Average Absolute Error (AAE)
AAE measures the difference between the Set Point and the measured Process Variable. This metric is used to track how well the controller is able to maintain its Set Point.
The output distribution identifies areas where the Final Control Element is spending its time. This allows users to understand if the control element is saturated (i.e. sitting at the extreme upper or lower operating range). This metric can be used to understand how well equipment is sized.
Output Reversals is a measure of the amount of times the controller output changes direction. A high number of output reversals suggests that excessive wear and tear on the Final Control Element or that the final control element is starting to fail (hunting for its position more erratically)
PID controllers are used to automatically regulate control and to maintain a process at a desired Set Point. When controllers are disabled, manual control is required. This metric tracks the number of changes (i.e. auto vs manual) in order to highlight potential problems
This is an estimation of the amount of high frequency noise present in the process variable. Increases in noise could indicate potential sensor failure.
This is the likelihood that the process variable is oscillating. It is the amount of the power of the dominant frequency divided by the total power from the spectral analysis. This gives an indication on whether the process variable has a single dominant oscillation.
The Overall Health is determined based upon the status and the user adjustable importance factor for each calculated KPI. The overall health provides a single performance factor to identify controller performance issues.
Percent Time in Normal
PID controllers are used to automatically regulate control and to maintain a process at a desired Set Point. When controllers are disabled, manual control is required. This metric tracks the percentage of time the controller spends in each available controller mode (i.e. auto vs manual) in order to highlight potential problems.
This metric showcases where your process is operating in terms of the Process Variable, Set Point, and Controller Output. Each gauge reflects values for Average, Variance and Controller Limits for the associated loop.
When PlantESP captures a window of data for analysis, in addition to the process data, it records an additional data value that is used to determine if the PID Loop is in service. The Uptime measures the percentage of time that the PID Loop is in service during the captured window.
Set Point Changes
This metric quantifies the number of times a control loop’s Set Point changes on a daily basis. A high and/or increasing number of Set Point Changes can indicate excessive operator intervention in the control of a given process.
The mechanical elements associated with moving a valve or damper can wear over time. As these linkages or other parts start to wear down, their ability to consistently move the valve stem or damper degrades. This metric measures the performance of the Final Control Element. This metric returns the likelihood and amount
PlantESP’s TuneVue module monitors process data to calculate PID Tuning Values. When the actual tuning values are outside of the recommended range, the tuning deviation metric can be used to highlight opportunities for PID Tuning
Output Travel is a measure of movement in the output signal. Output Travel gives us insight into how much effort by the Final Control Elements required to maintain control. Increasing amounts of controller effort (output travel) can indicate changes in performance of the controller or final control element
Some Other Metrics Include:
- PV & SP Travel – Travel is a measure of change in the Variable over a given time period. The process variable travel can indicate if your process is operating at steady-state or is consistently moving around. The SP Travel indicates how much movement your set-point exhibits.
- PV & SP Distribution – The distribution identifies areas where the variable is spending its time. This allows users to understand if the input sensor is saturated (i.e. sitting at the extreme upper or lower operating range). This metric can be used to understand how well sensors or processes are sized.
- PV & SP Variance – The Variable Variance is a measure of the spread of the Variable value. A tighter variance is desirable as it shows a well controlled process.
- CO Average – Average Controller Output is a measure of the typical value of the Controller Output signal.
- CO Variance – Output Variance is a measure of the spread of Controller Output values. A large controller output variance correspond to final control elements that have large amounts of travel.