Developing a G2 Software Bridge. WALLACE SHEEHAN (University of Colorado Boulder, CO 80301) JOSEPH FARMER (National Renewable Energy Laboratory, Golden, CO, 89401)

The Process Development Unit (PDU) -- a bioethanol plant -- at the National Renewable Energy Laboratory (NREL) is managed by control software called G2. Data from the PDU equipment is made available on the network by Opto22 controllers in the plant. To facilitate communications between the Opto22 controllers and G2 a piece of interpreting software called a G2 bridge is required. The G2 bridge is written in C or C++ and runs independently of either G2 or Opto22. It translates messages from G2 to Opto22 and back. Through the course of developing this bridge it has become apparent that the ideal solution for controlling the PDU is not G2. Rather, it seems more efficient to have a piece of software designed specifically for communicating with Opto22 controllers in the PDU. Because G2 is intended to be generic control software, it requires a great deal of maintenance. Although a custom application would be less robust than G2, it would be easier to use and the loss of generality would not affect the performance of the software as far as the PDU is concerned.

The Response Surface Method: Determining the Minimum Step Size and the Minimum Size of the Region of Interest. MICHAEL BARNARD (Portland State University Portland, OR 97207) WESLEY JONES (National Renewable Energy Laboratory, Golden, CO, 89401)

The Response Surface Method (RSM) was developed specifically to assist the experimental scientist in minimizing the number of experiments they must run in order to optimize a process or device. Current literature extensively covers the concept behind the method and the statistics involved when measuring the gradient of the response surface. However issues such as, "What size should the Region of Interest (RI) be?", "What should the step size be?", or "When should a quadratic model be used instead of a linear model?" are left to the knowledge of the scientist. By investigating the relationship between the confidence interval associated with the gradient and experimental error, statistically based analytical answers to these questions can be given. However, new questions arise that must be answered. In this paper, the minimum size of the Region of Interest is expressed as a mathematical function of the desired confidence in the measured gradient. Using the null hypothesis for coefficients, the minimum step size can be expressed as a function of the measured gradient. However, in order for this equation to be useful, more serious consideration needs to be given to the issue of determining when an factor is locally insignificant. A criterion for switching to a quadratic model is a minor result of the Step-Size Equation.

Wind Turbine Control. DANIEL RATHE (Iowa Lakes Community College Estherville, IA 51334) LEE JAY FINGERSH (National Renewable Energy Laboratory, Golden, CO, 89401)

Wind turbines are fine machines for generating electricity, but for practical purposes they must start, operate, and shut down automatically. Fortunately, the National Wind Technology Center (NWTC) works with organizations in the U.S. wind industry to design, test, and advance automatic wind turbine control. Currently, the NWTC is implementing a new wind turbine control system in LabVIEW that has the potential to increase the efficiency of wind turbine control research and testing. More efficient research and testing at the NWTC may facilitate quicker advances in wind turbine control systems and increased understanding of such systems; such advances and increased understanding will allow the U.S. wind industry to be more competitive with the global wind industry as well as other sources of energy. The following describes the NWTC's controls research facilities; it describes the basic components of a wind turbine and concurrently shows parameters of these components that are monitored and used to control the wind turbine. Also, the following describes to a limited level of detail how a wind turbine control system gathers and uses the parameters to control a wind turbine. Finally, it illustrates the NWTC's two wind turbine control systems.