Research On Analysis Of Steady-State Voltage Security And Mode Adjustment Strategy Of Power System

Today,with the steady increase in power demand,the operating point of the system is approaching the stable boundary,and large-scale power outages caused by voltage collapse have occurred throughout the world.The voltage security and stability of the power grid are facing serious challenges,and the underlying voltage stability mechanism has become a research hotspot.At the same time,the national power grid is interconnected across regions on a large scale,and the proportion of renewable energy sources such as wind and light connected to the grid continues to increase.The voltage stability characteristics of the system become more complex.In this context,the study of static voltage analysis methods for adapting new power systems has important theoretical and engineering value.Currently,calculating static voltage stability margin is a mainstream method for evaluating static voltage stability in power systems.The core idea is to determine the distance between the operating point and the limit point in the direction of power change,and to measure the voltage stability of the system.However,in uncertain scenarios involving renewable energy,the direction and amplitude of power change are not fixed,and the stability margin obtained along a single power growth direction cannot fully reflect the system voltage stability level.Further,if the load continues to grow,the operating point will approach or even exceed the limit point,and the convergence of power flow calculation will be affected.Inconvergence in power flow calculation cannot be solved by calculating stability margin,which may even affect the calculation process of stability margin,making it difficult to evaluate voltage stability.To solve the above problems,this paper proposes a new static voltage security analysis method for power systems from the perspectives of static voltage stability boundary analysis and power flow calculation.Based on the physical process of the movement of the operating point from the inside of the stable boundary to the outside during load growth,this paper analyzes the main contradictions when the power flow is in a good state,ill state,and no solution,and provides solutions.The innovative results included in this study are as follows:(1)Aiming at the construction of static voltage stability boundary,a construction strategy based on the general formula of boundary high-order derivative and the asymptotic numerical method is proposed.Firstly,the characteristic equations of stable boundary points are listed based on the power flow equation.Using the quadratic characteristics of the power flow equation in the rectangular coordinate system,the general formula of the higher-order partial derivative of the boundary is obtained by taking multiple derivatives on both sides of the equation to eliminate the higher-order tensor.In order to adapt stable boundaries with different topological properties,a boundary construction method based on asymptotic numerical method is proposed.Carry out multiple high-order Taylor expansions on the boundary curve,control the accumulated error by increasing the expansion order,and then realize the accurate fitting of the boundary,and the mathematical expression of the error and approximate order is given.When the general formula proposed by the algorithm is used to calculate the higher order partial derivatives,the matrix factor table can be reused to reduce the calculation time.Based on the boundary fitting results,the possible application scenarios and margin indicators of the algorithm are given to meet the engineering application requirements.Finally,according to the IEEE-118 node system,a numerical example is analyzed.For the generator node and the load node,two-dimensional and three-dimensional stable boundaries are constructed respectively.Compared with the point-by-point method,the rapidity and accuracy of this method are verified.(2)In order to solve the problem that power flow is difficult to converge when it is ill conditioned,an alternating iterative power flow algorithm based on the decomposition of node voltage equation is studied.First,list the network node voltage equation sorted by node type,take the PQ node load equivalent as admittance,and use the WARD equivalent to eliminate the coupling term to obtain the network equation of PQ subnet and PV subnet.According to the characteristics of nodes.Gauss iteration method and Newton method are used to solve the power flow of PQ subnet and PV subnet respectively.The PQ node voltage mode and PV node phase angle are used as the coupling variables of the two subnets to transfer the power flow information,improve the calculation framework of the alternate iteration of the subnets,and then complete the calculation of the power flow of the original network.The proposed algorithm belongs to fixed-point iterative method,and is compared with Gauss-Sider method of the same kind in detail,which proves theoretically that the proposed method has better convergence performance.In order to improve the calculation speed of the algorithm,a vectorization calculation format using advanced extended instruction set is given,which can significantly improve the matrix operation speed.In order to verify the robustness of the algorithm,the algorithm is tested in a standard example system of different scales and compared with a variety of power flow algorithms.The simulation results show that the proposed algorithm has better convergence and stronger adaptability to the selection of initial iteration value.(3)In order to propose effective power flow adjustment measures and restore power flow solvability,an extended power flow calculation model based on power droop node and two-step analysis is proposed.Firstly,the node characteristic curve in the injected power space is analyzed,the definition of power droop node is given,and the role of power droop node in restoring the solvability of power flow is demonstrated from the physical and mathematical levels.An extended power flow calculation model considering power droop node is given.A simple example is given to verify the effectiveness of the proposed model in restoring the solvability of power flow.The non-solution of power flow is often caused by local load imbalance.It is necessary to introduce reasonable power droop node conversion logic to provide power support in key areas to restore the solvability of power flow.In order to determine the reasonable node conversion logic and identify the weak areas from the level of voltage and power angle,a two-step analysis method is proposed to locate the nodes that need type conversion.In engineering,the adjusted power flow is generally required to keep the load unchanged.In order to restore the boundary conditions,the state backtracking process is defined.Determine the power backtracking direction according to the boundary conditions,and carry out the state backtracking calculation to the stable boundary.Solve the left eigenvector of the Jacobian matrix of the boundary point,determine the action generator and determine the adjustment scheme according to the relationship between its components.Based on IEEE-118 bus system and TEXAS-2000 bus system,several non-solution scenarios are constructed,and compared with the optimal power flow method,the shortest distance method and other methods to verify the effectiveness and robustness of the proposed model for restoring the solvability of power flow.(4)In order to further consider the influence of DC system on the convergence of power flow,a feasible solution acquisition algorithm of AC/DC power flow for LCC-HVDC is studied.Firstly,the coupling relationship between AC and DC power grids is analyzed.The main reason for the poor convergence of AC and DC power flow is clearly defined as the calculation logic of alternating iteration of two networks,and a semi-decoupling method with PQV nodes is proposed.The voltage mode of the AC/DC tie node is set to a constant value,so that the power transmitted from the DC system to the AC system will not jump.In theory,only one AC power flow and one DC power flow need to be calculated.which avoids the original alternating iterative calculation logic and improves the convergence.To ensure the basic solvability of the equation,the idea of power droop is extended to the AC/DC power flow calculation of LCC-HVDC,and an extended power flow calculation model with PQV-tancp node pairs is proposed.Draw the characteristic curve of tanφ node in the injected power space,and demonstrate the role of tanφ node in improving power flow convergence by analyzing the relationship between the curve and the stable boundary manifold.Based on the one-to-one correspondence between PQV and tanφ-node,the tanφnode conversion logic based on the calculation of electrical distance index is given.The effectiveness of the proposed algorithm is tested in different scale systems.The simulation results show that the algorithm can effectively deal with the non-convergence problem of LCC-HVDC AC/DC power flow calculation,recover the power flow solvability and give feasible solutions.