From Condition Monitoring to Real-Time Control: The Power of Live Feedback Loops
What Is Condition Monitoring?
Condition monitoring is the process of using data from sensors placed on a machine to understand its condition and predict failures or schedule maintenance. This is a rich field of work, and can involve bespoke sensors, complex models using limited data points, or even basic thresholding of data values. For decades, users have refined their condition monitoring models, to try and get more accurate predictions of failure, and to stretch maintenance schedules.
The Limits of Traditional Condition Monitoring
While condition monitoring has been a foundational tool in industrial operations, it has key limitations:
- Reactive, not proactive: Many systems only detect failure once symptoms are obvious
- Isolated signals: Data is often siloed, lacking broader system context
- Slow response time: Investigations can require full shutdowns to trace issues
In complex systems, these drawbacks can translate into costly downtime, safety risks, or missed performance optimization opportunities.
How Live Digital Twins Take Condition Monitoring Further
Until recently, it has not been possible in many cases to fully understand such complex systems, and thus to ensure that optimal performance is maintained without unacceptable safety risks. This is where live digital twins can help. For the first time, engineers and operators can have a full picture of machine behaviour in real time. This means not only are the insights from traditional condition monitoring systems replicated in the twin, but a far deeper understanding of what the machine is doing is possible.
This could mean that: unusual vibration patterns can be traced to wear debris from over loaded interfaces causing disruption to machine patterns; unusually low flow rates could be attributed to disturbed patterns within a mixing bowl caused by standing waves; poor product quality can be traced back to a blunt or broken cutting head which can be immediately identified and replaced. All of the issues described here would become a safety issue if left untreated, and traditional condition monitoring will not allow for such rapid diagnostics, requiring the whole machine or line to be shut down for investigations.
Better still, live insights on problems such as overfull mixers/separators, too high/low flow rates, or overheating can be addressed by control changes, which can change the loads on machine components, preventing damage or wear. These kind of rapid actions can be taken because of the insights gained from a live digital twin. In fact, using intelligent data pipelines, such reactions can even be automated using set point control.
Case Study: How SKF Uses Live Digital Twins
See how SKF are using digital twins to enhance their condition monitoring, and check out the full whitepaper here!
Key Definitions:
Digital twin: A virtual representation of what a complex real world system was doing
Live digital twin: A virtual representation of what a complex real world system is doing
Intelligent data pipeline: A microservice based system allowing for the transfer, processing and buffering of data from many different sources to many possible destinations.
Condition monitoring: The process of using data gathered from sensor on a machine to predict failures or schedule maintenance.
