Research Projects

Seismic inversion is a critical technique used in geophysics to transform seismic data into subsurface models that represent geological structures. These models play a crucial role in resource exploration, particularly in the oil, gas, and mining industries. Traditional seismic inversion techniques, however, face significant challenges. They require more computational power, are time-consuming, and often produce models with limited resolution or accuracy, especially in complex geological settings. Furthermore, seismic data can be noisy and affected by uncertainty, making accurate interpretation a difficult task. This project proposes to leverage quantum computing and quantum-enhanced artificial intelligence (AI) to address these challenges, offering good improvements in seismic inversion and interpretation processes. Quantum computing holds the potential to transform seismic inversion by performing complex calculations much faster than traditional high-performance computing (HPC) systems. Quantum algorithms can perform operations in parallel using quantum bits, allowing for the acceleration of the inversion process and enabling real-time processing of large seismic datasets. This capability will significantly reduce the computational bottlenecks faced by traditional methods, making seismic inversion faster and more efficient. One of the primary goals of this project is to develop quantum-enhanced seismic inversion algorithms that can handle large datasets while producing high-resolution and accurate subsurface models. Traditional inversion methods often struggle to capture fine details in complex geological formations, leading to models that lack precision. By enhancing image quality and resolution, these quantum algorithms will enable geophysicists to better understand complex geological structures, improving the accuracy of resource assessments and reducing exploration risks. Another significant challenge in seismic interpretation is the noise and uncertainty inherent in seismic data. Seismic measurements are often impacted by environmental factors, equipment limitations, and subsurface heterogeneities, which introduce noise into the dataset and make interpretation more difficult. Traditional algorithms may fail to effectively handle this noise, leading to unreliable models. This project will explore the application of quantum-enhanced machine learning models that can better manage uncertainty and reduce noise in seismic data. To achieve these objectives, we will develop a comprehensive quantum computing framework for seismic inversion and interpretation. We validate the developed algorithms with real-world seismic datasets by comparing them with traditional methods to evaluate improvements in processing speed, model accuracy, and interpretive reliability.