Research On Optimal Power Flow Optimization Dimension Reduction Method Considering Transient Voltage Stability Constraints

In recent years,with the increasing complexity and scale of the power grid,transient voltage safety issues threaten the safe operation of the power grid,and it is urgent to take appropriate voltage control measures to prevent it.In the face of large power grids,the characteristics of large power grids,tight coupling,numerous fault sets,and complex system structures have led to high computational complexity for optimal power flow(TSCOPF)considering transient safety constraints.Facing the TSCOPF problem of largescale and complex power grids,the existing methods still have the disadvantages of low model accuracy and slow solution speed.There is an urgent need to improve the speed and accuracy of online optimization calculations for current large power grids.This paper is funded by the "National Natural Science Foundation of China(52007017)",focusing on representative severe contingencies filtering methods,key bus identification methods,a optimization reduced-order equivalent methods based on dynamic equivalence,and a optimization reduced-order equivalent methods based on transient partitions,etc.The aim is to reduce the size of optimization problem and accelerate the solving speed.The main research contents are as follows:In order to reduce the scale of optimization problems,representative serious contingencies are selected from the preset contingency set to replace the original complete contingency set.For this reason,a two-stage representative serious contingencies filtering method with multiple predictive contingency sets is established.First,considering the effects of voltage sag,voltage fluctuation and voltage recovery level,a transient voltage stability index is proposed.Then the transient voltage stability index matrix is established to describe the characteristics of the system’s fault location,fault form and fault duration.Finally,a two-stage filtering approach for representative serious contingencies is proposed.In the first stage,serious contingencies are obtained through index thresholds.In the second stage,the condensed hierarchical clustering algorithm is used to classify the serious contingencies,and the representative serious contingencies are obtained by combining the two selection principles of the representative serious contingencies.Case studies of the IEEE 39 bus system and 57 bus system show that the number of contingencies can be greatly reduced through the filtering of representative serious contingencies,the scale of optimization problems can be reduced,and the original preset contingency set can be effectively represented.In order to reduce the computational time of transient safety constrained optimal power flow(TSCOPF)optimization problem,a key bus identification method and an optimal dimensionality reduction method based on dynamic equivalence are proposed.Firstly,a key bus identification method is proposed from the perspective of eliminating redundant constraints.Then using the idea of dynamic equivalence and the method of trajectory sensitivity analysis,the generator sets that have little influence on the system are selected,and the equivalence group is selected according to the electrical distance between the generator sets.Taking the difference of transient voltage stability index to reach the iterative threshold is the objective function,and the dynamic characteristics of the system before and after the dynamic equivalence are required to be consistent,and the equivalence model of the generator equipment is established.Finally,based on the representative severe contingency set selected in Chapter 2,a optimization reduced-order equivalent model based on dynamic equivalence is established,and the sensitivity model is used to replace the original nonlinear optimization model.Calculate the calculation time of the TSCOPF problem before and after dynamic equivalence under different number of contingencies.Case studies of the IEEE 39 bus system and 57 bus system show that the optimization problem can reduce the difficulty of solving the optimization problem and accelerate the solving speed while ensuring the accuracy of the system after key bus identification and dynamic equivalence.In order to further reduce the calculation time of the TSCOPF optimization problem,an optimization dimension reduction method based on dynamic partition is proposed in this paper to solve the problems of security constraints,too many optimization variables and strong degree of nonlinearity encountered by the existing optimization methods when dealing with large-scale complex power grids.First,based on the representative serious contingencies selected in Chapter 2,the system is initially partitioned,the dominant node is selected,and then partitioned.A two-stage system partition method based on transient voltage characteristics is proposed.Then the node splitting method is used to establish the physical model of the grid partition.Finally,based on the aforementioned dynamic partition results,an optimal decoupling model of the multi-region transient safetyconstrained optimal power flow is established,and the sensitivity model is used to replace the original nonlinear optimization model,and the time for each partition TSCOPF problem is calculated.Case studies of the IEEE 39 bus system and 57 bus system show that the scale of the optimization problem is greatly reduced and the solving speed is greatly improved after the dynamic partitioning of the TSCOPF optimization problem.