When you hear about predictive maintenance...

                       what is the first thing, the first metter you are thinking of?

                Is it engine...



                                      or other mechanical devise?

Are you thinking about vibroacustics, electricity consumption and temperature?


Or rather are you thinking about the risk of corrosion of the most critical parts of your instalaton - tanks, trunk pipelines or valves?

Well, you have good associations. You can predict an upcoming failure of any device, based on the data from the measuring sensors.

                                                                  At least as long as the data reflect reality...

All data needed to feed the predictive AI models come from measurment and control system.

But do you care about this sytem as much as about other critical elements of the installation?

How can you be sure that the values given by the sensor are real?

In Silesian Catalysts we know how to solve this problem at a very primary level.

By analysing the entire spectrum of the measurement signal (measured value and measurement  noise), the Advance Process Sentinell can predict an upcoming failure of a sensor or group of sensors. It will inform you both about sudden events and about slow, natural degradation of sensors, so that you have time to react before an emergency shutdown occurs