Optimization of Variable Cycle Engines by Using an Improved Differential Evolution

Abstract

The performance of variable cycle engine was optimized both in single and double bypass mode with an improved differential evolution algorithm. The variable cycle engine's major cycle parameters, the varying schedules of the geometry components and the fuel flow were used as optimization variables. The required thrust and compressor surge margins were used as constraints. The optimization goal was to minimize fuel consumption or maximize thrust performance. Two different base vectors and self-adaptive control parameters were applied to accelerate convergence speed. A uniformity mutation operation was used for poor fitness individuals and a Gaussian mutation operation for the current best fitness individual was added to enhance population diversity. It is shown that the developed differential evolution algorithms present better performance. The procedure provides variable cycle engine's design point optimization and off-design point optimization. Besides, an integrated optimization of both points optimizations was developed, presenting the most remarkable performance. The present work obtains the most favorable match on design parameters and the adjusting schedule of variable cycle engine.