Evaluation of a statistically based controller override on a pilot-scale flow loop.

Area of Science:

• Process Control
• Statistical Modeling
• Automation Engineering

Background:

• Conventional automatic controllers often struggle to balance variance reduction between controlled and manipulated variables.
• Existing filtering techniques may not provide optimal performance in all scenarios.

Purpose of the Study:

• To investigate a statistically based override for automatic controllers.
• To temper controller action for improved variance reduction.
• To achieve a better balance in variance reduction for both controlled and manipulated variables.

Main Methods:

• Implementation of a statistically based override on an automatic controller.
• Experimental validation on a pilot-scale flow loop.
• Supporting analysis through simulation and theoretical examination.

Main Results:

• Demonstrated benefits of the statistical override in experimental results.
• Improved balance of variance reduction in controlled and manipulated variables.
• Validation of the approach through simulation and theoretical analysis.

Conclusions:

• The statistically based override offers a superior method for tempering controller action.
• This approach enhances the balance of variance reduction compared to conventional filtering.
• The findings are robust, supported by experimental, simulation, and theoretical evidence.