In a groundbreaking development poised to revolutionize subsurface imaging, researchers have harnessed the power of artificial intelligence to enhance the clarity of ground penetrating radar (GPR) data. This innovation, detailed in a recent study published in the *NDT – Journal of Non-Destructive Testing* (translated to English as *Journal of Non-Destructive Testing*), promises to significantly improve the accuracy and efficiency of subsurface mapping, with profound implications for the energy sector.
At the heart of this research is Nikos Economou, a leading expert from the Applied Geophysics Lab at the Technical University of Crete. Economou and his team have developed an AI-aided method that focuses diffracted energy within GPR sections, a critical advancement for imaging subsurface structures. “Traditional methods often struggle with the randomness of diffractions in GPR data,” Economou explains. “Our approach uses AI to detect these diffractions and generate a velocity model that significantly enhances the clarity of the subsurface image.”
The conventional method for focusing diffracted energy in GPR data involves migration, a process that relies on a migration velocity model. However, obtaining an accurate velocity model from zero-offset data can be challenging due to the random distribution of diffractions. Economou’s team introduces a multipath summation method, which involves weighted stacking of constant velocity migrated sections. This method is particularly effective in low-heterogeneity environments but requires 2D weights when electromagnetic velocity varies dramatically.
The AI algorithm developed by Economou’s team detects diffractions and uses their kinematic information to generate a diffraction velocity model. This model is then used to assign 2D weights for the weighted multipath summation, focusing the scattered energy within the GPR section. “The AI algorithm not only improves the accuracy of the velocity model but also enhances the lateral continuity of reflections,” Economou notes. “This is a game-changer for subsurface imaging.”
The implications of this research are vast, particularly for the energy sector. Accurate subsurface imaging is crucial for exploring and extracting energy resources, as well as for assessing the integrity of existing infrastructure. “By improving the clarity and accuracy of GPR data, our method can help energy companies make more informed decisions,” Economou says. “This can lead to more efficient exploration and extraction processes, ultimately reducing costs and environmental impact.”
The study demonstrates the application of this methodology using simulated data and real-world GPR data from marble assessment. The results show a significant improvement in the lateral continuity of reflections, highlighting the potential of this AI-aided approach. “This research opens up new possibilities for subsurface imaging,” Economou concludes. “It’s an exciting time for the field, and we’re just scratching the surface of what AI can do.”
As the energy sector continues to evolve, the need for accurate and efficient subsurface imaging will only grow. Economou’s research provides a promising solution, one that could shape the future of energy exploration and extraction. With the aid of AI, the subsurface is becoming clearer, and the possibilities are endless.