| This thesis presents a hybrid architecture for a real-time, supervisor, Knowledge Based Control System (KBCS). This architecture provides a framework for the integration of knowledge obtained through operational experience on a process with the phenomenological processing model obtained through fundamental mechanisms occurring in the process. Key features in this architecture include: (1) The use of an on-line process model to assist the operator in the decision making process, (2) A blackboard system for the communication between various knowledge sources, (3) A metalevel control of the inference procedures involved, (4) The concept of continued operational plan refinement to meet the overall goals of the process operation in the presence of continuously changing external environment stimuli, and (5) The utilization of time-stamped data from the process.;The proposed architecture is applied to two composite manufacturing processes, autoclave curing process and pultrusion process, through simulation studies. The simulated study for autoclave curing process show that considerable improvement in performance can be gained by using this technology for on-line control. The study on pultrusion process illustrates the use of this architecture for qualitative off-line optimization of the process. |
