Explore a practical application of tinyML in battery-powered anomaly sensors for predictive maintenance of industrial assets in this 33-minute webcast. Discover how the combination of tinyML, low power wireless, integrated sensors, and IoT cloud enables a cost-effective and easily installable system for monitoring industrial equipment throughout a factory. Learn about the benefits of detecting anomalies early to prevent unplanned downtime and costly repairs. Witness a live demonstration of a tinyML-based end-to-end system solution. Delve into topics such as autoscaling, model training, anomaly triggering, raw acceleration data usage, low power wireless communication, machine learning approaches, physics-based modeling, sensor types, and other applications. Understand the differences between Grey Zone and Black Zone, explore the potential for sensor model updates, and assess the susceptibility of anomaly detectors. Gain valuable insights into this innovative approach to condition-based maintenance in industrial settings.
Practical Application of TinyML in Battery Powered Anomaly Sensors for Predictive Maintenance of Industrial Assets
Overview
Syllabus
Introduction
Demonstration
Autoscaling
Training the models
How do you trigger an anomaly
Why raw acceleration data
Low power wireless communication
Machine learning approach
Physicsbased modeling
Sensor types
Other applications
Grey Zone vs Black Zone
Can the sensor train update its model
How suspectable is the anomaly detector
Contact information
Sponsors
Taught by
tinyML
