In a modern Full-Authority Fly-By-Wire Flight Control System (FBW/FCS), the air data (static pressure, Mach number, and angles of attack and sideslip) are critical items for the safety of flight operations. As a matter of fact, air data are used for gain scheduling in the control laws and for the envelope protection. Such data have to be obtained by specific computation algorithms on the basis of local airflow measurements performed by redundant air data probes. The algorithms also have to manage the redundancy in order to detect possible failures and to provide consolidated outputs. This paper describes two different approaches to the development of air data computation algorithms. The first one, widely illustrated in [1], uses polynomial calibration functions tuned on wind tunnel test data relevant to the new jet trainer Aermacchi M-346. The second approach is based on neural networks trained in two ways: using the same wind tunnel data and using preliminary flight test data. The paper also illustrates the monitoring and voting algorithms developed in order to identify possible probe failures and to provide a voted value for each air data parameter. Finally, the results of the different approaches are presented by comparisons with the wind tunnel data and preliminary flight test data.

Fault-Tolerant Procedures for Air Data Elaboration

DENTI, EUGENIO;GALATOLO, ROBERTO;SCHETTINI, FRANCESCO
2006-01-01

Abstract

In a modern Full-Authority Fly-By-Wire Flight Control System (FBW/FCS), the air data (static pressure, Mach number, and angles of attack and sideslip) are critical items for the safety of flight operations. As a matter of fact, air data are used for gain scheduling in the control laws and for the envelope protection. Such data have to be obtained by specific computation algorithms on the basis of local airflow measurements performed by redundant air data probes. The algorithms also have to manage the redundancy in order to detect possible failures and to provide consolidated outputs. This paper describes two different approaches to the development of air data computation algorithms. The first one, widely illustrated in [1], uses polynomial calibration functions tuned on wind tunnel test data relevant to the new jet trainer Aermacchi M-346. The second approach is based on neural networks trained in two ways: using the same wind tunnel data and using preliminary flight test data. The paper also illustrates the monitoring and voting algorithms developed in order to identify possible probe failures and to provide a voted value for each air data parameter. Finally, the results of the different approaches are presented by comparisons with the wind tunnel data and preliminary flight test data.
2006
0953399176
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: https://hdl.handle.net/11568/190350
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? ND
social impact