According to the theory of variable heat, the turbofan engine component level model was established and the model effective was verified. Because of the poor real-time of nonlinear model, the airborne composite engine model which can meet the requirements of precision and real-time was obtained by using block modeling method. The target function and the constraint conditions of multi variable turbofan engine performance seeking control were analyzed based on the theory of nonlinear programming. The sequence of two quadratic programming algorithm(SQP) was used to solve the problem of performance optimization. The performance of aircraft and power unit increased significantly. In the minimum fuel consumption mode under engine constant thrust control, thrust fluctuation was very small (ΔF < 0.3%), engine fuel consumption rate decreased by 1.2% and the turbine temperature reduced by 2%, which achieved double optimization effect. In the maximum thrust mode, engine thrust increased by 0.5% and the aircraft acceleration time declined by 12.5%, which greatly improved the quality of flight.