Comparing the suitability of non-linear and linearized power flow models in congestion management (WISE-Project IIT Bombay)



In the course of the German phase-out of nuclear and coal energy, large power plants in the German transmission grid will be disconnected from the grid until 2038. With the shutdown of large power plants, infeed will be shifted from higher to lower voltage levels. This is accompanied by the need for plants connected to the distribution grid to provide more and more grid services. Due to the volatility of generation from renewable energy plants and the fact that grid expansion cannot keep up with the speed of renewable expansion, many tasks arise for grid management. Renewable energy plants in the distribution grid must be involved in congestion management and redispatch in the most cost-effective and intelligent way possible.


Optimal power flow (OPF) can be used in congestion management to determine the redispatch of distributed generation. Since the considered network area can be very large and consist of many nodes and lines, simplified load flow models may be necessary to calculate the redispatch in real time in 15 minutes intervals. In this research project, the suitability of different power flow models and OPF solvers for the usage in congestion management should be investigated. It should be evaluated which type of congestions (current related, voltage related) can be detected and handled with the respective models, how accurate these models are and how fast the solutions can be found using typical non-commercial solvers. Finally, it will be evaluated which models are most suitable at which voltage level.