 Hvala, mi je Roberto Sannina, da vse možemo vzlušati do stvarnosti CES 2018 v Vagavskih hodnih hodnih v Las Vegas. V tem demosetih vzlušajemo vzlušanje vzlušanje konditionu, kaj je solucija vzlušanje kaj je vzlušanje IOT. V zelo vzlušanjem vzlušajemo dve vzlušanje kaj je vzlušanje vzlušanje kaj je vzlušanje in počkaj je zelo izgleda. Kaj smo počkali na sistem, nekaj z dva motori izgleda, ker je zelo izgleda. Tako zelo semulujemo na red motoru, če je zelo izgleda. Kaj je zelo vse motori, mekanikali, nekaj elektrikali, There are two little boards, two little sensor boards, as shown in this diagram. You see, these little boards have a couple of interesting characteristics. They have an IoT architecture where an STM32 can manage signals taken from several sensors, mem sensors, accelerometer, temperature, humidity, and pressure, and do some computation and send the results out through an IO link, that is a typical industrial connection protocol. And the board has a shape that can host a typical in the C standard M12 connector. These two boards, as I was saying, are mechanically connected to each of the motor, but not electrically. So they can perform an analysis of the vibration that are ongoing in the motor and also of the environmental parameters. As we can see now, once the analysis is performed, we can see frequency by frequency up to two kilohertz in this case. The behavior of each of the two motors are presented with different color. The orange motor shows an amplitude of vibration that is much higher than the green motor. And this is a signature of a potential problem with this motor. We're not saying the engine is broken, but that there is something wrong with it. And if we don't take corrective actions, the motor will break. Probably what will happen in the next days is that the motor will heat more, heat up, and then later on it will break. So we can also check environmental parameters, such as temperature, which is very important in this case, but also humidity and pressure with this little board. Based on this analysis, we can basically perform with an IoT style of architecture that is a little board that is not done into an existing system, the condition monitoring operation, and we can do predictive maintenance. That is, we can predict when, if the system is about to be broken. Now we move to a different technology based on acoustics with this other system. Here we have a system where a probe implements on its tip an analog MEMS microphone. This analog microphone sends its signal into this little box where an STM32 interprets the audio signal and performs a frequency analysis for all frequencies from 0 to 97 kHz. Like most of us know, up to 20 kHz we have the audible range of frequencies, but then, from 20 kHz upwards, we are in the ultrasound range. The ultrasound range is particularly interesting. As you can see from this live spectrogram, all frequencies between 0 and 20 kHz have a reach with content generated by me speaking, other engines doing their work and so on and so forth. But for all frequencies higher than 20 kHz, we see basically a flat spectrum, no specific signature. Now, in this demonstrator, we turn on the simulation of a leakage in a gas pipe. So when there's a leakage, the leakage is produced by a small hole, the smaller the hole, the higher the frequency typically will be. So we don't know exactly what frequency will show the signature, but moving this microphone tip, microphone probe close to the pipe, we can check the presence of those signatures. And here we go. You can see a very high spike here at 60 kHz, and as we move the probe along the pipe, we can find in another position another even higher signature at 40 kHz in this case. So we have detected by means of ultrasound analysis two potential problems in this gas pipe. So finally, our proposal is to combine all this sensing capability into a single low-cost system that could be capable of doing its own computation on board, could be connected with any of the typical wireless connections, Bluetooth, subjigards, Wi-Fi, and so on and so forth, or could be connected via IO link, via typical industrial links so that the system can be constantly monitored and as soon as the problem arises, the operators are informed timely of the potential problems. Thank you for your attention.