Methods to ensure the adequate primary frequency response of low inertia power systems

primary frequency response;transient stability;governor response;interruptible load;demand response;frequency-constrained optimal power flow;frequency-constrained unit commitment;linear programming;under-frequency load shedding;droop control;integer programming

Bhana, Rajesh

关键词: primary frequency response;    transient stability;    governor response;    interruptible load;    demand response;    frequency-constrained optimal power flow;    frequency-constrained unit commitment;    linear programming;    under-frequency load shedding;    droop control;    integer programming;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/78342/BHANA-DISSERTATION-2015.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Power systems are changing and the trend toward renewable generation, lightweight gas turbines, and electronic load has resulted in systems with lower inertia and reduced governor response. The expected continuation of this decline motivates research on economical methods to ensure primary frequency response (PFR) adequacy and prevent frequency-related emergency actions. This dissertation describes both demand-side and generation-side methods for PFR improvement. Specifically, the commitment of autonomous interruptible load, the adjustment of governor gain settings, and the curtailment of resources are described. For these methods, the choice of the PFR resources is centrally coordinated, but the resource control actions are autonomous and based on local frequency measurements. Costs are attributed to the resources/control-decisions and minimum-cost optimization problems are formulated with nonlinear PFR constraints that require the time-domain simulation of a differential and algebraic equation (DAE) model of the system. Tractable iterative solution approaches, in which rapidly solvable linear approximations of the nonlinear problems are formulated, are proposed. For large systems, the burden of the iterative methods are further reduced through sensitivity-based estimation. This estimation exploits the near-linear power-sensitivity of the system, the similarity of electrically close buses, and the structure of a DAE power system model to accurately capture the temporal and spatial dependence of a PFR resource's contribution toward meeting a constraint.The computational benefits of the sensitivity-based estimates are demonstrated on systems with a large number of resources and the results show that the reductions in computational burden come with little increase in solution costs.