An integrated approach incorporating dynamic and static security limits in optimum power dispatch
Optimum power dispatch is performed in a power system to determine the most economic power dispatch condition for a certain system loading. In this thesis the main focus is on the investigation of this problem and its improved application by including both dynamic security and static security limits in its solution. The aim is to develop an efficient and practical on-line method of optimum power dispatch with due regard to the necessary security requirements. A critical review of the current practices of security and optimization in power system operation shows that they are essential elements in Energy Management System computer softwares. Since optimality and security present conflicting requirements on system operation, it is both logical and beneficial to develop an integrated approach to satisfy all the security limits in optimum dispatch. The classical approach to consider only the static security limits in optimum dispatch calculation is found to be insufficient in providi ng the essential informations on the dynamic security performance. This problem is causing increasing concern with the recent trend to load power systems more closer to their stability limits in order to achieve maximum economy. A new formulation of the security constrained optimum dispatch problem with an integrated approach to consider both dynamic and static security limits is thus proposed in this thesis. The Optimum Power Flow ( OPF ) formulation uses a Recursive Quadratic Programming algorithm applied in the compact modelling of the system. This formulation consists of a decoupling process of the active and reactive power optimizations. The investigation into on-line security control shows that insufficient attention on dynamic security in present practice could endanger the system integrity in the contingency state. This leads to the development of a new scheme to integrate both dynamic and static security assessments. Direct application of classical transient stability assessment methods using numerical integration of swing equations is found to be too slow and a new method based on reduced dynamic equivalent is investigated. The method is based on an efficient dynamic security assessment scheme which assesses the on-line operating state of the system. A dynamic security margin is defined to measure the robustness of the system when it is subjected to a selected scenario of dynamic contingency. The method also identifies the critical machine or cluster of machines that would cause transient instability, and proposes preventive control strategies to improve the dynamic security performance. This is integrated in the approach as a preventive control module. The module aims to prevent the system from reaching probable system collapse due to contingency that could cause cascading tripouts in the system. Extensive simulation tests are performed using the approach in several example networks together with validation case studies compared to full load flow and transient stability tests. The results demonstrate that the approach is fast and reliable with good potential for on-line application in stability limited power systems.