Short description
(English)
|
The project aims at improving the detection and prediction of critical conditions in railway power network systems and components by developing intelligent algorithms for Control System Data. In particular, the project aims at developing methodology for integrating the spatial-temporal information and underlying physical laws for detecting and predicting critical conditions that may cause significant disruptions. The developed methodology will also enable to prolong the maintenance intervals and extend the lifecycles of power system components, such as hydropower plants and reduce the maintenance costs. This pro-ject will, therefore, safe costs, improve security of supply and increase stability of grid operation.
Related documents
|
Project aims
(English)
|
The main goal of the project is to improve the detection and prediction of critical conditions in railway power network systems. To enable this, we will develop methodology for the detection and prediction of critical events in railway power networks and components (based on central dispatch big data archive):
1)Integrate spatial and temporal monitoring data from the railway power networks
2)Extend the developed methodology by integrating power system dynamics in deep learning algorithms
As an additional goal, erroneous measurements and sensor drifts will be automatically detected, improv-ing thereby the data quality significantly and making the decisions based on those measurements more reliable. Here, the manual efforts of data administration and quality assurance (ca. 15 FTE, which an in-ternal study showed) can be significantly reduced.
|
