Intelligent control of mixed-culture bioprocesses
- Idaho National Engineering Lab., Idaho Falls, ID (United States); and others
A hierarchical control system is being developed and applied to a mixed culture bioprocess in a continuous stirred tank reactor. A bioreactor, with its inherent complexity and non-linear behavior was an interesting, yet, difficult application for control theory. The bottom level of the hierarchy was implemented as a number of integrated set point controls and data acquisition modules. Within the second level was a diagnostic system that used expert knowledge to determine the operational status of the sensors, actuators, and control modules. A diagnostic program was successfully implemented for the detection of stirrer malfunctions, and to monitor liquid delivery rates and recalibrate the pumps when deviations from desired flow rates occurred. The highest control level was a supervisory shell that was developed using expert knowledge and the history of the reactor operation to determine the set points required to meet a set of production criteria. At this stage the supervisory shell analyzed the data to determine the state of the system. In future implementations, this shell will determine the set points required to optimize a cost function using expert knowledge and adaptive learning techniques.
- Research Organization:
- Argonne National Lab. (ANL), Argonne, IL (United States)
- DOE Contract Number:
- AC07-94ID13223
- OSTI ID:
- 175492
- Report Number(s):
- CONF-9505200-; ON: DE96000983; TRN: 96:003524
- Resource Relation:
- Conference: 13. symposium on energy engineering sciences, Argonne, IL (United States), 15-17 May 1995; Other Information: PBD: [1995]; Related Information: Is Part Of Thirteenth symposium on energy engineering sciences: Proceedings. Fluid/thermal processes, systems analysis and control; PB: 275 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Experimental Validation of a Resilient Monitoring and Control System
Development and implementation of a fuzzy optimal expert controller for intelligent buildings