 Hello, my name is Vipin Botra. Today we are at IoT World 2018 and I'm talking about here our demo that is vibration sensing and predictive maintenance. In this setup what we have is a motor that is connected to a central gateway board by means of a sensor board and IO-Link interface. There is also a motor driver that is driving this motor so I can basically by means of this run this motor by the motor driver board. There is a green color motor that is a good motor and there's a red color motor that is not so good motor. In the red color motor what we have done is that we have unbalanced the load and the rotor so that the motor is vibrating a little bit extra. The sensor board that is connected on top of each of these motors is the key piece. That board includes a MEMS accelerometer plus gyroscope. It is also including temperature plus humidity combo sensor, analog microphone and an STM32L4 for processing the data on the edge itself and then IO-Link interface takes care of taking the data from the board to the central board. This IO-Link master board in turn then takes the data and pass it to PC via USB interface. Same thing is happening on this side of the setup with the bad motor. So the data that we are capturing by means of these sensors on these motors is in turn passed on to PC and plotted on a graph that is showing X, Y, Z axis of frequency and amplitude that is being captured on these motors. The red color one is coming from the red motor. The green color one is coming from the green motor. Since the red motor is vibrating more we can see that at 50 Hertz the amplitude is almost four times more compared to the good motor. It's same thing is happening on the Y axis and on Z axis there is a huge bump in amplitude at about 1.5 kHz. So all of these sensor data coming from this motor are telling us that is significantly different compared to the good motor and that in turn is enabling our prediction that there might be something wrong with this motor. You can further combine the data with the temperature humidity and the MEMS microphone giving you ultrasound information to make this more powerful prediction based analysis. So vibration sensing is not new. It has been around for a long time. What ST is enabling is a very small compact cost effective power efficient evaluation kit to get you started for your prototype and replace power hungry complex piezo based sensors with a very small module that can potentially be applied to the motor. If you pair that with it also with the cloud connectivity solutions from ST you can potentially have a solution that can shift the data directly to cloud and gives you analysis in the cloud to monitor the motors remotely. Thank you for watching.