Background:

The advancement of new technologies enables the traditional power grid to move toward smart grids. The deployment of new technologies with high integration of renewable energy in smart grids as well as changes in production and consumption create additional planning and operational challenges. The demand-supply chain must be constantly maintained between the generation of power and also the demand for power from the customers to be able to maintain it with no interruptions. Different parts of the power system need to change in a more flexible way to cope with these new challenges and ensure a continued safe, secure and reliable operation. Increased risk of overload. For example, the increase in wind parks in the north of Germany, while shutting down the nuclear powerplants, has created an unbalance in the spatial distribution of the generation and consumption centres, which leads to an increment in the congestion level in the electrical network. However, the increase massive integration of renewables, and load demand may lead overhead transmission networks to overload, which leads to the failure of OHTLs. Consequently, it may cause a local outage that may propagate, thus causing large-scale blackout eventually.

Therefore, fast assessment of OHTL overload is useful in redispatch monitoring measures, including day-ahead scheduling, real-time operation, and long-term planning. The developments Machine Learning (ML) in modeling cascading failures in the grid may have significant improvement in the monitoring of smart grid operation, planning and control.

Tasks:

Literature Review
To develop machine learning algorithm.
To implement with dataset.

Required skills:

Student of electrical engineering, informatics or related .
Programming experience (Python, C/C++, Matlab).