Frequency Response from Inverter-Based Generation

Sammanfattning: The system inertia of the Nordic power system is expected decrease in the future. This is due to an expected reduction in synchronous generation from nuclear and thermal power. This will be compensated by an increase in power from inverter-based generation due to wind power’s expansion in the power system. In this thesis, a power system’s ability to handle a large under-frequency fault is studied and how this is affected by a lower system inertia. The thesis also studies how different types of frequency response from wind power plants affects the systems ability to handle the fault. The frequency of a modelled transmission system is studied through dynamic simulations in the software PSS/E. Changes are made to the system to replace nuclear power (and eventually also thermal power) with type 4 (full-sized converter) wind turbines. A large under-frequency event is induced and the frequency nadir and bus voltages are compared against system limits used by the Swedish TSO. While gradually adding wind power to the system as nuclear power is removed, different versions of the system with different amounts of wind power are compared. Two studies are carried out. In the first study, the voltage response study, nuclear power is decreased by scaling down the size of the generators. Besides reducing their power output and inertia, this also decreases their capacity for voltage support. To compensate for the reduced voltage support in the system, the wind power plants are made to provide voltage support. In the second study, the inertia study, the nuclear power in the system is instead decreased by reducing the active power output and the inertia constant of the nuclear generator. The size of the generator and reactive power output stay the same as in the original system so the nuclear power plant can continue to supply voltage support. In this way the problems caused by reduced inertia in the system are isolated. In both studies, three different frequency responses are applied to the wind power plants. The three frequency responses are aimed to as closely as possible follow the minimum requirements from the Swedish TSO for the frequency response products FCR-N, FCR-D and FFR alt A. In both studies, FCR-N provided by wind power causes a lower the system frequency during the fault compared to the original system without wind power. FCR-N can not fully compensate for the loss of inertia the replacement of synchronous generation is causing. In the voltage response study, FCR-D causes a slightly lower system frequency than the original system without wind. In the inertia study the frequency is slightly raised with wind power providing FCR-D. In both studies, FCR-D is or is very close to fully compensating for the loss of inertia. In both studies with FFR alt A, more wind power and therefor more FFR alt A in the system means a significantly improved frequency nadir. One conclusion from this project is that both FCR-D and FFR alt A from wind power are effective at counter-acting the effects reduced system inertia has on the system frequency stability in this system. For both FCR-D and FFR alt A, the frequency deviation during the fault is lower or close to equal to the frequency deviation when the original system (no wind power) is subjected to the same fault. The second conclusion is that in the studied system the reduced capacity of voltage support due to removal of nuclear power in the system is more influential than the reduced system inertia in the expansion of wind power. This indicates that there may need to be more focus from power producers, TSOs and legislators on voltage support in the future as the power system moves from large synchronous power plants to more distributed inverter-based production.