team photo

Figure 1
project photo

Figure 2
project photo


Team 20

Team Members

Faculty Advisor

Mike Rossi
Brandon Mino
Cory Harrington

Ion Mandoiu

sponsored by
sponsor logo

Securely Streaming PMU Data To The Cloud

We worked with ISO-NE for our senior design project.  ISO-NE is a non-for-profit organization responsible for managing the power grid in New England.  To give a little background, ISO-NE has on-premise systems that their control centers rely on for real-time operation.  These systems are currently configured for failover, however they may still be subject to system failures, thus a cloud based backup solution was needed.  The data that control centers rely on is called PMU data, which are sourced from Phasor Measurement Unit (PMU) devices located on the electric grid throughout New England.  They provide time-stamped measurements of voltage, current, and frequency and are important because they reflect real system states. Our group was tasked with securely moving this PMU data to the cloud from the on-premises infrastructure.  For security we utilized the TLS protocol and for our cloud service we used Amazon Web Services (AWS).  Our project can be divided into 4 components, the PMU Adapter, the Kinesis-DynamoDB Adapter, the Database, and Kinesis Data Analytics.  The PMU adapter receives data from the PMU devices, parses the data accordingly and pushes it to a Kinesis stream.  The Kinesis-DynamoDB Adapter processes the data from the Kinesis stream and writes the data to DynamoDB, which is where the PMU data ends up being stored.  This is our main workflow.  For deployment, the PMU Adapter and Kinesis-DynamoDB Adapter are written in C++ / NodeJS and Java respectively, both have a Docker image, and both will be deployed using the ECS container service on AWS.  Kinesis Data Analytics is integrated with the Kinesis stream, performs analytics, and sends them to another DynamoDB table.  This allows us to monitor system status and detect issues in real-time. The featured image is a diagram of our whole workflow, figure 1 expands on the PMU Adapter workflow, and figure 2 expands on the Kinesis-DynamoDB Adapter workflow.