A novel pattern-based approach for diagnostic controller performance monitoring

This paper details a novel method for monitoring the disturbance rejection performance of controllers by applying a second-order underdamped model as a pattern recognition tool. A controller performance index based on the second-order model parameters classifies the patterns into diagnostic categories of sluggish, well-behaved and overly aggressive. The autocorrelation function (ACF) has been used in numerous performance assessment capacities, and this work builds on these successes by applying the pattern recognition method to automate the ACF assessment across the full range of disturbance rejection performance. In addition to the performance diagnostic, a pattern-based visual tuning guide is presented for retuning PI controllers to regain desired performance. The performance assessment method has been tested on numerous control loops in a 25 MW cogeneration power plant and results of the application are presented.