How to Detect Wear and Tear on Equipment Before It Breaks Down
Share
Big data and AI promise to catch problems and tuning opportunities in your plant before they turn into costly failures. So, when I first heard about SID, Smart Industrial Diagnostics, I assumed it was another AI-based monitoring tool.
I was wrong.
SID uses smart mathematics, not artificial intelligence, to monitor PID loops and identify when a loop starts working harder to do the same job. It operates locally, with no cloud connection and no additional sensors required. Instead, SID analyses data your plant is already generating.
The idea comes from an IEEE journal article published by SID's creators, titled "Inverse-PID: A Mathematical Approach Towards Detecting Real-World Wear and Tear in Industrial Machines." You can download and read the original article here.
Why PID Loops Can Provide an Early Warning
Proportional-Integral-Derivative (PID) control is the backbone of modern factory and plant automation.
A PID controller compares the actual process output with its setpoint and continuously adjusts the control input to close the gap. It does this by using three tuning parameters – proportional, integral and derivative – commonly referred to as the P, I and D gains.
Under normal circumstances, we generally think about those gains as values used by the controller to control the process.
SID approaches the problem from the other direction.
It calculates gains based on the observed behaviour of the PID loop. By tracking how those calculated parameters change over time, SID can provide a measure of how hard the PID loop is working to maintain the desired process condition.
"SID can flag developing changes before equipment wear results in unplanned downtime"
From Gain Oscillation to a Wear Signal
This is where the concept becomes particularly interesting.
Trials showed that a properly functioning PID loop has a small amount of oscillation in these calculated parameters. As components wear and the PID loop has to work harder to achieve the same result, that oscillation increases.
By monitoring that change over time, SID can identify a developing wear condition before the process can no longer maintain its expected performance.
In other words, SID isn't adding another measurement to the machine. It is looking differently at information already contained in the behaviour of the control loop.
That's how the PID loop itself can become a source of information about equipment condition.
WHAT CAN SID DETECT?As equipment begins to wear, changes in the PID loop can provide an early warning. SID can identify changes associated with conditions such as:
No additional vibration sensors. No thermal cameras. No new wiring. SID uses control-loop data your PLC or DCS is already producing. |
From Research to a Practical Diagnostic Tool
The inverse-PID concept described in the research paper has now been developed into easy-to-use software for monitoring PID loops in an operating plant.
That distinction is important. You don't need to send process data to the cloud, add another sensor to the equipment or build an AI model. SID uses existing control-loop data and mathematics to identify changes in how the loop is behaving.
For maintenance and automation teams, that creates another way of looking for equipment wear using information the control system is already producing.
Want to See How It Works?
We're hosting a detailed webinar on September 21, 2026, where we'll take a closer look at SID, the inverse-PID concept and how it can be applied to industrial PID loops.
Please fill out the form below to register for the SID webinar →