Date of Award
Master of Science
Electrical and Computer Engineering
Power systems over the recent past few years, has undergone dramatic revolution in terms of government and private investment in various areas such as renewable generation, incorporation of smart grid to better control and operate the power grid, large scale energy storage, and fast responding reactive power sources. The ongoing growth of the electric power industry is mainly because of the deregulation of the industry and regulatory compliance which each participant of the electric power system has to comply with during planning and operational phase.
Post worldwide blackouts, especially the year 2003 blackout in north-east USA, which impacted roughly 50 million people, more attention has been given to reactive power planning. At present, there is steady load growth but not enough transmission capacity to carry power to load centers. There is less transmission expansion due to high investment cost, difficulty in getting environmental clearance, and less lucrative cost recovery structure. Moreover, conventional generators close to load centers are aging or closing operation as they cannot comply with the new environmental protection agency (EPA) policies such as Cross-State Air Pollution Rule (CSAPR) and MACT. The conventional generators are getting replaced with far away renewable sources of energy. Thus, the traditional source of dynamic reactive power support close to load centers is getting retired. This has resulted in more frequently overloading of transmission network than before. These issues lead to poor power quality and power system instability. The problem gets even worse during contingencies and especially at high load levels.
There is a clear need of power system static and dynamic monitoring. This can help planners and operators to clearly identify severe contingencies causing voltage acceptability problem and system instability. Also, it becomes imperative to find which buses and how much are they impacted by a severe contingency. Thus, sufficient static and dynamic reactive power resource is needed to ensure reliable operation of power system, during stressed conditions and contingencies. In this dissertation, a generic framework has been developed for filtering and ranking of severe contingency. Additionally, vulnerable buses are identified and ranked.
The next task after filtering out severe contingencies is to ensure static and dynamic security of the system against them. To ensure system robustness against severe contingencies optimal location and amount of VAR support required needs to be found. Thus, optimal VAR allocation needs to be found which can ensure acceptable voltage performance against all severe contingency. The consideration of contingency in the optimization process leads to security constrained VAR allocation problem. The problem of static VAR allocation requirement is formulated as minlp. To determine optimal dynamic VAR installation requirement the problem is solved in dynamic framework and is formulated as a Mixed Integer Dynamic Optimization (MIDO).
Solving the VAR allocation problem for a set of severe contingencies is a very complex problem. Thus an approach is developed in this work which reduces the overall complexity of the problem while ensuring an acceptable optimal solution. The VAR allocation optimization problem has two subparts i.e. interger part and nonlinear part. The integer part of the problem is solved by branch and bound (B&B) method. To enhance the efficiency of B&B, system based knowledge is used to customize the B&B search process. Further to reduce the complexity of B&B method, only selected candidate locations are used instead of all plausible locations in the network. The candidate locations are selected based upon the effectiveness of the location in improving the system voltage.
The selected candidate locations are used during the optimization process. The optimization process is divided into two parts: static optimization and dynamic optimization. Separating the overall optimization process into two sub-parts is much more realistic and corresponds to industry practice. Immediately after the occurrence of the contingency, the system goes into transient (or dynamic) phase, which can extend from few milliseconds to a minute. During the transient phase fast acting controllers are used to restore the system. Once the transients die out, the system attains steady state which can extend for hours with the help of slow static controllers.
Static optimization is used to ensure acceptable system voltage and system security during steady state. The optimal reactive power allocation as determined via static optimization is a valuable information. It's valuable as during the steady state phase of the system which is a much longer phase (extending in hours), the amount of constant reactive power support needed to maintain steady system voltage is determined. The optimal locations determined during the static optimization are given preference in the dynamic optimization phase.
In dynamic optimization optimal location and amount of dynamic reactive power support is determined which can ensure acceptable transient performance and security of the system. To capture the true dynamic behavior of the system, dynamic model of system components such as generator, exciter, load and reactive power source is used. The approach developed in this work can optimally allocate dynamic VAR sources.
The results of this work show the effectiveness of the developed reactive power planning tool. The proposed methodology optimally allocates static and dynamic VAR sources that ensure post-contingency acceptable power quality and security of the system. The problem becomes manageable as the developed approach reduces the overall complexity of the optimization problem. We envision that the developed method will provide system planners a useful tool for optimal planning of static and dynamic reactive power support that can ensure system acceptable voltage performance and security.
Tiwari, Ashutosh, "Optimal allocation of static and dynamic reactive power support for enhancing power system security" (2013). Graduate Theses and Dissertations. 13504.