 |
|
|||
Research
Despite the challenges, my students and I were quite successful in creating a quantitative theory of the measured spectrum from laser anemometer experiments. This theory was put to work in designing and making several innovative laser anemometer systems. We then went on to analyze and perfect other methods and uses for laser light scattering such as devices for measuring surface tension and zeta potential. In the course of the work discovered above, we did a lot of work with statistical data processing and techniques for using the computer in measurement systems. Early in the process we started to look for a method to rank order or evaluate various proposed schemes for making light scattering measurements. With a lot of nudging from my colleague in Denmark, Lars Lading, we settled on the expected measurement error (or reproducibility) as a measure of the quality of a measurement. The smaller the expected measurement error, the better. The main method used to calculate expected measurement errors is the Maximum Likelihood Parameter Estimation Process. (MLE). MLE can be used to process data or to make a priori estimates of the expected measurement error. My latest work has focussed on applying MLE methods to estimating measurement errors and optimization procedures for measurement systems, primarily light scattering systems. However, the techniques can be applied to other types of systems.
Selected PublicationsErrata Sheet for "Processing Random Data" by RV Edwards Mark P. Wernet and Robert V. Edwards, "Real Time Optical Correlator Using a Magnetooptic Device Applied to Particle Imaging Velocimetry," Applied Optics, Vol. 27, No. 5, March 1, 1988, p. 813-815. Edwards, R. V., Kolodzy, P. J., "Computation of the Autocorrelation Function of Velocity Fluctuations Using a Laser Anemometer in Sparsely Seeded Flows," Laser Anemometry in Fluid Mechanics III, R. J. Adrian, T. Asanuma, D.E.G. Durão, F. Durst, J. H. Whitelaw, LADOAN, (1988), Lisbon. Lading, L., Mann, J. A., Edwards, R. V., "Analysis of a Surface -- Scattering Spectrometer," J. Opt. Soc. Am. A, 11, No. 11, November 1989. L. Lading, R.V. Edwards, "Laser Velocimeters: Lower Limits to Uncertainty", Applied Optics, 32, 21, 1993. R.V. Edwards, “Processing of Random Data”, Chapter in Optical Diagnostics for Flow Processes, eds. Lars. Lading, Graham Wigley, Preben Buchhave, Plenum Press, New York and London (1995). Patrick Howard, Robert V. Edwards, “Maximum Likelihood Based Autocorrelation Processor for a Photon Resolved Laser Doppler Anemometer,” Measurement Science and Technology 7, (1996), p 801-822. Lading, L., Saffman, M., Edwards, R.V., "Laser Anemometry Based on Collective Scattering: the Effects of Propagating and Nonpropagating Fluctuations," Optics and Lasers in Engineering, 27, (1997), p 531-542. Howard, P., Edwards, R.V., “Implementation of a likelihood ratio test for laser doppler velocimeter burst detection,” Applied Optics, 36, 30, October 20, 1997, p. 7629-7638. Smart, Anthony E., Edwards, Robert V., Meyer, William V., “Quantitative simulation of errors in correlation analysis,” J. Applied Optics, 40, Nr. 24, 20 August, 2001, p 4064-4078 Edwards, R. V., Processing Random Data: Statistics for Engineers and Scientists, World Scientific Press, 2006. |
||||
I started out simply trying to adapt the newly invented laser anemometer to measure flows in an artificial kidney. I like to think that my initial failure to get one working was due to the fact that I was misled about how it worked. Figuring out a quantitative theory for how it really worked was quite daunting since it involved randomly placed particles that were convected along in the fluid flow and were undergoing Brownian motion at same time. Further, we were using laser light to measure the motions, so we had to take into account the optical parameters of the system and the special statistics for photon detection of scattered laser light. I didn't think this was the kind of problem that us chemical engineers were trained to solve.