TY - GEN
T1 - Research on Optimal Sensor Placement for Aircraft Structural Health Management
AU - Wan, Fangyi
AU - Yu, Xingliang
AU - Yu, Qi
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/31
Y1 - 2017/7/31
N2 - Prognostic and Health Management (PHM) systems can manage health conditions of airplane real-time. As crucial subsystem of PHM, Aircraft Structural Health Management (ASHM) mainly focus on structural health monitoring and assessment, microstructure fault monitoring and isolation, overload, corrosion monitoring, residual life assessment and so on. It is overwhelmingly difficult for large aircraft structure to optimize sensor placement in ASHM on account of diversity of failure modes, hiddenness of fault position, incalculability of corrosion failure. Especially, corrosion detection, this is an intractable problem because corrosion degree is gradually accumulated with time. However, the Optimal Sensor Placement (OSP) is a vital aspect and plays a crucial role in data acquisition and data identification for the ASHM. Thus, thousands of sensors are installed for aircraft structure. Besides, failure models such as accidental damages, fatigue damages and corrosion, are often synergistic and interactive. Therefore, the OSP must be enough underlined for ASHM. In this paper, the OSP of aircraft wing structure had been discussed. Firstly, the test matrix was gained through modal test and the finite element model of wing structure was built by ANSYS. Secondly, sensor measuring points were optimized by the Singular Value Decomposition (SVD), the QR decomposition and the Fuzzy Measurement Coverage. Adopting singular value difference method and system fixed order method based on singular entropy had coped with sensor placement. The final result of the OSP had been verified by the QR decomposition and the Fuzzy Measurement Coverage. Finally, the scheme of OSP had been analyzed for aircraft wing structure. For the demand of the safety and reliability, at least 92 measuring points should be monitored for wing structure. Through analysis, about 18 measuring points could meet the demand for a wing. The final scenario of the OSP is that the strain of 10 measuring points and longitudinal and horizontal strain of eight wing-body joints are monitored. Numerous measuring points have been simplified and the effective information and data we need can be acquired for the wing.
AB - Prognostic and Health Management (PHM) systems can manage health conditions of airplane real-time. As crucial subsystem of PHM, Aircraft Structural Health Management (ASHM) mainly focus on structural health monitoring and assessment, microstructure fault monitoring and isolation, overload, corrosion monitoring, residual life assessment and so on. It is overwhelmingly difficult for large aircraft structure to optimize sensor placement in ASHM on account of diversity of failure modes, hiddenness of fault position, incalculability of corrosion failure. Especially, corrosion detection, this is an intractable problem because corrosion degree is gradually accumulated with time. However, the Optimal Sensor Placement (OSP) is a vital aspect and plays a crucial role in data acquisition and data identification for the ASHM. Thus, thousands of sensors are installed for aircraft structure. Besides, failure models such as accidental damages, fatigue damages and corrosion, are often synergistic and interactive. Therefore, the OSP must be enough underlined for ASHM. In this paper, the OSP of aircraft wing structure had been discussed. Firstly, the test matrix was gained through modal test and the finite element model of wing structure was built by ANSYS. Secondly, sensor measuring points were optimized by the Singular Value Decomposition (SVD), the QR decomposition and the Fuzzy Measurement Coverage. Adopting singular value difference method and system fixed order method based on singular entropy had coped with sensor placement. The final result of the OSP had been verified by the QR decomposition and the Fuzzy Measurement Coverage. Finally, the scheme of OSP had been analyzed for aircraft wing structure. For the demand of the safety and reliability, at least 92 measuring points should be monitored for wing structure. Through analysis, about 18 measuring points could meet the demand for a wing. The final scenario of the OSP is that the strain of 10 measuring points and longitudinal and horizontal strain of eight wing-body joints are monitored. Numerous measuring points have been simplified and the effective information and data we need can be acquired for the wing.
KW - Aircraft wing structures
KW - Fuzzy Measurement Coverage
KW - Optimal Sensor Placement
KW - QR decomposition
KW - Singular Value Decomposition
UR - http://www.scopus.com/inward/record.url?scp=85028554693&partnerID=8YFLogxK
U2 - 10.1109/ICPHM.2017.7998322
DO - 10.1109/ICPHM.2017.7998322
M3 - 会议稿件
AN - SCOPUS:85028554693
T3 - 2017 IEEE International Conference on Prognostics and Health Management, ICPHM 2017
SP - 160
EP - 166
BT - 2017 IEEE International Conference on Prognostics and Health Management, ICPHM 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 IEEE International Conference on Prognostics and Health Management, ICPHM 2017
Y2 - 19 June 2017 through 21 June 2017
ER -