R&D Projects

  Fiber Optics Research Program


Distributed fiber optic sensing technology enables continuous, real-time measurements along the entire length of a fiber optic cable deployed down a well.  This growing technology provides a reliable, cost-effective solution for many borehole geophysics applications since all data acquisition hardware is located at the surface.  In distributed acoustic sensing (DAS), time-varying strain along the fiber is recorded, and applications include:

  • Vertical seismic profiling
  • Microseismic monitoring
  • Completions evaluation
  • Production flow profiling
  • Fracture characterization

The same fiber optic cable can also be used for distributed temperature sensing (DTS).

The Fiber Optics Research Program (FORP) is working to improve the acquisition, processing, and modeling of DAS and DTS data.  We are pursuing laboratory tests by deploying fiber optic cable in a flow loop on the CSM campus and a horizontal borehole in the Edgar Research Mine.  We are also developing software to interpret the copious amounts of data produced by DAS acquisition.  Further activities support the analysis of DAS data from the Eagle Ford, Wolfcamp and Chalk Bluff projects.  This is a collaborative project with the Mines Petroleum Engineering Department, Prof. Jennifer Miskimins and Prof. Hossein Kazemi.


 Microseismic Guided Waves Analysis


The advent of distributed acoustic sensing (DAS) technique enables seismic monitoring in restrictive downhole conditions and provides unique opportunities for geophysicists to observe microseismic phenomena in unconventional reservoirs during hydraulic fracturing operations. The DAS fiber instrumented along the horizontal well can effectively detect the microseismic-induced guided waves that travel within the unconventional reservoir. The strongly dispersive guided waves trapped within the shale formation bear critical information about the reservoir structure and microseismic source parameters, including:

  • Isotropic/anisotropic seismic velocity profile
  • Layer thickness
  • Source radiation pattern
  • Source location

RCP is developing accurate physics-based models to characterize guided wave behaviors in downhole DAS records and to test our theoretical analysis against field data and synthetic guided waves generated by well-developed wave propagation modeling methods. Our goal is to establish generic analytic procedures to extract useful information from field observations of guided waves and to provide qualitative and quantitative constraints on the reservoir structure and the microseismic source mechanism. The knowledge gained through the guided wave analysis can shed light on field development by providing important engineering insights into the hydraulic fracturing operations, such as shale layer thickness calibration, time-lapse variation of seismic velocity, and stimulated rock volume interpretation.


 Machine Learning and Data Analytics


The growth of both data volume and data types in the development of a modern oil field is stretching the capabilities of traditional manual workflows. Moreover, the complex physics behind the interactions between dynamic reservoir properties and well completion processes in unconventional reservoirs is only partially understood through a deterministic, model-based analysis. The oil and gas industry is turning to new analysis tools to overcome these challenges. These tools are more data-driven and take into account the stochastic nature of unconventional resource plays.

These tools are known by various names like data analytics, data science and machine learning. These methods allow us to make sense of lots of data with many variables utilizing the power of computer systems that automatically improve with experience. The ultimate aim of this technology is to enhance the productivity of geologists, geophysicists and petroleum engineers by automating tasks and performing the majority of time-consuming analysis.

RCP is working with the Chalk Bluff, Eagle Ford and Wattenburg project data to generate new statistical analysis workflows. We are also working on the application of new machine learning technologies like Deep Learning to solve a wide range of exploration and development problems. The near future focus is on four specific problems:

  • Machine learning based geostatistical estimation and simulation
  • Automation of microseismic processing workflows through Deep Learning
  • Land seismic noise reduction via Dictionary Learning
  • Sweet spot identification using pressure and flow data in conjunction with seismic and well data

 Compressive Sensing


Compressive Sensing (CS) is a new sampling paradigm. While sampling an analog signal, one needs to be sure that a discrete version of the signal preserves all signal information. According to the Nyquist rate, if the sampling rate is at least twice the maximum signals’ frequency, then we say that the signal is recorded without loss of information. CS gives an alternative view of the sampling rate and exploits signal sparsity.

In terms of seismic application, CS places receivers and sources irregularly, and the distance between some adjacent sources/receivers might be below the Nyquist rate. If the sampling grid, sparse transform, and recovery algorithm are selected wisely, then the CS-acquired data plus recovered data are almost identical to data recorded on the dense uniform grid.

CS makes seismic data acquisition cheaper and faster. CS has been successfully tested in the field and has been adopted by companies to shoot land and marine surveys. However, CS application to seismic problems still faces challenges due to the non-sparse feature of seismic data under sparse transforms, wide dynamic range and noise contamination of data, and large-scale data volumes.

This project is addressing these challenges towards better implementation and use of CS concepts for seismic data acquisition. We investigate different sparse reconstruction techniques, i.e., 5D interpolation. Also, we study machine learning-based noise attenuation, acquisition footprint removal, and regularization for improving land data quality. Our recent findings show that extensively used mutual coherence does not correctly differentiate different irregular sampling patterns. Currently, we are searching for a more reliable metric to guide the CS survey design.

To make our research relevant and applicable to a complex land seismic data environment, we are working with the SEAM Barrett dataset and 3D field datasets from sponsoring companies. This project is a collaborative project with the CSM Electrical Engineering Department (Prof. Michael B. Wakin).

 Enhanced Oil Recovery for       Unconventionals


EOR has the potential to improve the cumulative oil production from liquid-rich unconventional shale reservoirs that is currently only around 6%. RCP, in a collaborative project with the Mines Petroleum Engineering Department and the Lawrence Berkeley National Laboratory, are conducting experiments on core samples to calibrate EOR processes in unconventionals. Several field tests and numerical modeling (e.g., RCP Wattenberg project) have indicated that gas recycling is a viable EOR method to produce more oil from liquid-rich shale reservoirs. Our goal is to understand and quantify the fundamentals of gas and CO2 injection EOR in shale reservoirs.

Contact us for more detail on these new and exciting R&D programs.