Find fresh insight in some core PI tags you've probably had in your PI server since the beginning of time. Reduce losses and improve safety when you see issues sooner. What you need to know is a new way of listening.
Get ready to hear voices! Once you implement this strategy across a unit or section there will be lots of new voices talking. It will be like the folks who never say anything are finally speaking up.
Here are some things you might find: machine wear/age, process fouling, flow restrictions, cavitation, contaminants, vapor in a liquid service, poor calibration, machine replacements/repairs, material properties, chemistry, feed stock quality, upstream conditions, buffered energy, ambient conditions.
It's based on a very simple machine learning metric. Consider the PID controllers in your processes. You can model them as providers delivering something at a cost. The PID controller's output (OUTPUT) produces a variable process condition or process value (PV). For example, a pump's speed (OUTPUT) delivers varying flow rate (PV). The ratio then of (OUTPUT/ PV) is essentially a measurement of the controller's strain. Strain is often fixed during healthy operation. Degrading machine health or disturbances in your processes can show up right away in controller strain but not be noticed by watching PV or OUTPUT alone. For example, pump cavitation shows up immediately as drop in flow rate. The PID controller responds instantly by increasing the pump speed. Flow rate then matches setpoint but the operator is unaware of the higher erosion that is damaging the pump if allowed to run in this state.
Shown below is a pump involved in a flow controller as indicated in the control sketch.
Your PI server (example kit attached below) can continually calculate and qualify shifts in controller Strain. The surrounding data in your PI system can account for downtime, dismiss noise and scrub your results to produce meaningful alerts. You can Backfill these PI analytics as far back in your past as you like to instantly see the big picture of what the controller Strain has been. You will see isolated events and recurring events. Some anomalies you find will be progressive over time (such as fouling, corrosion, blockage) lend themselves to being forecast with additional PI analytics.
Controller Strain tracking is a primary diagnostic for control valves. Control valves are used throughout many processes so the impact of this technique at scale can be large. Again, All you need is PV and OUT performance data to track Strain. The insight you find can be significant. Shown below are typical control valve installations.
A PI AF analytic can be used to analyze patterns in Controller Strain over time and detect when significant shifts have occurred. Industry-standard pattern recognition techniques for Statistical Quality Control (SQC) are useful in this application. PI Templates are a must to allow for mass admin across a large installed base. One new PI tag is required to store the Strain for each controller. Periods of unusual performance are captured as PI event frames and stored by asset. Let PI sends detailed email/Text messages with all of the pertinent stats for a shift in Strain to multiple subscribers. Keep everyone in informed immediately. Emails contain detailed instructions for investigating root causes and next steps. Users need not monitor dashboards but rather wait to be notified of the shifts. PI Vision, PI System Explorer, or PI Datalink for Excel can each be used to study event history. Here it is important to also let PI send eMail messages when the Strain event returns to normal. (Sample PI AF message formats and best practices are provided in the attached kit).
This technique can reveal dependencies and interactions between variables in your process. This can be a natural first step in understanding your interactions. This can form a baseline for describing multi-variable model predictive control strategies. The question of linearity is a good one here too. Many devices operate within the linear range of their output but some do not. Devices working over a larger range of their output often produce non-linear behavior in normal operation. These will require additional allowances for linearization in your PI templates beyond the samples provided here.
Feel free to post any questions here or contact me directly. Don't hesitate to reach out to OSIsoft for support in getting started.
Best of luck,