The analogML™ core delivers the lowest power always-on edge processing strategy for handling the massive amounts of vibration data that are collected for predictive maintenance solutions. Built on Aspinity's unique RAMP™ technology platform, the analogML core offloads essential computations from the MCU that can be done more efficiently in analog circuitry to reduce the size and extend the battery-life of each sensor node by:
1. Minimizing the quantity of data being sent to the MCU by pulling specific, important information out of raw, analog sensor data
2. Generating actionable data for the MCU through analog data analysis (crest factor, peak velocity, etc.)
3. Alerting users to early changes in analog spectral content that indicate upcoming machine fault
By intelligently identifying and analyzing the most important datapoints while they are still analog and keeping digital processing to a minimum, the analogML core greatly allows maintenance teams to focus on early failure warnings and minimizing down time instead of on sifting through irrelevent data or changing batteries.
With the analogML core, you decide how much processing and analysis you need. From receving all extracted information in order to follow trends and monitor industrial health to only receiving alerts when a condition fault is detected, the functionality of the analogML core is fully software programmable using Aspinity's front-end development environment.