Implement a filtering method for the ANDES model (long term beast prediction model, 1 to 40 years) on the measurements of more than 500 substations. This model is used for customer integration decisions and investment plans and choices.
Improved STORM algorithm
AI & Data Science
- Duration
- 2020
- Project member(s)
- R.C. Bouman (Roel) , Dr Y. Shapovalova (Yuliya) , Jacco Heres
- Project type
- Research
Results
- Before implementation, substation data was filtered manually, which took about 6 weeks of man-hours per year. Now 80 percent of the data is filtered automatically. This means only 2 to 3 weeks of filtering time, with better quality filtering on top of that because the network analysts have more time.
- Alliander can rely more on the measurements. That gives confidence to create new types of contracts and make decisions with greater impact.
- You can read more about this project and its results here: Data scientists help find space on crowded power grid
Partners
Alliander
