Chaotic behavior control in fluidized bed systems using artificial neural network. Quarterly progress report, October 1, 1996--December 31, 1996
Pressurized fluidized-bed combustors (FBC) are becoming very popular, efficient, and environmentally acceptable replica for conventional boilers in Coal-fired and chemical plants. In this paper, we present neural network-based methods for chaotic behavior monitoring and control in FBC systems, in addition to chaos analysis of FBC data, in order to localize chaotic modes in them. Both of the normal and abnormal mixing processes in FBC systems are known to undergo chaotic behavior. Even though, this type of behavior is not always undesirable, it is a challenge to most types of conventional control methods, due to its unpredictable nature. The performance, reliability, availability and operating cost of an FBC system will be significantly improved, if an appropriate control method is available to control its abnormal operation and switch it to normal when exists. Since this abnormal operation develops only at certain times due to a sequence of transient behavior, then an appropriate abnormal behavior monitoring method is also necessary. Those methods has to be fast enough for on-line operation, such that the control methods would be applied before the system reaches a non-return point in its transients. It was found that both normal and abnormal behavior of FBC systems are chaotic. However, the abnormal behavior has a higher order chaos. Hence, the appropriate control system should be capable of switching the system behavior from its high order chaos condition to low order chaos. It is to mention that most conventional chaos control methods are designed to switch a chaotic behavior to a periodic orbit. Since this is not the goal for the FBC case, further developments are needed. We propose neural network-based control methods which are known for their flexibility and capability to control both non-linear and chaotic systems. A special type of recurrent neural network, known as Dynamic System Imitator (DSI), will be used for the monitoring and control purposes.
- Research Organization:
- Tennessee State Univ., Nashville, TN (United States). School of Engineering and Technology
- Sponsoring Organization:
- USDOE Assistant Secretary for Fossil Energy, Washington, DC (United States)
- DOE Contract Number:
- FG22-94MT94015
- OSTI ID:
- 493394
- Report Number(s):
- DOE/MT/94015-T7; ON: DE97052875; TRN: 97:004324
- Resource Relation:
- Other Information: PBD: 27 Feb 1996
- Country of Publication:
- United States
- Language:
- English
Similar Records
Neural Network Based Montioring and Control of Fluidized Bed.
Iterative prediction of chaotic time series using a recurrent neural network. Quarterly progress report, January 1, 1995--March 31, 1995