Project No: FG02-08ER84944
Performer: Cybosoft General Cybernation Group Inc


Contacts
Robert Romanosky
Advanced Research Technology Manager
National Energy Technology Laboratory
3610 Collins Ferry Road
P.O. Box 880 P03D
Morgantown, WV 26507-0880
304-285-4721
robert.romanosky@netl.doe.gov

Susan Maley
Project Manager
National Energy Technology Laboratory
3610 Collins Ferry Road
P.O. Box 880 P03D
Morgantown, WV 26507-0880
304-285-1321
susan.maley@netl.doe.gov

George Cheng
Principal Investigator
CyboSoft, General Cybernation Group Inc.
2868 Prospect Park Drive
Suite 300
Rancho Cordova, CA 95670-6065
916-631-6313
gscheng@cybosoft.com

Duration
Award Date:  07/02/2008
Project Date:  12/31/2013

Cost
DOE Share: $849,369.00
Performer Share: $0.00
Total Award Value: $849,369.00

Performer website: Cybosoft General Cybernation Group Inc - http://www.cybosoft.com/

Crosscutting Research - Plant Optimization Technologies

Intelligent Actuation Control Using Model Free Adaptive Control Technology

Project Description

The project will develop, and bring to production, an intelligent actuation control solution technology that will adapt to deal with changes in valve gain, time constant, delay time, nonlinearity, and other non-optimum behavior. The project will evaluate, simulate, and determine the feasibility of using MFA control technology to develop an intelligent control solution for valve positioning control capable of meeting stringent performance specifications and effectively handling large variations in valve behavior. The expected/ anticipated solution will also provide more robust and precise position control and remove potential valve oscillations irrespective of the variability of valve characteristics and operating conditions. The commercial products will be designed to enable easy integration with new and installed valves, dampers, and other actuation devices, thus offering a smooth transition and minimal operator training and maintenance.


Program Background and Project Benefits

The Advanced Research Sensors and Controls Program is leading the effort to develop sensing and control technologies and methods to achieve seamlessly integrated and intelligent power systems. The program is led by the U.S. Department of Energy (DOE) Office of Fossil Energy National Energy Technology Laboratory (NETL) and is implemented through research and development agreements with other national laboratories, industry, and academia. The program strategy is to develop robust sensing approaches using durable materials and highly automated process controls to optimize advanced power systems operation and performance. The control and optimization of coal-fired power plants highly depend on coordinated and integrated sensing, control, and actuation technologies and products. Prior efforts to develop novel sensing and control technologies have been successful, but little work has been devoted to the coordinated control and actuation of power plant processes

In alignment with the programs mentioned above, NETL has partnered with General Cybernation Group Inc. to develop a solution for intelligent actuation control using Model-Free Adaptive (MFA) control technology. The project is funded through the DOE’s Small Business Innovation Research (SBIR) Program, a highly competitive program that encourages small businesses to explore their technological potential and provides them with incentives to profit from its commercialization.

The MFA actuation control solution addresses the need for clean coal technology and will help power and process industries improve efficiency, quality, safety, and emission reduction. Additionally, this technology can be commercially available software and easily embedded in digital valves or damper positioned products to benefit the U.S. energy plants of the future, as well as other industries where control valves are used.


Accomplishments

Phase I objectives were successfully completed, with encouraging results. A novel MFA actuation controller was developed that can effectively control valve position processes that have nonlinear backlash problems and large changes in valve gain, time constant, and delay time. The MFA valve position controller was able to control the linear and nonlinear valve position process models with consistent control performance in all 48 step test conditions. In contrast, the high-gain proportional–integral–derivative (PID) control, which is typically used for valve position control, failed in more than 50 percent of the test conditions. During Phase I, a novel, anti-delay, nonlinear MFA controller for valve position control was developed. This controller effectively managed large changes in valve gain, time constant, and delay time, and included a unique anti-backlash scheme to effectively deal with the backlash problem of the valve position process. Numerous closed-loop control simulations for valve position control were performed using the MFA controller and a PID controller. The linear and nonlinear valve position process models developed in this project were controlled using 24 varying test conditions for each model. Research was conducted on valve positioner products currently available with a viable MFA valve positioner product developed. A development plan was created to demonstrate that the MFA valve positioner can be tested side-byside with leading valve positioners for air and liquid flow control.

In Phase II, the process of developing multiple MFA actuation control products to effectively control actuation processes that have nonlinear backlash problems and large variations in valve gain, time constant, and delay time is under development.