Machine Monitoring with Smart Sensors
Tough demands on sensors
Sensors play a key role in Industry 4.0 as the artificial “eyes and ears” of machines and workpieces, for which they capture information about condition and performance. To facilitate intelligent management and connectivity in manufacturing, the sensors have to collect and process huge amounts of data in real time. They also need to be as energy efficient as possible and be easy to integrate into complex production systems. The industry sensors currently in widespread use are limited in their usefulness for Industry 4.0. For many applications, they are not smart or flexible enough, consume too much energy, and are too expensive.
MEMS sensors for industry
To fulfill the objectives of the AMELI 4.0 research project, the researchers are turning to one of the key technologies of the connected world: MEMS sensors (MEMS stand for microelectromechanical systems). Even now, it is impossible to imagine cars and consumer electronics without MEMS sensors. They are the core component of the ESP® anti-skid system, for example, and also ensure that the display on a smartphone screen rotates when the device is turned. Compared to conventional industrial sensors, MEMS sensors are small, smart, energy efficient, and economical. However, in many respects they are not yet robust or powerful enough for the demands of an industrial environment. This means that some of the potential to apply condition monitoring in production systems is going untapped. The AMELI 4.0 research team plans to further develop MEMS sensors to make them suitable for industrial applications. Energy supply plays a major role here: the new system will not require either power cables or batteries. It is designed to be completely self-sufficient by generating the necessary power itself from the machines’ vibrations (energy harvesting).
The difference is in the sound
To monitor the machines, the new sensor system will measure two types of noise: structure-borne sound, meaning vibrations inside the machine, and acoustic sound, meaning noise emitted by the machine. When a machine is not working as planned, it vibrates and sounds different than it does when operating normally. The system compares the measured signals with stored profiles. It continues learning, and takes action only if the changes in the signals indicate a defect or wear and tear. As a result, in the future the sensor system will be able to detect when a machine needs maintenance or repair. In more complex systems, this smart evaluation can be handled by the gateway (or router as it is sometimes called), to which the sensors transmit their data, or the manufacturing facility’s computer network.
Комментарии