The Awesense Energy Transition Platform’s AI Data Engine tool ingested and integrated data from measurement sources active in the grid. During the ingestion process, the AI Data Engine executed its Validation, Estimation and Error Correction (VEE) algorithms against all geospatial, connectivity and time series data entering the system. The AI Data Engine discovered over 500,000 quality issues. Following multiple iterations, 97% were corrected in a matter of weeks.
The Engine also synchronized all time series data and associated time series measurements with their geospatial representation in the network model. The result was a uniform grid model built according to Awesense’s Energy Data Model (EDM), from which applications could easily access all data and other Duke internal systems requiring access to critical data. Utilizing the EDM, various applications provided by Awesense performed grid-wide geospatial analysis.
The final product was a situational awareness engine capable of displaying higher quality real-time data anywhere in the distribution grid from various sources. Historical data was used to understand trends and issues occurring over time in the network, and data was continuously ingested to maintain a thorough understanding of grid performance.
