Mission: AI-supported condition monitoring creates valuable time advantage

Siemens | June 2022

Honestly: Every plant operator or maintenance technician knows them: the 'problem children'. I'm referring to plant components or machines for which you'd rather do an extra monitoring round just to make sure that everything is really under control. After all, failures in mechanical components can often lead to prolonged plant downtime. An efficient condition monitoring solution such as SITRANS SCM IQ can do these checks for you.


The availability of your machines and systems is the prerequisite for your productivity - and only with efficient condition monitoring for the early detection of mechanical machine damage can this availability be guaranteed in the long term.  SITRANS SCM IQ offers a condition monitoring solution that can be applied to any plant component, e.g. compressors, pumps, agitators or heat exchangers. And the best part: From installing the appropriate sensor to logging into the cloud, you will only need 15 to 30 minutes - without any prior knowledge! How is that supposed to work? I'll be happy to explain.


Plant-wide or multi-site asset monitoring

SITRANS SCM IQ stands for Smart Condition Monitoring using IIoT sensors. The sensors are located on rotating or vibrating plant components and collect data that is then sent to a cloud application via gateways. There, artificial neuronal networks analyze vibration characteristics and monitor them continuously. Event-based warnings from the system alert you in a timely manner of an impending failure of these assets.


The solution is designed for plant-wide or even multi-site asset monitoring. If you are already taking advantage of MindSphere, you can add it as an additional module to your existing digitalization solution. But the system is also ideal for beginners: Existing processes are not affected in any way, and the cabling and installation effort is almost zero. The entry price for the system is so low that there is often a return on investment after just three months.


But how does this work exactly?

What components do you need and what are the benefits? First of all, you must connect Sitrans MS200 multisensors directly to the plant components to be monitored. Depending on how closely meshed your monitoring concept is designed, multiple sensors on one component are also possible. Wiring is not necessary: Vibration, acceleration or temperature data is sent via Bluetooth Low Energy (BLE) to one of our SITRANS CC220 cloud gateways. They collect the data and transmit it securely to the MindSphere industrial IoT-as-a-service solution. End-to-end encryption is guaranteed from the sensor to the cloud and within the applications. The real intelligence then finally comes from the Sitrans SCM IQ app which can be operated and output via mobile or stationary devices. 


Sounds too good to be true?

But it is! With SITRANS SCM IQ, your time and effort is truly manageable. You just need to consider which components you want to monitor. The integrated intelligence of our system takes care of the remaining effort.


Once the sensors are online and the cloud application is supplied with the raw data, the learning phase can begin: For a period of approximately two weeks, the system learns about the condition of an asset. We train an individual model for each sensor. This allows us to dispense with exact installation specifications. Again, there is hardly any effort on your part: You simply need to mark in the app the period of time during which a sensor collected data on a machine in normal operation. Based on the data from this time period, the solution's artificial neuronal network independently learns to recognize the operating states recorded during the test period as good states. Really smart, isn't it?


Early detection of anomalies

However, the system also detects anomalies that are not due to regular load or operating conditions. The maintenance specialists are informed by SMS or e-mail when a machine part is on its way to "bad condition". Based on this, they check the corresponding machine, search for the cause, e.g. signs of early bearing damage, which is then entered into the app. It is precisely this step that constantly optimizes the system. In addition, you can also document which maintenance steps were performed on the machine. If the same characteristic deviation occurs at any time, SITRANS SCM IQ reports that it is very likely the same incipient fault. But what happens if an unlearned condition occurs during ongoing plant operation? In this case, SITRANS SCM IQ first reports an anomaly for you to check. The time period is marked by the user and provided to the system for so-called re-learning. In the future, the system will evaluate the new plant condition as good condition.


Prevent downtime before it occurs

Not only do you benefit from easy installation, fast commissioning and hassle-free scalability from very small installations to extensive plant monitoring. Smart Condition Monitoring also gives you a valuable time advantage! This is because you can effectively avoid unplanned plant shutdowns, since you have sufficient time to take care of machine anomalies in advance.

I think it is important to move away from ad hoc error detection, which is still common practice in many companies today, and to focus on optimizing planned maintenance measures - the only way to ensure significantly improved plant performance. Do you have any questions or need more information about our IIoT solution for Smart Condition Monitoring? Don't hesitate to contact your Siemens representative.