GNGTS 2017 - 36° Convegno Nazionale

722 GNGTS 2017 S essione 3.3 energy of the stack trace associated to the DAGA CMP gather is very close to the energy of the stack trace pertaining to the reference CMP. Note that differently from the NGA CMP, the final DAGA CMP could be used as a valid starting model for any local optimization method for a further refinement of the residual statics estimation. Due to the limited number of model evaluations the DAGA resulted to be only 1.11 times slower than NGA. Conclusions. We presented an innovative strategy to attenuate the genetic drift and to increase the exploration of the model space in a GA optimization. In the proposed DAGA approach, a standard NGA is hybridized with principles coming fromMCA. The test on analytic objective functions and on residual statics computation, demonstrated that, differently from the NGA, the implemented DAGA approach does not suffer the genetic drift effect. In particular, our tests confirmed that the DAGA approach grants the convergence in case of objective functions with very complex topology, where the standard NGA and MCA fail to converge. Differently, in cases of simpler topologies the NGA and the DAGA algorithms achieved very similar performances. The implemented DAGAmethod is more computational demanding than the standard NGA and for this reason an accurate code optimization has been performed. After additional code optimization (i.e. parallel implementation), the next step of our research is to apply the DAGA method to 2D acoustic full-waveform inversion that is a highly non-linear geophysical optimization problem with expensive forward modelling. References Aleardi M., and Mazzotti, A. (2017). 1D elastic full-waveform inversion and uncertainty estimation by means of a hybrid genetic algorithm–Gibbs sampler approach . Geophysical Prospecting, 65(1), 64-85. Eldos T. (2008). Distributed genetic algorithms: a scheme for genetic drift avoidance , Journal of electrical engineering, 59(1), 45. Holland J. (1975). Adaptation in Natural and Artificial Systems , The MIT Press, Cambridge. Rothman D.H. (1985). Nonlinear inversion, statistical mechanics, and residual statics estimation . ����������� Geophysics, 50(12), 2784-2796. Sajeva A., Aleardi M., Stucchi E., Bienati N., Mazzotti A. (2016). Estimation of acoustic macro models using a genetic full-waveform inversion: Application to the Marmousi model. ����������� ������ ���������� Geophysics, 81(4), R173-R184. Sajeva A., Aleardi M., Galuzzi B., Stucchi E., Spadavecchia E., Mazzotti A. (2017). Comparing the performances of four stochastic optimisation methods using analytic objective functions, 1D elastic full-waveform inversion, and residual static computation . Geophysical Prospecting. In print. doi:10.1111/1365-2478.12532. Contribution of Stern Layer and Membrane Polarization to the Spectral Induced Polarization of Porous Media M. Rossi 1 , A. Brovelli 2 , G. Cassiani 3 , S. Johansson 1 , T. Dahlin 1 1 Lund University, LTH 2 Isamgeo italia srl, Italy 3 University of Padova, Italy Introduction. There is an increasing interest in understanding the Spectral Induced Polarization (SIP) phenomenon in porous media. The complex electrical conductivity of porous geological materials has a frequency dependent behavior than can be associated to three main mechanisms: 1. �������������� ���� ������������� ������������ �� ��� ����������� ����� ��������� ��� Maxwell-Wagner (MW) polarization. ������������ ��� ����������� ����� ��������� ��� Polarization of the water-solid grain interface due to the accumulation of charges. It occurs at frequencies higher than 10-100 Hz. There are a number of mechanistic models that describe this kind of polarization (���� ���� ��� ��� e.g. Chen and Or, 2006)�; 2. �������� ������������� ��� ����������� �� ��� ���� ��������� ��� ��� ���������� ������ membrane polarization. The combination of the pore structure and the Electrical Double Layer (EDL) on the surface of the grains are responsible for the local accumulation of