Phase Statistics and Spectral Analysis of Ultrasonic Signals for CFRP Component Assessment

Abstract : The established and industrially applied ultrasonic inspection and monitoring techniques typically use the amplitude values of the ultrasonic signals for assessment of flaws. However every individual signal value has the phase information, in addition to the amplitude information. Through analysis of the statistical distribution of phase values provided by state-of-the-art measurement systems and an appropriate consideration of this information in the signal evaluation, a significant improvement in flaw detection capability of an ultrasonic test can be achieved. Moreover, the phase and the frequency of ultrasonic signals carries information that can be correlated to the inner structure and existing inhomogenities of the material to be assessed. The two novel approaches for increasing the information content of the classic ultrasonic test - phase statistics and spectral analysis - are new and promising. Use of these approaches, even for on-line measurements has become possible particularly through modern computer science and signal processing. It is particularly significant that these methods are based exclusively on the analysis and processing of "conventionally" acquired ultrasonic signals, so that in-depth rework of the measurement technology or ultrasonic hardware can be added with additional useful information on the material's condition to be obtained.