Application of metaheuristic and deterministic algorithms for aircraft reference trajectory optimization

Aircraft reference trajectory is an alternative method to reduce fuel consumption, thus the pollution released to the atmosphere. Fuel consumption reduction is of special importance for two reasons: first, because the aeronautical industry is responsible of 2% of the CO2 released to the atmosphere, and second, because it will reduce the flight cost. \n \nThe aircraft fuel model was obtained from a numerical performance database which was created and validated by our industrial partner from flight experimental test data. A new methodology using the numerical database was proposed in this thesis to compute the fuel burn for a given trajectory. \n \nWeather parameters such as wind and temperature were taken into account as they have an important effect in fuel burn. The open source model used to obtain the weather forecast was provided by Weather Canada. A combination of linear and bi-linear interpolations allowed finding the required weather data. \n \nThe search space was modelled using different graphs: one graph was used for mapping the different flight phases such as climb, cruise and descent, and another graph was used for mapping the physical space in which the aircraft would perform its flight. \n \nThe trajectory was optimized in its vertical reference trajectory using the Beam Search algorithm, and a combination of the Beam Search algorithm with a search space reduction technique. \n \nThe trajectory was optimized simultaneously for the vertical and lateral reference navigation plans while fulfilling a Required Time of Arrival constraint using three different metaheuristic algorithms: the artificial bee’s colony, and the ant colony optimization. \n \nResults were validated using the software FlightSIM®, a commercial Flight Management System, an exhaustive search algorithm, and as flown flights obtained from flightaware®. All algorithms were able to reduce the fuel burn, and the flight costs.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.