System health monitoring
The starter kit provides a reference implementation for system health monitoring use cases. The solution is designed to reliably evaluate system health and detect anomalies in a large number of IoT metrics received from multiple sensors and machines.
Multiple system types
The starter kit includes a design guide that helps determine the right model type based on the properties of the monitored system and availability of the labeled data.
Power of Vertex AI AutoML
The starter kit leverages native Vertex AI AutoML components to evaluate system health. AutoML sharply reduces the feature and model engineering efforts, enabling developers to focus on more business-oriented problems.
We use only native Google Cloud services and open source libraries, so the starter kit can be easily extended and customized.
The starter kit provides out of the box system health monitoring and anomaly detection pipelines. These reference pipelines can be used as a starting point for implementing a broader range of use cases related to Industrial IoT and Smart Manufacturing.
Remaining useful life estimation
System health monitoring and anomaly detection capabilities are important in many industries. The starter kit includes several templates that cover some of the most relevant ones.
Monitor equipment health and prevent downtimes.
Continuosly control the product quality and detect anomalies.
Get the most value from your smart grid investments.
Oil and gas
Implement sensor-based tank monitoring and asset monitoring.
Monitor electric vehicles and optimize EV charging.
Monitor production processes, ensure quality and compliance.
Why develop an IoT analytics solution in Google Cloud?
How it works
The starter kit includes health scoring models and detection algorithms that make binary decisions and generate alerts.
Anomaly detection in industrial IoT data using Google Vertex AI: A reference notebook
In this blog post, we focus on IoT data analysis challenges associated with system health monitoring and how to resolve them.
Anomaly detection in industrial applications: solution design methodology
In this article, we describe a solution design methodology for anomaly detection based on the labeled data types and availability. We delineate the general strategies for three types of data labeling here and point to some hidden pitfalls that frequently pass unnoticed and may result in productivity disruptions and even project failures.
Get in touch
We'd love to hear from you. Please provide us with your preferred contact method so we can be sure to reach you.
Thank you for getting in touch with Grid Dynamics!
Your inquiry will be directed to the appropriate team and we will get back to you as soon as possible.
Something went wrong...
There are possible difficulties with connection or other issues.
Please try again after some time.