Looking forward to seeing more ML content from RealPars. It seems like ML is going to change the way we approach maintenance but there are very few practical resources to learn about ML in industrial automation.
Yes and no. If you are able to know the "normal" value of the parameter before running the equipment, then yes, tolerance would be the guideline for determining a problem. For new equipment, or processes where the conditions are dynamic, machine learning (modelling) can be very useful to determine the behavior of the machine in normal operation. But in either case, detection of a condition outside of the expected range will lead to an alarm, or alert.
Except that it isnt that obvious when a motor is deviating from its optimal value, usually, for a point to be that far from the normal operation point, what you require is not a predictive maintenance, is more about corrective maintenance and replacing failing pieces that already damaged the motor
Most of the time, you require more expertise to be able to tell if a parameters is an anomaly, perhaps the same vibration could be considered normal at high speed and an anomaly at low speed.
Predictive maintenance tools foresee equipment failures before they occur. SmythOS incorporates these tools to improve maintenance schedules and reduce downtime. #PredictiveMaintenance #ManufacturingAI #SmythOS
I still don't see the benefits of machine learning, if I can only use a single check, eg. IF (Sensor_A > 7) OR (Sensor_B > 7) THEN ... Or if I want to prevent false alarms, then I'd use rising triggers and a counter.
Thanks for the video. Short and direct, perfect for my students
Looking forward to seeing more ML content from RealPars.
It seems like ML is going to change the way we approach maintenance but there are very few practical resources to learn about ML in industrial automation.
Wait, isn't it the same as generating alarms based on tolerance?
Yes and no. If you are able to know the "normal" value of the parameter before running the equipment, then yes, tolerance would be the guideline for determining a problem. For new equipment, or processes where the conditions are dynamic, machine learning (modelling) can be very useful to determine the behavior of the machine in normal operation. But in either case, detection of a condition outside of the expected range will lead to an alarm, or alert.
RealPars is best in the world, from Brazil.
Thanks a millino, Alexandre!
Realpars just freaking rock ,thats awesome
Thank you so much, Ayman!
That's great , looking forward for the new courses as machine learning is becoming an integral part of multiple industries
Thank you, Sam!
Hope to see more videos on this topic!
Thanks for your feedback, Jorge! I will happily go ahead and pass that on to our course developers. Happy learning!
Except that it isnt that obvious when a motor is deviating from its optimal value, usually, for a point to be that far from the normal operation point, what you require is not a predictive maintenance, is more about corrective maintenance and replacing failing pieces that already damaged the motor
Most of the time, you require more expertise to be able to tell if a parameters is an anomaly, perhaps the same vibration could be considered normal at high speed and an anomaly at low speed.
@@wtomas250 we don’t ever consider Vibration normal specially at high speed ⚠️
Predictive maintenance tools foresee equipment failures before they occur. SmythOS incorporates these tools to improve maintenance schedules and reduce downtime. #PredictiveMaintenance #ManufacturingAI #SmythOS
Congratulations! Real parts is heading to an interesting direction!
Thanks for your feedback, Mohammed!
Great video! Thanks.
Glad you liked it!
RealPars u 4 the best !!!
Thanks a million!
Can you make detailed video on siemens kinematic tech objects? I think you would make an interesting video.
Thanks for your topic suggestion! I will happily go ahead and pass this on to our course developers.
This is very informative lecture. Thank you very much sir.
You are most welcome, Shamsi!
Perfect
Dear sir please post the resolver or RVDT Video
Thanks for your suggestion, Mansoor! Happy learning
I still don't see the benefits of machine learning, if I can only use a single check,
eg. IF (Sensor_A > 7) OR (Sensor_B > 7) THEN ...
Or if I want to prevent false alarms, then I'd use rising triggers and a counter.
I think the point of this is for the model to adapt on its own to changing situations, otherwise you're right.
👍
Hope to see more videos on this topic!
More to come! Thanks for your support, Mohamed!