In order to improve customer phase and meter-to-transformer associations, PwC utilized the Awesense AI Data Engine to cleanse and structure FortisBC’s GIS and time series data into a single grid connectivity model. GIS data often has extensive connectivity errors and is typically more difficult to resolve. Time series data, comprising data from 10,593 meters, required structuring and synchronization in order to work in the model.
“Awesense’s ability to work with large volumes of energy data has helped streamline FBC’s efforts to leverage the value of the significant amounts of data generated from various systems, including our Advanced Metering Infrastructure system.” – Michael Leyland, Manager of Innovative Initiatives at FortisBC
Awesense’s AI Data Engine validated, estimated, and corrected the data at scale. Thanks to AI processes employed by the engine, PwC was able to deliver the project 12X faster than standard utility data delivery methods. The AI Data Engine then synchronized the data in time and space according to Awesense’s Energy Data Model (EDM). The output, a populated data model, provided the PwC team robust grid data accessible via Awesense’s TGI tool and APIs.
The PwC team was empowered to develop their algorithms freely, having reliable access to cleansed and structured grid data. They achieved this by developing AI-driven correction algorithms that successfully aligned phases and meters to transformers within the territory subset. Furthermore, prior to working with FortistBC’s grid data, the PwC team was able to test out their use case using synthetic grid data on The Energy Transition Platform’s Sandbox Environment, which accelerated PwC’s work rapidly for Project Greenlight.
The PwC data science team ended the project equipped with AI-driven algorithms that improve the FortisBC meter-to-transformer connectivity model. These algorithms enhance reliable insight into FortistBS’s customer phase association, which helps the utility understand its distribution system with greater accuracy, assist in further system planning, and improve load growth studies.
