Optimal Reactive Power Reserve Dispatch Method Considering Uncertainties

The reactive power reserves(RPRs)are sensitive to both active and reactive load changes,and the correlation between the RPR and the voltage stability makes the RPR as an important security index for the power system.However,the RPRs with uncertainties are beyond the current research focus.In this dissertation,models of uncertainties,such as branch outages,generation uncertainties of renewable energy and normal energy,loads' uncertainties,are established to formulate the framework of RPR optimization considering uncertainties.The main results are as follow.(1)The models of various uncertainties emerging in power systems are established.Among above four types of uncertainties,the branch outages are 0-1 types,which are assessed by the static and the transient severity selection.Generation uncertainties of renewable energy and normal energy,loads' uncertainties are assumed continuous,which are modelled as the probabilistic distribution or the hyper-pyramid model.(2)The model of stochastic RPR optimization considering operating point uncertainties is established.At first,a reactive partitioning method based on the modularity index and the agglomeraive clustering is proposed.Then a convex relaxation method is proposed to transform the chance constraints.At last,a solution based on a sequential linear optimization is proposed.(3)The model of robust RPR optimization considering load increment direction(LID)uncertainties is established.At first,a calculating approach for the closest point to voltage stability limit surface is p roposed based on the second-order approximation of surfaces.Then an alternative iteration method is proposed to solve the above problem.(4)The model of interval RPR optimization considering LID uncertainties is established.At first,a provement for the operating risk reduction is given.Then a two-stage alternative iteration solution is proposed to solve the above problem.(5)The model of a single-time intersection transient RPR optimization is proposed based on the trajectory sensitities.This model gives different weights to the reactive power sources.Moreover,a linearized objective function method is proposed for the solution.(6)The model of a multiple-time intersection transient RPR optimization is proposed.Then the numerical differentiation is utilized to transform the original model to a dynamic optimization.The time-varied reactive power reserve weights are therefore calculated accurately.