A computational methodology for an experimental design problem based on the concept of potential and repulsive fields is presented. The problem consists of finding the optimal distribution of test points in a two-dimensional domain, pursuant to hard constraints (permitted boundaries of the domain) and soft constraints (minimization of potential). Each test point is assumed to be the source of different fields which expose all other points to repulsive forces, thus accelerations, acting in different directions. The result of the mutual repulsive forces is a dynamic evolution of the configuration of test points in the domain, which eventually converges to a condition of minimum potential, where forces are balanced. An iterative process is adopted to find a numerical solution where residual accelerations are below a desired threshold. When a change of the number of test points (either an increment or reduction) is needed, the method has been extended to the additional task of dynamically relocating the remaining test points, after an initial subset has been performed. The proposed technique allows for an easy accomplishment of the task with minor modifications to the algorithm. The idea has been tested against a practical case: the definition of a flight test matrix for the evaluation of the aero-elastic and environmental characteristics of an aircraft. The goal is to distribute flight test points in the flight envelope in such a way to satisfy the requirements of structural engineers, interested in an optimal distribution in terms of airspeed and Mach, and systems engineers, more interested in the altitude and airspeed distribution. The method provides to combine all objectives in a single test campaign through an optimization of the test point distribution, in line with classical facility location problems. A large degree of flexibility in the algorithm is allowed to tune the relative weights assigned to the different requirements. The presented method shows effective and computationally efficiency, exhibiting satisfactory results in both the test matrix design task and the dynamic relocation task.

Design of a Flight Test Matrix and Dynamic Relocation

MALLOZZI, LINA;D'ARGENIO, ALESSANDRO;DE PAOLIS, PIERLUIGI
2013

Abstract

A computational methodology for an experimental design problem based on the concept of potential and repulsive fields is presented. The problem consists of finding the optimal distribution of test points in a two-dimensional domain, pursuant to hard constraints (permitted boundaries of the domain) and soft constraints (minimization of potential). Each test point is assumed to be the source of different fields which expose all other points to repulsive forces, thus accelerations, acting in different directions. The result of the mutual repulsive forces is a dynamic evolution of the configuration of test points in the domain, which eventually converges to a condition of minimum potential, where forces are balanced. An iterative process is adopted to find a numerical solution where residual accelerations are below a desired threshold. When a change of the number of test points (either an increment or reduction) is needed, the method has been extended to the additional task of dynamically relocating the remaining test points, after an initial subset has been performed. The proposed technique allows for an easy accomplishment of the task with minor modifications to the algorithm. The idea has been tested against a practical case: the definition of a flight test matrix for the evaluation of the aero-elastic and environmental characteristics of an aircraft. The goal is to distribute flight test points in the flight envelope in such a way to satisfy the requirements of structural engineers, interested in an optimal distribution in terms of airspeed and Mach, and systems engineers, more interested in the altitude and airspeed distribution. The method provides to combine all objectives in a single test campaign through an optimization of the test point distribution, in line with classical facility location problems. A large degree of flexibility in the algorithm is allowed to tune the relative weights assigned to the different requirements. The presented method shows effective and computationally efficiency, exhibiting satisfactory results in both the test matrix design task and the dynamic relocation task.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11588/598916
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact