An Enhanced Genetic Algorithm and its Application on Nonlinear Geophysical Problems
Y.P. Villa Acuna, Y. Sun and R. Van Borselen
Event name: 81st EAGE Conference and Exhibition 2019
Session: Poster: Near-Surface Methods and Applications
Publication date: 03 June 2019
Info: Extended abstract, PDF ( 1.76Mb )
Price: € 20
Previously we proposed an advanced Genetic Algorithm and successfully applied it to large-scale static correction problems. In this paper we introduce an enhanced Genetic Algorithm which further improves both the global search and the local search capabilities of our previous algorithm, dramatically boosting the convergence speed. Our enhanced Genetic Algorithm adds two modifications to our previous algorithm: the “island model” gives a big uplift on the global search capability, and the “self-adaptive differential evolution fine tuning scheme” greatly benefits the local search capability. We first demonstrate the improved convergence speed of the enhanced Genetic Algorithm over our advanced Genetic Algorithm using three multi-modal test functions, and next successfully apply our enhanced Genetic Algorithm to two nonlinear geophysical problems, namely receiver-side static corrections and common reflection surface stacking.