CN110137967B - A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes - Google Patents

A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes Download PDF

Info

Publication number
CN110137967B
CN110137967B CN201910388541.9A CN201910388541A CN110137967B CN 110137967 B CN110137967 B CN 110137967B CN 201910388541 A CN201910388541 A CN 201910388541A CN 110137967 B CN110137967 B CN 110137967B
Authority
CN
China
Prior art keywords
data
power flow
convergence
power
key
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910388541.9A
Other languages
Chinese (zh)
Other versions
CN110137967A (en
Inventor
安军
宋俊达
周毅博
葛维春
周东皓
乔雪婧
邓子晗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
State Grid Liaoning Electric Power Co Ltd
Northeast Electric Power University
Original Assignee
Northeast Dianli University
State Grid Liaoning Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northeast Dianli University, State Grid Liaoning Electric Power Co Ltd filed Critical Northeast Dianli University
Priority to CN201910388541.9A priority Critical patent/CN110137967B/en
Publication of CN110137967A publication Critical patent/CN110137967A/en
Application granted granted Critical
Publication of CN110137967B publication Critical patent/CN110137967B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for AC mains or AC distribution networks
    • H02J3/04—Circuit arrangements for AC mains or AC distribution networks for connecting networks of the same frequency but supplied from different sources
    • H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

本发明是一种针对关键节点的大规模电力系统潮流收敛性调整方法，其特点是通过牛顿迭代法(Newton‑Raphson)迭代的中间数据，建立潮流收敛性的判断指标；根据指标值所代表的具体意义辨识关键影响节点和因素；根据关键影响节点和因素刻画潮流收敛域；依赖不同调整目标对关键数据量进行调整等内容。具有科学合理，适用性强，效果佳等优点，能够为电网潮流调整提供直观依据。

Figure 201910388541

The present invention is a method for adjusting the power flow convergence of a large-scale power system for key nodes. The specific meaning is to identify the key influencing nodes and factors; to describe the power flow convergence region according to the key influencing nodes and factors; to adjust the key data volume depending on different adjustment targets. It has the advantages of scientific rationality, strong applicability, and good effect, and can provide an intuitive basis for power grid flow adjustment.

Figure 201910388541

Description

Large-scale power system power flow convergence adjusting method for key nodes
Technical Field
The invention relates to the field of power flow analysis of a power system, in particular to a method for judging power flow convergence by using iteration intermediate information, identifying key nodes, defining a power flow convergence domain and adjusting key node data by using the power flow convergence domain.
Background
The tidal current calculation has important significance on planning design and optimized operation of the power grid. For a large-scale power system with heavy load, the situation that the power flow calculation is not converged easily occurs. The factors causing the non-convergence of the power flow are numerous, the factors include algorithms, model parameters, injection data and the like, and since the Newton method is widely applied when solving the power flow problem and the power grid parameters are corrected for many times, the irrational injection data is generally considered as an important reason for the non-convergence of the power flow in engineering. For large-scale power system analysis, how to locate key data influencing the power flow convergence from a plurality of injected data and provide an adjusting strategy to improve the power flow convergence has practical value. The existing adjusting method for the power flow convergence mainly emphasizes that the interval for power flow convergence is expanded by using an algorithm improvement mode, so that a series of power flow calculation methods such as a planning method, a homotopic method, a Levenberg-Marquard method and the like are formed. However, the adjustment method proposed by the existing research emphasizes the compensation of the power generation node on the power of the whole grid, and a direct identification and adjustment strategy aiming at key injection data is lacked.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method for adjusting the power flow convergence of the large-scale power system aims at the key nodes, overcomes the defects that the existing integral adjustment method cannot be positioned and the specific influence nodes and factors are affected, emphasizes the identification and adjustment of the key nodes causing the power flow non-convergence, and is scientific, reasonable, high in applicability and good in effect and capable of providing visual basis for power grid power flow adjustment.
The technical scheme for solving the technical problem is as follows: a method for adjusting the power flow convergence of a large-scale power system aiming at key nodes is characterized by comprising the following steps: establishing a judgment index of the power flow convergence through intermediate data of Newton-Raphson iteration; identifying key influence nodes and factors according to the specific meaning represented by the index value; depicting a trend convergence domain according to key influence nodes and factors; adjusting the key data volume depending on different adjustment targets; the concrete contents are as follows:
1) establishing a judgment index of the power flow convergence through intermediate data of power flow iteration of a Newton iteration method (Newton-Raphson): the intermediate process data of the power flow iteration contains important information of a power grid, and viewed in isolation, the solving result of each power flow iteration can represent an independent power flow form, namely the electrical quantity relation of each node meets the physical rule, the redundant power is borne by balance nodes, and repeated iteration is needed only because the error between the calculating result and the injected numerical value cannot meet the precision requirement, so that the change condition of the data between two iterations can reflect the convergence capacity of a power flow equation continuously, if the calculating data gradually approaches to the injected data in the iteration process, the iteration process is shown to be developing towards the convergence direction, the probability of power flow convergence is higher, and in the calculation process of the Newton iteration method, | Deltay(k)L is used as the unbalance amount of each node power, the change condition of the error between the calculated value of each node power and the injected data in the iteration process is reflected, and therefore the index is established:
Figure BDA0002055652070000021
the power flow convergence condition is as follows:
Figure BDA0002055652070000022
in the formula
Figure BDA0002055652070000023
Judging indexes for key factors; k represents the number of iterations; k is a radical ofmaxRepresenting a preset maximum iteration number; Δ y represents the node power imbalance; mu represents preset iterative solution precision;
2) and identifying key influence nodes and factors according to the specific meaning represented by the index value: by means of Δ y(k)Reflecting the error between the calculated power value and the injected data, providing a method for identifying the key factors of load flow non-convergence, Newton iterationIn the process of normal load flow calculation, delta x(k)And Δ y(k)The accuracy of the tidal current solution can be reflected, but the significance of the tidal current solution and the significance of the tidal current solution are different, and delta x(k)Representing the difference between two iterations of a state variable, when Δ x(k)When the change condition of the state variable gradually tends to be stable, the change condition of the state variable can be reflected to gradually tend to be stable; and Δ y(k)Represents the error between the calculated value and the true value, which is defined by | Δ y for a system of equations containing more than two arguments(k)The maximum value in | is determined when | Δ y(k)When | gradually tends to decrease, it indicates that in the iterative process, the iterative computation quantity is approaching towards the direction of injecting data gradually, and the comparison Δ x(k)In other words,. DELTA.y(k)Has more definite practical significance, EmaxThe larger the number of iterations, the larger the error between the calculated value and the injected data, and conversely, EmaxSmaller indicates that the calculated value is closer to the injected data in that iteration;
3) and (3) describing a trend convergence domain according to key influence nodes and factors: the load flow convergence domain refers to a set of operating point data meeting load flow calculation convergence conditions, the load flow convergence domain is described by taking load flow calculation convergence as boundary conditions according to identified load flow non-convergence key factors, a data correction scheme is provided by taking the position relation of the data points relative to the boundary of the convergence domain, and when the data points are positioned in the convergence domain, the load flow calculation convergence is realized; when the data points are located outside the convergence domain, the load flow calculation cannot be converged, theoretically, the convergence domain can be drawn to an infinite dimension, but the convergence domain higher than the two-dimensional convergence domain is not beneficial to analyzing subsequent load flow correction, so that more than two identified problem factors can be drawn, and a plurality of two-dimensional domains can be respectively analyzed;
4) the key data volume is adjusted depending on different adjustment targets: when the load flow calculation is not converged, namely the data point is positioned outside a convergence domain, correcting data point data, and pulling the data point into the convergence domain, namely the aim of adjusting the non-converged load flow to be converged through correcting key control variables is achieved, a data correction scheme taking single data to be corrected as a target and a data correction scheme taking the total amount of the corrected data as the minimum are adopted, and the core of the data correction scheme taking single data to be corrected is that the shortest distance of the data point to the convergence boundary in a direction parallel to a coordinate axis is obtained, but when the data deviation is more and the deviation amount is larger, the adjustment amount is more; when a data correction scheme aiming at the minimum total corrected data amount is adopted, a boundary point can be found theoretically so that load flow calculation is converged, but in the actual operation process, the distance from a data point to a convergence boundary is not easy to obtain, the two targets have respective advantages and disadvantages, and the two targets are correspondingly combined and used according to specific targets in actual application.
According to the method for directly positioning and adjusting the key problem of the non-convergence of the power flow caused by the direct positioning and adjustment, due to the fact that Newton-Raphson iteration calculation is utilized, the change rule of intermediate data of power flow iteration is analyzed, the change condition of a control variable in the iteration process is used as a diagnosis standard of the key data of the non-convergence of the power flow caused by the power flow calculation, the key data of the non-convergence of the power flow caused by the comparison of the error relation between calculated amount and injected data in the power flow iteration is identified, and an index for judging the convergence of the power flow calculation is established; and (3) providing a power flow convergence domain, introducing the power flow convergence domain into evaluation and correction of power flow data, solving the maximum load capable of converging the power flow in the current operation state by taking power flow calculation convergence as a boundary condition, obtaining a power flow convergence domain boundary, and correcting the key data by means of the position relationship between the data point and the convergence domain boundary. Has the advantages of scientific and reasonable structure, strong applicability, good effect and the like. The method can provide an intuitive reference basis for power grid operators, and the node data can be directly adjusted.
Drawings
FIG. 1 is a schematic diagram of an iterative process;
FIG. 2 is a schematic diagram of the objective of correcting individual data;
FIG. 3 is a schematic diagram of the objective of minimizing the total amount of correction data;
fig. 4 is a schematic diagram of an active power correction amount Δ P of each node at the 9 th iteration;
FIG. 5 is P15-P12And (4) a convergence domain schematic diagram.
Detailed Description
The method for adjusting the power flow convergence of the large-scale power system for the key nodes according to the present invention is further described with reference to the accompanying drawings and embodiments.
A method for adjusting the power flow convergence of a large-scale power system aiming at key nodes is characterized by comprising the following steps: establishing a judgment index of the power flow convergence through intermediate data of Newton-Raphson iteration; identifying key influence nodes and factors according to the specific meaning represented by the index value; depicting a trend convergence domain according to key influence nodes and factors; adjusting the key data volume depending on different adjustment targets; the concrete contents are as follows:
1) establishing a judgment index of the power flow convergence through intermediate data of power flow iteration of a Newton iteration method (Newton-Raphson): the intermediate process data of the power flow iteration contains important information of a power grid, and viewed in isolation, the solving result of each power flow iteration can represent an independent power flow form, namely the electrical quantity relation of each node meets the physical rule, the redundant power is borne by balance nodes, and repeated iteration is needed only because the error between the calculating result and the injected numerical value cannot meet the precision requirement, so that the change condition of the data between two iterations can reflect the convergence capacity of a power flow equation continuously, if the calculating data gradually approaches to the injected data in the iteration process, the iteration process is shown to be developing towards the convergence direction, the probability of power flow convergence is higher, and in the calculation process of the Newton iteration method, | Deltay(k)L is used as the unbalance amount of each node power, the change condition of the error between the calculated value of each node power and the injected data in the iteration process is reflected, and therefore the index is established:
Figure BDA0002055652070000031
the power flow convergence condition is as follows:
Figure BDA0002055652070000041
in the formula
Figure BDA0002055652070000042
Judging indexes for key factors; k represents the number of iterations; k is a radical ofmaxRepresenting a preset maximum iteration number; Δ y represents the node power imbalance; mu represents preset iterative solution precision;
2) and identifying key influence nodes and factors according to the specific meaning represented by the index value: by means of Δ y(k)Reflecting the characteristic of the error between the power calculation value and the injected data, and providing a method for identifying the key factors of load flow non-convergence(k)And Δ y(k)The accuracy of the tidal current solution can be reflected, but the significance of the tidal current solution and the significance of the tidal current solution are different, and delta x(k)Representing the difference between two iterations of a state variable, when Δ x(k)When the change condition of the state variable gradually tends to be stable, the change condition of the state variable can be reflected to gradually tend to be stable; and Δ y(k)Represents the error between the calculated value and the true value, which is defined by | Δ y for a system of equations containing more than two arguments(k)The maximum value in | is determined when | Δ y(k)When | gradually tends to decrease, it indicates that in the iterative process, the iterative computation quantity is approaching towards the direction of injecting data gradually, and the comparison Δ x(k)In other words,. DELTA.y(k)Has more definite practical significance, EmaxThe larger the number of iterations, the larger the error between the calculated value and the injected data, and conversely, EmaxSmaller indicates that the calculated value is closer to the injected data in that iteration;
3) and (3) describing a trend convergence domain according to key influence nodes and factors: the load flow convergence domain refers to a set of operating point data meeting load flow calculation convergence conditions, the load flow convergence domain is described by taking load flow calculation convergence as boundary conditions according to identified load flow non-convergence key factors, a data correction scheme is provided by taking the position relation of the data points relative to the boundary of the convergence domain, and when the data points are positioned in the convergence domain, the load flow calculation convergence is realized; when the data points are located outside the convergence domain, the load flow calculation cannot be converged, theoretically, the convergence domain can be drawn to an infinite dimension, but the convergence domain higher than the two-dimensional convergence domain is not beneficial to analyzing subsequent load flow correction, so that more than two identified problem factors can be drawn, and a plurality of two-dimensional domains can be respectively analyzed;
4) the key data volume is adjusted depending on different adjustment targets: when the load flow calculation is not converged, namely the data point is positioned outside a convergence domain, correcting data point data, and pulling the data point into the convergence domain, namely the aim of adjusting the non-converged load flow to be converged through correcting key control variables is achieved, a data correction scheme taking single data to be corrected as a target and a data correction scheme taking the total amount of the corrected data as the minimum are adopted, and the core of the data correction scheme taking single data to be corrected is that the shortest distance of the data point to the convergence boundary in a direction parallel to a coordinate axis is obtained, but when the data deviation is more and the deviation amount is larger, the adjustment amount is more; when a data correction scheme aiming at the minimum total corrected data amount is adopted, a boundary point can be found theoretically so that load flow calculation is converged, but in the actual operation process, the distance from a data point to a convergence boundary is not easy to obtain, the two targets have respective advantages and disadvantages, and the two targets are correspondingly combined and used according to specific targets in actual application.
According to the power flow iteration process shown in fig. 1, | Δ y for the converged power flow(k)Each quantity in | should gradually tend to decrease in the iteration process, and | Δ y when the flow iteration is difficult to converge(k)If the l is oscillating or even suddenly changing, the accuracy requirement cannot be met all the time. Taking | Δ y of each iteration in the calculation process(k)The maximum absolute value of | is recorded as Emax. When E ismaxWhen the power flow gradually tends to decrease in the iterative process, the power flow calculation is developed towards the convergence direction, and when E ismaxAnd finally, the tidal current calculation is judged to be convergent when the tidal current calculation is reduced to meet the precision requirement. If EmaxThe oscillation occurs and can not be reduced all the time under the specified iteration times until the accuracy requirement is met, and the load flow calculation is not converged.
II, taking the calculation process EmaxThe minimum iteration whose calculated value is closest to the injected data, if E is the timemaxRepresenting the active deviation value, the active data of the system is considered to have problems and reactive dataData is not changed; if at this time EmaxAnd representing the reactive deviation value, considering that the reactive data of the system have problems, and not changing the active power data. By observing this time Δ y(k)Which are larger indicate that when the calculated values are closest to the injected data, these quantities remain difficult to converge, identifying key nodes and factors that lead to non-convergence of the power flow.
Thirdly, supposing that the identified key factor is the active power P of the ith and the j nodesi、PjThen with PiIs a horizontal axis, PjAnd (3) taking the vertical axis and the power flow convergence as boundary conditions, and making a power flow convergence domain:
(1)Pjsetting to zero, P in fixed stepiIncreasing from zero to no convergence of power flow to obtain PjIs zero time PiMaximum value of (P)imaxThis P isimaxThe maximum value of the convergence domain on the horizontal axis is obtained;
(2) in the interval [0, Pimax]In which n nodes P are inserted equidistantlyi (k)(where k is 0,1,2, … …, n), each P is sought using the dichotomyi (k)P for causing flow non-convergencejMaximum value of
Figure BDA0002055652070000051
Thereby obtaining boundary points of n power flow convergence domains;
(3) and fitting a curve synthesized by the boundary points to form a power flow convergence boundary, wherein a region enclosed by the boundary and the coordinate axes is a power flow convergence region, and a region outside the convergence region is called a non-convergence region.
Fourthly, when the single data is corrected as the target: the position of the intersection point of the convergence boundary and the coordinate axis is perpendicular to the coordinate axis to form a perpendicular line, so that the non-convergence region can be divided into four regions, i, ii, iii and iv, as shown in fig. 2.
(1) When the data point is in region i, data point 1 is taken as an example. Since the data points do not exceed the maximum value of the convergence boundary, the power flow can be converged by correcting any one of the abscissa and ordinate axes alone. And comparing the transverse distance and the longitudinal distance of the data point relative to the convergence boundary, and taking the shortest distance as a measure for flow correction under the data point.
(2) Data point 2 is taken as an example when the data point is in region ii. Since the data point at this time exceeds the maximum value P of the convergence boundary on the vertical axisjmaxAt this time, if P is corrected aloneiThe data has failed to reconverge the power flow, so P will bejAs a correction amount. The longitudinal distance from the data point to the convergence boundary is taken as P at this timejThe correction amount of (1).
(3) Data point 3 is taken as an example when the data point is in region iii. Since the data point at this time exceeds the maximum value of the convergence boundary on the horizontal axis, P is corrected if onlyjThe data has failed to reconverge the power flow, so P will beiAs a correction amount. The longitudinal distance from the data point to the convergence boundary is taken as P at this timeiThe correction amount of (1).
When the minimum total amount of correction data is targeted: the minimum total amount of correction data, i.e., the shortest distance between the data point and the convergence boundary, is shown in fig. 3. When the minimum corrected data total amount is taken as a target, a vertical line from a data point to a convergence boundary is made through the data point, the boundary is intersected at a point A, a connecting line AM between the data point and the intersection point is the shortest distance from the data point to the convergence boundary, the AM is decomposed into a horizontal component and a vertical component, and the horizontal component OM is PiThe longitudinal component OA is PjThe correction amount of (1).
The feasibility of the above protocol was verified in conjunction with specific tests, described in detail below:
a power grid simulation model with 220kV and above voltage class in a certain province in China is adopted, and basic information of a power grid is as follows: the total number of the nodes is 177, wherein 17 nodes of 500kV, 126 nodes of 220kV, 34 intermediate nodes of the transformer, and 301 branches of the transmission line and the transformer winding are provided. And the load of the whole network is expanded by 3 times, so that the load flow simulation calculation is not converged.
With successive iteration of the power flow, after the delta y is iterated for four times, the larger value is gradually concentrated in a plurality of nodes adjacent to the 15 nodes, the correction quantity of other nodes is obviously smaller, the inverse delta x does not show obvious mutation characteristics in each iteration, the delta x is used as an indirect quantity, the delta y needs to be calculated by means of a Jacobian matrix, and a relation is established between the delta y and injected data, so that the method for calculating the key data which is not converged by the power flow and is judged according to the delta y has certain advantages.
E at each iterationmaxSee Table 1, see E at the ninth iterationmaxAt minimum, E at this timemaxRepresenting the 15 th node active power correction quantity delta P15When the calculated quantity of the power flow iteration is closest to the injected data, the 15 th node has the largest active correction quantity. The ninth iteration full active power correction case is shown in fig. 4. The first three positions of the correction quantity value from large to small are respectively as follows: delta P15、ΔP12、ΔP10And several nodes with large correction deviation form a ring network on the topological connection, and the 15 th node in the ring network has heavy active load, which causes bad influence on the ring network tide.
TABLE 1 cases at each iteration
Figure BDA0002055652070000061
And describing a power flow convergence domain by using the two nodes with the maximum active power correction amount in the ninth iteration, as shown in fig. 5. The current data point position has not crossed the maximum horizontal and vertical axes of the convergence region, i.e., is within region i shown in fig. 2. If the current power flow is adjusted by selecting and modifying single data, the delta P is obtained15The minimum adjustment is 0.511 p.u.]The minimum adjustment amount is 0.1205[ p.u ].]Since the adjustment amount of the 12 th node is small, the 12 th node active power is selected for adjustment, and the adjustment amount 0.1205[ p.u ].]。
And (4) revising the load flow data, and performing load flow calculation again. The adjusted power flow calculation reaches convergence after 12 iterations, and the iteration error before and after adjustment is shown in table 2 below.
TABLE 2 iterative errors before and after tidal current adjustment
Figure BDA0002055652070000071
Through the analysis, the invention provides a method for judging the key data causing the non-convergence of the power flow calculation, and the method establishes the judgment index of the convergence of the power flow calculation based on the change characteristics of the power flow iteration intermediate data of a Newton-Raphson (Newton-Raphson), thereby directly positioning the key input data causing the non-convergence of the power flow calculation and providing a direct basis for the power flow adjustment. In addition, by means of the domain idea, a power flow convergence domain is defined, and data correction under the corresponding target is carried out on the pathological power flow by comparing the relative position of the current data point to the boundary of the convergence domain, so that the power flow calculation can be converged again. The method emphasizes the evaluation of the influence of the injected data on the power flow convergence, and solves the problem that the dominant influence node is difficult to identify when the power flow is not converged.
The above embodiments are only intended to illustrate the present invention, but not to limit it, and it should be understood by those skilled in the art that any modifications and equivalent changes made with reference to the above embodiments are within the scope of the claims of the present invention.

Claims (1)

1.一种针对关键节点的大规模电力系统潮流收敛性调整方法，其特征包括：通过牛顿迭代法(Newton-Raphson)迭代的中间数据，建立潮流收敛性的判断指标；根据指标值所代表的具体意义辨识关键影响节点和因素；根据关键影响节点和因素刻画潮流收敛域；依赖不同调整目标对关键数据量进行调整；具体内容为：1. A large-scale power system power flow convergence adjustment method for key nodes, characterized by comprising: establishing a judgment index for power flow convergence through intermediate data iterative Newton-Raphson method (Newton-Raphson); The specific meaning is to identify the key influencing nodes and factors; to describe the power flow convergence region according to the key influencing nodes and factors; to adjust the key data volume depending on different adjustment targets; the specific contents are: 1)通过牛顿迭代法(Newton-Raphson)潮流迭代的中间数据，建立潮流收敛性的判断指标：潮流迭代的中间过程数据蕴含着电网的重要信息，孤立地看，每次潮流迭代求解结果都能够代表一个独立的潮流形式，即每一节点的电气量关系满足物理规律，冗余功率皆由平衡节点承担，由于计算结果与注入数值之间的误差无法满足精度要求，所以才需要进行反复迭代，因此，从连续上看，两次迭代之间数据的变化情况能够反映潮流方程的收敛能力，牛顿迭代法计算过程中，|Δy(k)|作为各节点功率的不平衡量，体现了迭代过程中各节点功率的计算值与注入数据之间误差的变化情况，由此建立指标：1) Through the intermediate data of the Newton-Raphson power flow iteration, the judgment index of the power flow convergence is established: the intermediate process data of the power flow iteration contains important information of the power grid. Viewed in isolation, the solution results of each power flow iteration can be Represents an independent power flow form, that is, the electrical quantity relationship of each node satisfies the physical law, and the redundant power is borne by the balance node. Since the error between the calculation result and the injected value cannot meet the accuracy requirements, repeated iterations are required. Therefore, from a continuous point of view, the change of data between two iterations can reflect the convergence ability of the power flow equation. In the calculation process of the Newton iteration method, |Δy (k) | The change of the error between the calculated value of the power of each node and the injected data is used to establish the index:
Figure FDA0003342512320000011
Figure FDA0003342512320000011
 则潮流收敛条件为：Then the power flow convergence condition is:
Figure FDA0003342512320000012
Figure FDA0003342512320000012
 式中
Figure FDA0003342512320000013
为关键因素判别指标；k表示迭代次数；kmax表示预设的最大迭代次数；Δy表示节点功率不平衡量；μ表示预设的迭代求解精度；in the formula
Figure FDA0003342512320000013
is the key factor discriminating index; k represents the number of iterations; k max represents the preset maximum number of iterations; Δy represents the node power imbalance; μ represents the preset iterative solution accuracy; 2)根据指标值所代表的具体意义辨识关键影响节点和因素：借助Δy(k)反映功率计算值与注入数据之间误差的特性，提出一种辨识潮流不收敛关键因素的方法，牛顿迭代法潮流计算过程中，Δx(k)与Δy(k)都能够体现潮流解的精度，但两者所代表的意义有所不同，Δx(k)代表状态变量在两次迭代计算之间变化的差值，当Δx(k)逐渐趋于减小时，能够反映状态变量的变化情况逐渐趋于稳定；而Δy(k)则代表了计算值与真值间的误差，对于含有高于两个自变量的方程组而言，计算值与真值间的误差就由|Δy(k)|中的最大值决定，当|Δy(k)|逐渐趋于减小时，则说明在该迭代过程中，迭代计算量正逐渐向注入数据方向逼近，相比Δx(k)而言，Δy(k)更具有明确的实际意义，Emax越大表示该次迭代下，计算值较注入数据之间误差越大，相反的，Emax越小则表示在该次迭代中计算值与注入数据误差越小；2) Identify the key influencing nodes and factors according to the specific meaning represented by the index value: With the help of Δy (k) to reflect the error between the power calculation value and the injected data, a method to identify the key factors of the non-convergence of power flow is proposed, the Newton iteration method In the process of power flow calculation, both Δx (k) and Δy (k) can reflect the accuracy of the power flow solution, but the meanings of the two are different. Δx (k) represents the difference between the changes of state variables between two iterations When Δx (k) tends to decrease gradually, it can reflect that the change of state variables gradually tends to be stable; while Δy (k) represents the error between the calculated value and the true value. For the equation system of , the error between the calculated value and the true value is determined by the maximum value in |Δy (k ) |. When |Δy (k) | gradually decreases, it means that in the iterative process, the iterative The calculation amount is gradually approaching the direction of the injected data. Compared with Δx (k) , Δy (k) has a more clear practical significance. The larger E max is, the larger the error between the calculated value and the injected data is in this iteration. , on the contrary, the smaller the E max is, the smaller the error between the calculated value and the injected data is in this iteration; 3)根据关键影响节点和因素刻画潮流收敛域：潮流收敛域指的是所有满足潮流计算收敛条件的运行点数据的集合，根据辨识出的潮流不收敛关键因素，以潮流计算收敛为边界条件刻画潮流收敛域，以数据点相对于收敛域边界的位置关系，提供数据的修正方案，当数据点位于收敛域以内时，潮流计算收敛；当数据点位于收敛域以外时，潮流计算不能收敛，理论上讲收敛域可以刻画至无限维，但高于二维的收敛域不利于分析后续的潮流修正，因此对于辨识出的问题因素多于两个的，可以绘制多个二维域分别进行分析；3) Characterize the power flow convergence region according to the key influencing nodes and factors: the power flow convergence region refers to the collection of all operating point data that meet the convergence conditions of power flow calculation. The power flow convergence region provides a data correction scheme based on the positional relationship of the data points relative to the boundary of the convergence region. When the data points are located within the convergence region, the power flow calculation converges; when the data points are outside the convergence region, the power flow calculation cannot converge. As mentioned above, the convergence region can be described to infinite dimensions, but the convergence region higher than two-dimensional is not conducive to the analysis of subsequent power flow correction. Therefore, if there are more than two problem factors identified, multiple two-dimensional domains can be drawn for analysis respectively; 4)依赖不同调整目标对关键数据量进行调整：当潮流计算不收敛时，即数据点位于收敛域以外时，修正数据点数据，将数据点拉入收敛域以内，即实现了通过修正关键控制变量将不收敛的潮流调至收敛，采用以修正单个数据为目标的数据修正方案和以修正数据总量最小为目标的数据修正方案，以修正单个数据为目标的数据修正方案的核心在于，求取数据点到收敛边界平行于坐标轴方向上的最短距离。4) Adjust the amount of key data depending on different adjustment objectives: when the power flow calculation does not converge, that is, when the data points are outside the convergence area, correct the data point data and pull the data points into the convergence area, that is, the key control by correcting is realized. The variable adjusts the non-convergent power flow to convergence, and adopts a data correction scheme aiming at correcting a single data and a data correction scheme aiming at the minimum amount of correction data. The core of the data correction scheme aiming at correcting a single data is to find Take the shortest distance from the data point to the convergence boundary in the direction parallel to the coordinate axis.
CN201910388541.9A 2019-05-10 2019-05-10 A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes Active CN110137967B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910388541.9A CN110137967B (en) 2019-05-10 2019-05-10 A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910388541.9A CN110137967B (en) 2019-05-10 2019-05-10 A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes

Publications (2)

Publication Number Publication Date
CN110137967A CN110137967A (en) 2019-08-16
CN110137967B true CN110137967B (en) 2022-03-22

Family

ID=67573183

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910388541.9A Active CN110137967B (en) 2019-05-10 2019-05-10 A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes

Country Status (1)

Country Link
CN (1) CN110137967B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112564113B (en) * 2020-12-15 2023-05-12 深圳供电局有限公司 Convergence adjusting method based on approximated tide voltage stabilization
CN113098018B (en) * 2021-04-14 2022-06-03 湘潭大学 A Convergence Adjustment Method for Multi-energy Flow Calculation in Electric-Gas Regional Integrated Energy System

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104485661A (en) * 2014-11-17 2015-04-01 中国南方电网有限责任公司 Tidal current automatic regulating method based on node type conversion
CN104732459A (en) * 2015-03-31 2015-06-24 上海交通大学 Large-scale power system ill-condition load flow analysis system
CN104953585A (en) * 2015-07-22 2015-09-30 西南石油大学 Method for calculating distribution network flow
US20150378387A1 (en) * 2013-05-14 2015-12-31 Rensselaer Polytechnic Institute Methods of computing steady-state voltage stability margins of power systems
CN108376984A (en) * 2018-01-25 2018-08-07 云南电网有限责任公司 A kind of Electric Power System Node Voltage automatic adjusting method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150378387A1 (en) * 2013-05-14 2015-12-31 Rensselaer Polytechnic Institute Methods of computing steady-state voltage stability margins of power systems
CN104485661A (en) * 2014-11-17 2015-04-01 中国南方电网有限责任公司 Tidal current automatic regulating method based on node type conversion
CN104732459A (en) * 2015-03-31 2015-06-24 上海交通大学 Large-scale power system ill-condition load flow analysis system
CN104953585A (en) * 2015-07-22 2015-09-30 西南石油大学 Method for calculating distribution network flow
CN108376984A (en) * 2018-01-25 2018-08-07 云南电网有限责任公司 A kind of Electric Power System Node Voltage automatic adjusting method

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
An Improved Approach for AC-DC Power Flow Calculation With Multi-Infeed DC Systems;Chongru Liu;《IEEE TRANSACTIONS ON POWER SYSTEMS》;20110531;第26卷(第2期);全文 *
Power Flow Methods for Improving Convergence;P. J. Lagace;《IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society》;20121224;全文 *
几种高阶收敛的Levenberg-Marquardt 方法在潮流计算中的应用;曹佳;《电网技术》;20170430;第41卷(第4期);全文 *
基于节点类型转换的潮流收敛性调整方法;李智欢;《电力系统自动化》;20150410;第39卷(第7期);全文 *
大电网病态潮流的识别和修正方法;彭慧敏;《电力系统保护与控制》;20181116;第46卷(第22期);全文 *

Also Published As

Publication number Publication date
CN110137967A (en) 2019-08-16

Similar Documents

Publication Publication Date Title
CN107332240B (en) Boundary search method of power system static voltage stability domain based on optimization model
CN105978003B (en) A kind of additional wide area damping control design method of power system for considering time lag
CN111224404B (en) Fast power flow control method for power system with controllable phase shifter
CN111181164B (en) An improved master-slave split transmission and distribution coordinated power flow calculation method and system
CN110137967B (en) A Power Flow Convergence Adjustment Method for Large-scale Power Systems for Key Nodes
CN108075480B (en) A state estimation method and system for an AC/DC system
CN111682557B (en) Optimization method for configuration place and optimal compensation capacity of reactive power equipment of power system
CN115688276A (en) Aircraft appearance automatic optimization method, system, equipment and medium based on discrete companion method
CN108564205A (en) A kind of load model and parameter identification optimization method based on measured data
CN110247390A (en) A kind of polymorphic type distributed generation resource Optimal Configuration Method based on immunity particle cluster algorithm
CN106410811A (en) Power flow calculation method for the first iterative small-impedance branch end-point change Jacobian matrix
CN105914752B (en) Pilot bus selection method based on fast search and density peaks cluster
CN115563719A (en) Single-pipeline double-layer optimized layout method in aircraft fuel tank
CN111697575B (en) Flexibility-improved quantitative analysis method, device and system
CN105305484B (en) A Distributed Power Optimal Configuration Method Applicable to Distribution Network
CN119010187A (en) Power grid control optimization system and method based on continuous optimization iteration
CN113270867B (en) Automatic adjustment method for weak power grid tide without solution
CN106712029B (en) Newton's method for power flow calculation method of variable Jacobian matrix at PQ end point of small impedance branch
CN113809779B (en) Stability inspection method for wind farm when connected into flexible direct system
CN109698516A (en) The maximum capacity computing system and method for renewable energy access power distribution network
CN115714421A (en) Distributed power supply site selection and volume fixing method based on intelligent single particle optimization algorithm
CN109698498A (en) Using the power grid security assessment system and method for voltage stability margin index
CN108123434A (en) It is a kind of to calculate the PV slopes of curve to ask for the method for PV curve motions point
CN113659578A (en) An optimal configuration method of UPFC and STATCOM considering the available transmission capacity of the system
CN112653152A (en) High-dimensional space similarity-based power flow convergence adjusting method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant