english  | Home | Impressum | Datenschutz | KIT

Deep learning applied to spatial interpolation of weather data along simulated overhead lines

Deep learning applied to spatial interpolation of weather data along simulated overhead lines
Typ:Bachelor-/ Masterarbeit
Datum:offen (zu vergeben)
Betreuer:

 M. Eng. Gabriela Molinar

Deep learning applied to spatial interpolation of weather data along simulated overhead lines

.
.

Background

Weather conditions in the surroundings of the overhead lines and the transmitting power are the main factors for changes in the conductor temperature. This cannot be higher than a fixed maximum value, which determines a current capacity (or ampacity) for the conductor.
Therefore, a measurement of the atmospheric conditions along the line helps the Transmission System Operators (TSOs) to use efficiently the infrastructure (maximum possible transmitting power without overheating the conductor). It also allows to forecast the capacity of the network. This makes possible to plan ahead the power transmission and energy market.
One challenge when describing the weather conditions for overhead line monitoring is the interpolation of the punctual measured values along the overhead line route. This should be done considering the terrain where the overhead line goes through. 

 

Your tasks

  • To interpolate weather data along simulated overhead lines based on terrain information,
  • To evaluate the possibility to solve the problem using deep learning algorithms. 

 

We would need from you

  • Motivation and creativity!
  • To enjoy programming (Python knowledge is advantageous but not necessary)
  • Students of electrical, mechanical engineering, informatics, meteorology or similar are welcome!