Optimising power system frequency stability using virtual inertia from inverter-based renewable energy generation

[摘要] ENGLISH ABSTRACT: Inverter-based renewable energy generation are integrated into power systemsat an increasing rate. Governments continuously set higher goals for renewable energy generation considering only the impact on the environment, developmenttime, financial and economical side, while ignoring the impact of high penetration of inverter-based renewable energy generation on the current power system's stability. This thesis investigates the impact of increased variablerenewable energy generation integration into the power system, with a specific focus on inertial response for system frequency stability. The focus is on wind and solar power generation, which uses inverters to interface withthe power system network. These generation sources have a detrimental effect on the generation/load power balance, which reduces the system frequency stability. The power system becomes more sensitive, with increased RoCoF,lower frequency nadir and increased difficulty to control system frequency with generation/load balancing. Current mitigation measures and regulations for the decreased frequency stability are reviewed, indicating that for high share of renewable generation not just long-term energy storage is required, but also short-term energy storage with fast power response capabilities. To evaluatethe impact, power system components are examined and modeled to implement in a power system simulation. Power system operation and stability (rotor, voltage and frequency) are reviewed and discussed. The research thenfocuses on transient frequency dynamics and stability. For the mitigation of reduced frequency stability the concept of virtual inertia is introduced. Virtual inertia is then explained and simulated for wind and solar PV plants.Lastly the H2 -norm metric is used to evaluate power system frequency stability, rather than using the amount of inertia present in a power system. The metric follows from Lyapunov theory for analyzing non-linear system stabilitythrough energy functions. The distribution of virtual inertia in a network is then optimised using the Genetic Algorithm with the H2 -norm, which is used to analyze the system robustness against disturbances, as the cost-function.The results show significant performance improvement in transient stability for the Western Transmission network of the Eskom power system in South Africa.

[发布机构] Stellenbosch University