INVERSION OF COMPOSITE-MATERIAL ELASTIC-CONSTANTS FROM ULTRASONIC BULK WAVE PHASE-VELOCITY DATA USING GENETIC ALGORITHMS

In this paper, efforts on the reconstruction of material stiffness pro perties of unidirectional fiber-reinforced composites from obliquely i ncident ultrasonic bulk wave data, employing an inverse technique base d on genetic algorithms, is described. Computer-generated ultrasonic p hase velocity data, as a function of the angle of refraction in both s ymmetry and nonsymmetry through the thickness planes of unidirectional composites were used as the input to the genetic algorithm. A simple genetic algorithm, with optimal parameters chosen from the literature and numerical analysis (reported here), was implemented for the recons truction. The inversion using this novel technique was found to be ext remely promising in the characterization of the material stiffness pro perties. Stability to noise in the input phase velocity data set was a lso investigated. Advantages and disadvantages of the genetic reconstr uction technique over conventional methods are also discussed. (C) 199 8 Elsevier Science Limited. All rights reserved.