An Improved Genetic Algorithm for the P-wave AVO Inversion

An improved genetic algorithm (IGA) is presented for geophysical P-wave amplitude variation with offset (AVO) inversion. The optimization method includes two phases: a new real coded GA with hybrid selection operator and hybrid crossover operator serves as a base level search. It makes a quick decision to direct the search towards the optimal region. In phase-2, the diversity of the population is maintained by creating new population and the particular parent with the best fitness is preserved in the next generation. The elitism method increases the probability of finding the global optimum. The experimental results show that the IGA not only improves the solution accuracy but also increases the convergence speed. Applying the IGA to P-wave AVO inversion, four elastic parameter contrasts are extracted. The proposed algorithm presents good stability in noisy condition.