Predicting generation and consumption in real time of all grid assets with telemetering
With the electrification of society and new consumption patterns, distributed energy generation has become a reality. As the system's complexity grows, so does the importance of predicting the flow of energy across the grid while meeting supply safety and energy quality criteria.
The PREDIS project uses Big Data technologies to forecast, in real time, the consumption and generation of all grid assets, with load diagrams.
This E-REDES project has had the support of EDP Inovação, the academic universe, and external partners.
PREDIS is mounted on a Big Data cluster in Ermesinde.
With this final version, PREDIS can be sent as a module to the corporate BDA, thus ensuring a 24/7 service.
Existing Interfaces
- PREDIS receives information from IPMA (weather data) via FTP, and from SGT-ES (load diagrams) by checking BD;
- It is therefore dependent on the availability of information coming from these data sources;
- APIs have been developed to display the result of the calculations so that machines and/or users can access the information (predis.edp.pt/predis/índex.html);
- Due to data quality issues (communication failures, non-daily reading cycles, configuration errors, etc.), there are some limitations on accessing data from previously reported PTDs;
- For production forecasts, the mapping between SIT assets and SGT-ES load diagrams is yet to be validated.
How PREDIS Works
PREDIS' main goal is real-time consumption/production Distributed Forecasting.This project is important for EDP because it enables us to develop advanced analytics capabilities and implement Big Data technologies. Furthermore, the project is expected to have a direct impact on improving the energy balance by supporting dispatch and planning processes.
