Développement d’un modèle de performance aéro-propulsive en vol de croisière – Application sur un avion Cessna Citation X

Abstract

The study of the performance of an aircraft is a critical aspect in the world of aviation. Whether it is for a safety, ecological or economic concern, there are many advantages to predicting the flight performance of an aircraft. With the constant increase in the number of flights around the world, fuel consumption is becoming a major issue. However, it is estimated that approximately every 5 years, an aircraft increases its fuel consumption by approximately 2% (Airbus, 2001). Whether it’s due to the degradation of the aerodynamic structure or the engines, it becomes increasingly difficult over time to predict the amount of fuel required for a given flight. This thesis deals with the development of a model predicting the aero-propulsive performance of a Cessna Citation X aircraft in cruise flight. This performance model is represented by adaptive look-up tables, one for aerodynamic performance, one for propulsion performance and a third, serving to validate results, for aero-propulsion performance, i.e. using the results of the aerodynamic model to make the prediction of the propulsive performance. The model is divided into three major stages. The first one is to establish the initial performance of the aircraft in cruise flight, i.e. the starting point of the model. The second consists in locally adapting the lookup tables according to flight data coming from the flight simulator. The third consists in making a global adaptation of the look-up tables and thus obtaining the final aeropropulsive performance model. The results of this master's thesis demonstrate that by using the proposed aero-propulsive performance model, it is possible to predict aerodynamic performance with an average accuracy of up to 0.99%, propulsive performance at 3.38% and performance combined (propulsive model using the results of the aerodynamic model) up to 6.25%. These results were obtained using the neural network estimation method.

Date	12 Nov 2022
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Ruxandra Botez (Supervisor)