Interesting Projects

Standoff Seismology

As part of an Army-sponsored project led by John Scales (a professor in the Physics Department of the Colorado School of Mines), some colleagues and I have been working on techniques for exciting and recording low-frequency (20-500 Hz) seismic energy as part of an effort to find safe ways to detect land-mines and other buried hazards.  So far we have successfully built and tested

  1. a non-linear parametric acoustic array for producing a well-collimated beam of low-frequency sound from a source element that’s smaller than a wavelength, and
  2. a millimeter-wave Doppler radar system to record ground motion at distances of a few tens of meters.

Inverse Theory for Fractures

Steve Brown and I extended an existing theory due to S. Oda for the effects of fractures on certain geophysical field measurements (elastic velocities, resistivity, permeability) and derived the appropriate inverse theory to invert field measurements into fracture distributions.  We were able to make substantial use of both a priori information as well as resolving kernels.

Core-scale acoustic tomography

At NER we designed and built a system for high-resolution acoustic tomography on core samples (at laboratory pressure), including the sample holder, acquisition hardware, and all the necessary supporting software.  The system included servo-positioned ultrasonic sensors, chirp-based ultrasonics, software to semi-automatically pick large numbers (tens of thousands) of arrival times with very good accuracy, and inversion and display software.  We were able to reuse much of the analysis and display software in a high-pressure system with acoustic-emission capability.

Lab Automation

I have had substantial experience in the development of lab-automation software and, to a lesser extent, hardware.  Some of the products and projects I’ve been involved with are

  • AutoLab  Gui-driven software that controls NER’s high-pressure core measurement systems and is used extensively at NER and at customer locations around the world.
  • ODM  The outboard data module is a linux-based embedded system that manages all of the analog and digital i/o between AutoLab hardware and the control software of the same name and communicates with the latter via ethernet.  The system is built around our own device driver and is designed to perform synchronous i/o at rates of up to a kilohertz on 32 channels, without interruption, for months at a time.
  • Acoustic Separator    This device is an ultrasonic system for measuring the relative volumes of brine and oil in a multi-component fluid streaming through the apparatus.  At its heart is an embedded linux system running software I wrote that uses very efficient signal-processing techniques. It also provides a useful graphics display.

Global Optimization Methods in Geophysics

At the instigation of John Scales, he, Terri Fischer, and I (while at Amoco research) undertook an extensive comparison of two methods for global optimization – that is, computational schemes to find the minima of a function which avoid, or try to avoid, getting caught in local dips in a way that doesn’t depend too heavily on a priori knowledge of the particular function being minimized.  We examined both simulated annealing and genetic algorithms.

Course in Geophysical Inverse Theory

In the winter semester of 1990 I taught a graduate course, as a visiting professor with support from both Amoco Research and Princeton University, in geophysical inverse theory.  The course was principally centered on Tarantola’s excellent, if challenging, book together with some visiting lecturers I was able to  induce.

Full-waveform Sonic Logging

A number of my Amoco Research colleagues and I developed a complete field data acquisition system for full-waveform sonic logging.  We also implemented an extensive suite of processing software centered on a (then) new and very effective processing algorithm.

 

Comments are closed.