Browsing by Issue Date, starting with "2022-02-02"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
- 4D commercial trajectory optimization for fuel saving and environmemtal impact reductionPublication . Ahmed, Kawser; Bousson, KouamanaThe main purpose of the thesis is to optimize commercial aircraft 4D trajectories to improve flight efficiency and reduce fuel consumption and environmental impact caused by airliners. The Trajectory Optimization Problem (TOP) technique can be used to accomplish this goal. The formulation of the aircraft TOP involves the mathematical model of the system (i.e., dynamics model, performance model, and emissions model of the aircraft), Performance Index (PI), and boundary and path constraints of the system. Typically, the TOP is solved by a wide range of numerical approaches. They can be classified into three basic classes of numerical methods: indirect methods, direct methods, and dynamic programming. In this thesis, several instances of problems were considered to optimize commercial aircraft trajectories. Firstly, the problem of optimal trajectory generation from predefined 4D waypoint networks was considered. A single source shortest path algorithm (Dijkstra’s algorithm) was applied to generate the optimal aircraft trajectories that minimize aircraft fuel burn and total trip time between the initial and final waypoint in the networks. Dijkstra’s Algorithm (DA) successfully found the path (trajectory) with the lowest cost (i.e., fuel consumption, and total trip time) from the predefined 4D waypoint networks. Next, the problem of generating minimum length optimal trajectory along a set of predefined 4D waypoints was considered. A cubic spline parameterization was used to solve the TOP. The state vector, its time derivative, and control vector are parameterized using Cubic Spline Interpolation (CSI). Consequently, the objective function and constraints are expressed as functions of the value of state and control at the temporal nodes, this representation transforms the TOP into a Nonlinear Programming (NLP) problem, which is then solved numerically using a well-established NLP solver. The proposed method generated a smooth 4D optimal trajectory with very accurate results. Following, the problem considers generating optimal trajectories between two 4D waypoints. Dynamic Programming (DP) a well-established numerical method was considered to solve this problem. The traditional DP bears some shortcomings that prevent its use in many practical real-time implementations. This thesis proposes a Modified Dynamic Programming (MDP) approach which reduces the computational effort and overcomes the drawbacks of the traditional DP. The proposed MDP approach was successfully implemented to generate optimal trajectories that minimize aircraft fuel consumption and emissions in several case studies, the obtained optimal trajectories are then compared with the corresponding reference commercial flight trajectory for the same route in order to quantify the potential benefit of reduction of aircraft fuel consumption and emissions. The numerical examples demonstrate that the MDP can successfully generate fuel and emissions optimal trajectory with little computational effort, which implies it can also be applied to online trajectory generation. Finally, the problem of predicting the fuel flow rate from actual flight data or manual data was considered. The Radial Basis Function (RBF) neural network was applied to predict the fuel flow rate in the climb, cruise, and descent phases of flight. In the RBF neural network, the true airspeed and flight altitude were taken as the input parameters and the fuel flow rate as the output parameter. The RBF neural network produced a highly accurate fuel flow rate model with a high value of coefficients of determination, together with the low relative approximation errors. Later on, the resulted fuel flow rate model was used to solve a 4D TOP by optimizing aircraft green cost between two 4D waypoints.