About the Author
Jose Flores is passionate about developing and providing solutions for the Maintenance and Reliability Industry, strong manufacturing and engineering background which helps understand industry’s needs and challenges.
Imagine trying to measure the a 100° F temperature with a Thermometer with a high value of 80° F, or your local police trying to catch speeding cars with a radar gun that only goes to 20 MPH. Both situations share the same problem, the sensor is not appropriate for their use!
Hence the concept of Observability, the ability to measure the internal states of a system by examining its outputs. A system is considered” observable” if the current state can be estimated by only using information from outputs, namely sensor data.
It can also be defined as a process in which you can infer the internal running status of a system based on external outputs.
Reliability Professionals are faced with a wide variety of sensors and systems promising to be the best, yet OEMs rarely spend the time with their customer deciding if the sensor, or sensors, they are recommending are appropriate for their needs.
The best approach for Sensor Selections is an Observability Analysis and Fault Assessment, some important aspects you, and your supplier, need to establish are:
– Sensor type, quantity, location and mounting.
– Does the sensor provide enough information? What is the frequency response?
– Sensor arrangement and Redundancy
An Unobservable system will create an incorrect estimation which in turn provide useless results. Make sure to consider all sensor types and brand, consult with unbiased advisors and save yourself and your organization time, money and headaches in the long run.
#conditionmonitoring #predictivemaintenance #vibrationanalysis
Jose Flores is passionate about developing and providing solutions for the Maintenance and Reliability Industry, strong manufacturing and engineering background which helps understand industry’s needs and challenges.
I would like to amend the great article by adding the inputs as well as the outputs as a set that “bounds” the “system” under study.
The mai reasons are:
1) Sometimes the output changes due to variations in the inputs that might trigger false alarms.
2) For nonlinear systems (which is the case for more engineering systems), observability depends of the inputs and region of the state-space where the system is.