Study of detecting impact damage for composite material based on intelligent sensor

Abstract

A system of impact damage detection for composite material structures by using an intelligent sensor embedded in composite material is described. In the course of signal processing, wavelet transform has the exceptional property of temporal frequency localization, whereas Kohonen artificial neural networks have excellent characteristics of self-learning and fault-tolerance. By combining the merits of abstracting time-frequency domain eigenvalues and improving the ratio of signal to noise in this system, impact damage in composite material can be properly recognized.

Keywords

wavelet transform / neural network / intelligent sensor / composite material / impact damage detection

References

[1] Miller G E. Fiber Optic Sensors for Aircraft. SPIE Fiber Optic Smart Structures and Skins IV, 1991, 1538: 178-181.

[2] Lee C K. Theory of Laminated Piezoelectric Plates for the Design of Distributed Sensors/Actuators. Part I: Governing Equations and Reciprocal Relationships. J. Acoust Soc. Am., 1990, 87(3): 1144-1158.

[3] Salzano T B, Calder C A. Vibration Measurement in Composite Members Using Embedded Constantan Wire. J. of Intelligent Materials and Structures, 1992, 3(3): 536-546.

[4] Brignell J, White N. Intelligent Sensor Systems, 1994, Bristol: IOP Publishing Inc.

[5] Xue Q, Reddy RBS. Late Potential Recognition by Artificial Neural Networks. IEEE Trans, on Biomed. Eng., 1997, 44: 132-143.

[6] Wang Lei, Klein Meir, Yirga Solomon, Kundur Prabha. Dynamic Reduction of Large Power Systems for Stability Studies. IEE transactions on PS, 1997, 12(2): 2103-2108.

[7] Qianqing Qin, Zhongkai Yang. Practicality Wavelet Analysis, 1994, Xi-An: XiAn University Press of Electron Technology.

[8] Wei Liu, . The Compression of IR Spectra by Using Wavelet Neural Network. Chinese Science Bulletin, 1997, 10: 822-825.

[9] Baiquan Lu, . Wavelet Neural Network for Function Leaming. Automatic Sinice, 1998, 24(4): 548-551.

[10] Yong Yu, Dejun Wan. Simulation Research on an Approach of Fault Detection Using Wavelet Neural Networks. Computer Simulation, 2000, 17(1): 24-27.