conceptual Design of Mixed-Stream Turbofan Engines

Abstract

Conceptual engine design is currently done essentially using a trial-and-erro r process based on the experience of engineers. This paper presents an alternative approach using mult {disciplinary optimization to automate the conceptual design phase and ensure that an optimal design is reached. A robust integrated computer program was created to find the values of eight basic engine parameters that minimized fuel usage over a given mission. Because of the character of the objective function, a new approach was devised incorporating a genetic algorithm with a switch to a gradient-based algorithm to refine the design. The new procedure was used to optimize the engine design for a conceptual short-range interceptor. The process is general, however, and can be applied to a wide variety of missions and aircraft.