Probabilistic Power System Contingency Analysis Considering Wind

The worldwide demand for energy is steadily growing. Environmental pressure has driven the power generation shift from fossil fuel to more 'green' sources of supply, such as wind or solar. However, the intermittent nature of wind has also brought many uncertainties to network operation. Traditionally, to have a sufficiently secure operational margin, a transfer limit is calculated based on iterative simulations using the worst case scenario. The derived transfer limit normally is very conservative, which reduces the efficient use of the transmission network and curtails the use of renewable energy. Therefore, the traditional approach has many disadvantages when dealing with the developing electricity market. This thesis presents a fast screening probabilistic approach to help operators avoid possible problems with optimal use of transmission networks and renewable energy. Rather than using a long-term based worst case scenario in an invariant environment, we generalised the wind generation model based on historical studies to complete the analysis in a short time interval. The radar-like mechanism aims to sweep a power system for potential problems within 5 minutes with no time lag between the analysis and operation. The thesis also provides pathways for implementing this algorithm in different power system assessment methods, such as direct method and time domain simulation. This research has investigated the wind power fluctuation effect on power system stability in detail, from the linearised domain to the nonlinear domain, from the classic model to the most realistic model. It has demonstrated the robustness of this approach and its feasibility for implementation in a real system.