Self-Validating Control Loops: Extensions for Some Special Processes

摘要

The possibility of developing self-validating control loops has recently been examined else where. That work has focused on the isolation of faults in open loop stable processes, of type number zero, whose control loops inherently eliminate steady state error. The approach is to reason about abnormal, but steady state, deviations in measured, control loop related variables. These ideas are extended here to consider cause-effect knowledge pertaining to the open loop unstable case and to processeswith type numbers greater than zero. The former is applied to a simulated CSTR process.