Phase shifter locating and sizing method and system for extreme scene

Abstract

The invention discloses an extreme scene-oriented phase shifter locating and sizing method and system, and the method comprises the steps: constructing an annual hourly operation data set based on production simulation data; carrying out power flow out-of-limit judgment and N-1 safety verification on the operation data set, identifying a power flow out-of-limit scene and an unsafe scene in the operation data set, and constructing an extreme candidate scene set; selecting a representative extreme scene from the extreme candidate scene set by using a K-Medoids + + algorithm based on a weighted Euclidean distance; based on representative extreme scenes, under the conditions of various cost constraints, investment decision constraints and power grid operation constraints, a linearized phase shifter locating and sizing optimization model is constructed, a mixed integer programming solver is utilized to solve the phase shifter locating and sizing optimization model, and a locating and sizing result is obtained. The method can be suitable for locating and sizing of a plurality of phase shifters in various extreme scenes, and the solving process is fast and accurate.

Description

TECHNICAL FIELD

[0001] The application relates to a phase shifter site selection and capacity configuration technology for extreme scenarios and belongs to the technical field of power system operation and planning. BACKGROUND

[0002] With the increasing penetration of renewable energy sources such as wind power and photovoltaic power, the proportion of green power is increasing, but it also brings severe challenges to the operation and planning of power systems. On the one hand, the output of new energy is highly volatile, which can easily cause problems such as line overload and frequency change. On the other hand, in some areas, due to the limitations of the power grid structure, the new energy consumption rate is limited. In order to improve the power flow regulation capability of the system and the new energy consumption level, it is urgent to configure flexible resources. As a device that can effectively regulate active power flow, the phase shifter is becoming an important measure under the background of high penetration of new energy, because it can achieve power flow reconstruction by changing the phase angle difference. The performance of the phase shifter is highly dependent on the appropriate deployment location and capacity configuration, therefore, it is necessary to study the phase shifter site selection and capacity configuration method.

[0003] In the prior art of phase shifter site selection and capacity configuration, one type of method optimizes the location and phase angle of the phase shifter based on power flow sensitivity, linear search or intelligent algorithm, so as to relieve the system power imbalance; another type of method configures the phase shifter by establishing a double-layer optimization model or based on an alternating current power flow model to improve the consumption capacity of new energy. However, most of these methods fail to fully consider the volatility characteristics or scenario diversity of new energy output, and have problems such as insufficient accuracy, limited scope of application or excessive simplification.

[0004] With the frequent occurrence of extreme weather, comprehensively considering different operating scenarios can help ensure the safety of the system and improve the adaptability of the planning. However, the introduction of a large number of operating scenarios significantly increases the scale and solution difficulty of the optimization problem. Some studies propose a method of using a small number of typical scenarios to replace a large number of operating scenarios to reduce the amount of calculation, but this method often ignores the influence of extreme scenarios, and it is difficult to guarantee the safety and stability of the system under extreme conditions. Some studies consider extreme scenarios, but mostly use empirical methods or simple clustering methods to select extreme scenarios, which have problems of insufficient representation or omission, and it is difficult to systematically cover multiple complex extreme operating conditions.

[0005] In summary, the existing technology lacks an effective method that can consider the volatility characteristics of new energy output and systematically identify extreme scenarios in the problem of phase shifter site selection and capacity configuration, which has become a technical problem that needs to be solved in power system planning and operation under the background of increasing penetration of new energy and frequent occurrence of extreme weather. SUMMARY

[0006] The technical problems to be solved by the present application are: under the background of high penetration of new energy, how to identify and screen extreme operation scenarios, and on this basis, how to build an optimization model to determine the installation location and capacity configuration of the phase shifter, so as to improve the power flow regulation capability of the power grid and the new energy consumption level.

[0007] To solve the above technical problems, the technical scheme adopted by the present application is as follows.

[0008] A phase shifter site selection and capacity determination method for extreme scenarios, comprising the following steps:

[0009] Step S1, based on production simulation data, an annual hourly operation data set is constructed;

[0010] Step S2, the operation data set is subjected to power flow overrun judgment and N-1 safety check, and the power flow overrun scenarios and unsafe scenarios in the operation data set are identified to construct an extreme candidate scenario set;

[0011] Step S3, a representative extreme scenario is selected from the extreme candidate scenario set by using a K-Medoids++ algorithm based on weighted Euclidean distance;

[0012] Step S4, based on the representative extreme scenario, a linearized phase shifter site selection and capacity determination optimization model is constructed under the conditions of various cost constraints, investment decision constraints and power grid operation constraints, and a mixed integer programming solver is used to solve the phase shifter site selection and capacity determination optimization model to obtain the site selection and capacity determination result.

[0013] The foregoing phase shifter site selection and capacity determination method for extreme scenarios, in step S1, the step of constructing an annual hourly operation data set comprises:

[0014] Step S11, input various parameters required for production simulation, including a node set , a branch set , the resistance and reactance of the branch , , , the rated capacity of the branch , the segmented linearization generation cost function of the unit , the ramp-up / ramp-down constraints of the unit and , the active load of the node at time , the photovoltaic irradiance at time , the environmental temperature at time , and the wind speed at time . ​​​​​​

[0015] Step S12, calculate the renewable energy output including photovoltaic output and wind power output hour by hour ;

[0016] Step S13, construct a production simulation model based on direct current flow, use a mixed integer programming solver to solve the hourly calculation, get the unit output and flow distribution data under the annual scenario, and constitute the annual hourly operation data set.

[0017] The aforementioned phase shifter site selection and capacity determination method for extreme scenarios, in step S12, the photovoltaic output calculation formula is:

[0018] (1)

[0019] Where, is the efficiency, is the component area, is the temperature coefficient, is the cell temperature, is the maximum active power output of the photovoltaic unit of node n;

[0020] The wind power output calculation formula is:

[0021] (2)

[0022] Where, , , are the cut-in, rated and cut-out wind speeds respectively, is the unit rated power.

[0023] The aforementioned phase shifter site selection and capacity determination method for extreme scenarios, in step S13, the objective function of the production simulation model is:

[0024] (3)

[0025] Where, is the unit output, indicates that the optimized unit output minimizes the objective function, and Ci(⋅) is a piecewise linear cost function.

[0026] The constraint conditions are:

[0027] (4)

[0028] (5)

[0029] (6) ​

[0030] wherein, Y is the admittance coefficient, , are the voltages of nodes , are the voltage phase angles of nodes at time t, are the active power outputs of nodes at time t, are the voltage magnitudes of nodes at time t, are the lower voltage limits of nodes , are the upper voltage limits of nodes , and represent the ramp-up and ramp-down constraints of units , i.e., the maximum power that can be increased and decreased between two adjacent time instants.

[0031] The aforementioned phase shifter siting and sizing method for extreme scenarios, in step S2, the step of identifying the power flow out-of-limit scenarios and unsafe scenarios in the operation data set comprises:

[0032] Step S21, compare the absolute value of branch power flow and the rated capacity of branch , if there exists a branch , that satisfies , wherein represents the absolute value, then the scenario is included in the power flow out-of-limit set ;

[0033] Step S22, in the remaining scenarios excluding the out-of-limit scenarios, identify the scenarios that do not satisfy the N-1 principle, and include the identified scenarios in the unsafe set ;

[0034] Step S23, construct the extreme candidate scenario set , represents the union.

[0035] The aforementioned phase shifter siting and sizing method for extreme scenarios, in step S22, for each scenario , the N-1 checking operation performed comprises:

[0036] Step S221, cut off branch from the network topology , , , construct a new node admittance matrix ;

[0037] Step S222, solve the DC power flow equation under the condition of keeping the power injection of each node unchanged:

[0038] (7)

[0039] wherein, is the voltage phase angle vector of each node, is the active power injection of the node in the scenario ;

[0040] Step S223, calculate the branch flow after the removal:

[0041] (8)

[0042] wherein, is the branch flow after the removal of the line in the scenario , is the branch admittance, and are the voltage phase angles of the node and the node after the removal of the line in the scenario ; S224, if there exists a branch ,

[0043] satisfying , then it is determined that the scenario does not satisfy the N-1 safety under the working condition of removing the line , and the scenario is included in the unsafe scenario set .

[0044] The foregoing phase shifter site selection and capacity determination method for extreme scenarios, in the step S3, the step of selecting a representative extreme scenario includes:

[0045] S31, construct a node feature vector and a node weight , wherein, and are the active load and the active output of new energy of the node in the scenario ; is the set of all nodes;

[0046] S32, set the number of clusters to , and randomly select one scenario in the extreme candidate scenario set as a first center , record the current center set = ;

[0047] S33, for any scene , calculate the weighted Euclidean distance of the scene to the nearest existing center, wherein, is the existing cluster center, is the weighted Euclidean distance;

[0048] S34, according to the probability , extract the next center from the extreme candidate scene set , add it to the current center set , until initial centers are obtained , wherein is the initial cluster center, is the weighted Euclidean distance of the scene to the nearest selected cluster center, denotes the weighted Euclidean distance of any scene in the extreme candidate scene set to the nearest selected cluster center;

[0049] S35, assign each scene to the nearest center to form a cluster , in each cluster, calculate the sum of the weighted Euclidean distances between each sample and other samples, and take the sample with the smallest sum of weighted Euclidean distances as the new cluster center of the corresponding cluster;

[0050] S36, if all cluster centers no longer change or the algorithm reaches the preset maximum number of iterations, stop; otherwise, return to step S35;

[0051] S37, the representative extreme scene set is composed of the center samples at the time of convergence , wherein denotes the center sample in the cluster corresponding to the cluster center.

[0052] The aforementioned extreme scene-oriented shifter site selection and capacity determination method, in step S31, the node weight formula is:

[0053] (9)

[0054] wherein, is the weight coefficient of the node with new energy access determined according to expert experience.

[0055] The aforementioned extreme scene-oriented shifter site selection and capacity determination method, in step S33, the weighted Euclidean distance is calculated as follows:

[0056] (10)

[0057] wherein, , is the scenario clustering center the normalized load and new energy component under the scenario, , is the scenario the normalized load and new energy component under the scenario, the formula is as follows

[0058] (11)

[0059] (12)

[0060] wherein, and is the active load and new energy output of the node under the scenario , and is the mean value of the active load and new energy output of the node , and is the standard deviation of the active load and new energy output of the node .

[0061] The aforementioned phase shifter site selection and capacity determination method for extreme scenarios, in step S4, the step of constructing and solving the phase shifter site selection and capacity determination model comprises:

[0062] Step S41, considering the investment cost, operation cost and penalty cost caused by overloaded lines and new energy curtailment of the phase shifter, the objective function is constructed:

[0063] (11)

[0064] wherein, represents minimizing the total cost, represents the investment cost of the phase shifter, represents the grid operation cost, represents the penalty cost;

[0065] Step S42, according to the total number limit of the phase shifter, the phase angle discretization selection, the expression of the phase angle of each branch, the upper and lower limits of the phase shifter capacity, the investment decision constraints of the phase shifter are constructed;

[0066] Step S43, according to the node power balance, the linearized alternating current flow expression, the new energy curtailment constraint and the linearized expression of the absolute value of the branch flow, as well as the conventional flow upper and lower limit constraints, the reference bus constraint and the phase angle difference constraint, the grid operation constraints are constructed;

[0067] Step S44, using a mixed integer programming solver to solve the established model to obtain the shifter installation location and installation capacity.

[0068] The aforementioned shifter site selection and capacity determination method for extreme scenarios, in step S41, the calculation formula of each cost is:

[0069] (12)

[0070] wherein, is the annual fixed investment cost per unit, is the annual cost coefficient per unit capacity, is

[0071] branch shifter capacity, is the total number of branches;

[0072] (13)

[0073] wherein, is the generation cost, is the simulated network loss cost, is the total number of scenarios, is the weight of scenario , and are the total generation cost coefficient and network loss cost coefficient of scenario , is the total number of generators, is the total number of segments, is the segment cost coefficient, is the unit segment m output, is the resistance of branch , is the active power flow from node to node in scenario ,

[0074] (14)

[0075] (15)

[0076] wherein, and are the abandoned power cost coefficient and load rate penalty coefficient in scenario , is the new energy abandoned power in scenario , is the branch In the scenario where the security load rate threshold is exceeded, as the heavy load threshold.

[0077] The aforementioned phase shifter site selection and capacity determination method for extreme scenarios, in step S42, the formula for phase shifter total number limit, phase shift angle discretization selection, each branch phase shift angle expression, and upper and lower limits of phase shifter capacity is:

[0078] (16)

[0079] where, ∈{0,1}, indicates whether the branch is installed with a phase shifter, is the number of phase shifters, and is the minimum and maximum number of phase shifters;

[0080] (17)

[0081] where, ∈{0,1}, indicates whether the branch is selected in the scenario of the first discrete phase shift angle, if the branch is installed with a phase shifter, i.e. =1, then a phase angle must be selected from the preset discrete angles; if no phase shifter is installed, then no angle is allowed to be selected;

[0082] (18)

[0083] where, is the preset angle candidate set, and equation (18) indicates that the phase shift angle of the branch in the scenario is determined by the weighted discrete variable;

[0084] (19)

[0085] where, is the capacity of the phase shifter installed on the branch , and is the maximum installed capacity of the phase shifter.

[0086] The aforementioned phase shifter site selection and capacity determination method for extreme scenarios, in step S43, the formula for node power balance, linearized alternating current power flow expression, new energy curtailment quantity constraint, and linearized expression of branch power flow absolute value is:

[0087] (20)

[0088] (21)

[0089] wherein, is the set of generators connected at node , is the active power output of generator at node , is the active power flow from node , is the new energy output of node at scenario , is the active load of node , represents the active power flowing in, is the reactive power output of generator at node , is the reactive load of node , is the reactive power flow from node , is the reactive power flow from node , is the reactive power flow from node , is the reactive power flow from node , ;

[0090] (22)

[0091] (23)

[0092] wherein, and represent the voltage magnitude and phase angle of node at scenario , and are the conductance and susceptance of branch , and are the voltage phase angles of node at scenario , is the phase shift angle of branch at scenario , represents the phase angle difference after considering the phase shift angle;

[0093] (24)

[0094] wherein, is the scenario the reduction amount of new energy output, is the theoretical maximum value of new energy output;

[0095] (25)

[0096] wherein, is an auxiliary variable introduced.

[0097] A computer system comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the steps of the method as described above.

[0098] The application achieves the beneficial effects: the phase shifter site selection and capacity selection method for extreme scenarios, by introducing extreme scenario identification and representative scenario screening, the planning scheme obtained can take into account various faults and operation uncertainties, maintain stable operation of the power grid under extreme working conditions, and significantly improve the safety and adaptability of the power grid under high penetration of new energy and extreme operating conditions. Both the comprehensiveness and scientificity of the extreme scenario coverage are ensured, and redundancy and deviation are avoided, so that the selected scenarios can accurately reflect the operating characteristics of the power grid under various extreme conditions, thereby providing reliable input for subsequent optimization models and improving the safety and adaptability of the planning results. Both the data integrity, the scenario representativeness and the model solvability are taken into account, the operating characteristics of the power grid under extreme conditions are accurately reflected, the calculation efficiency and engineering implementability are ensured, and the method has strong popularization and application value. BRIEF DESCRIPTION OF DRAWINGS

[0099] Figure 1 is a whole flowchart of a phase shifter site selection and capacity selection method for extreme scenarios in embodiment 1 of the application;

[0100] Figure 2 is a N-1 safety check flowchart in the method provided in embodiment 1 of the application;

[0101] Figure 3 is a representative extreme scenario selection flowchart of the method provided in embodiment 1 of the application;

[0102] Figure 4 is a whole framework diagram of the site selection and capacity selection optimization model in the method provided in embodiment 1 of the application. DETAILED DESCRIPTION

[0103] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0104] Example 1

[0105] like Figure 1 As shown, this embodiment provides a phase shifter addressing and calibrating method for extreme scenarios, including the following steps:

[0106] Step S1: Construct an annual hourly running dataset based on production simulation data;

[0107] Step S2: Perform power flow limit violation judgment and N-1 safety verification on the running dataset to identify power flow limit violation scenarios and unsafe scenarios in the running dataset and construct an extreme candidate scenario set;

[0108] Step S3: Use the K-Medoids++ algorithm based on weighted Euclidean distance to select representative extreme scenes from the extreme candidate scene set;

[0109] Step S4: Based on representative extreme scenarios, under various cost constraints, investment decision constraints, and power grid operation constraints, a linear phase shifter location and capacity optimization model is constructed, and a mixed-integer programming solver is used to solve the phase shifter location branch. The resistance-based capacity optimization model was used to obtain the location and capacity selection results.

[0110] In step S1, the steps for constructing the annual hourly running dataset include:

[0111] Step S11: Input the various parameters required for the production simulation, including the node set. Branch set , and reactance ( , ), branch road Rated capacity ,unit Piecewise linearized power generation cost function ,unit Climbing / descent constraints and Node i at time active load ,time Photovoltaic irradiance, at any time ambient temperature ,time wind speed ;

[0112] Step S12, calculate the renewable energy output including photovoltaic output and wind power output hour by hour ; ;

[0113] Step S13, construct a production simulation model based on direct current flow, use a mixed integer programming solver to solve the hourly calculation, get the unit output and flow distribution data under the annual scenario, and constitute the annual hourly operation data set.

[0114] In the step S12, the photovoltaic output calculation formula is:

[0115] (1)

[0116] Wherein, is the efficiency, is the component area, is the temperature coefficient, is the cell temperature, is the maximum active power output of the photovoltaic unit of the node .

[0117] The wind power output calculation formula is:

[0118] (2)

[0119] Wherein, , , are the cut-in, rated and cut-out wind speeds respectively, is the unit rated power.

[0120] In the step S13, the objective function of the production simulation model is:

[0121] (3)

[0122] Wherein, is the unit output, indicates that the optimized unit output makes the objective function minimum, and Ci(⋅) is a piecewise linear cost function.

[0123] The constraint condition is:

[0124] (4)

[0125] (5)

[0126] (6)

[0127] Wherein, is the admittance coefficient of the branch, 、 is the voltage phase angle of the node at the time is the voltage amplitude of the node at the time is the active power output of the new energy of the node at the time is the sum of the photovoltaic output and the wind power output, is the voltage amplitude of the node at the time is the lower limit of the node voltage, is the upper limit of the node voltage, and respectively represent the climbing rising and falling constraints of the unit , that is, the maximum output that can be increased and decreased between adjacent two times.

[0128] The present application generates the load and new energy output data of 8760 hours in a year through a production simulation method, can systematically mine the key time sequence characteristics affecting the power flow distribution and new energy consumption, avoids the representativeness problem caused by relying on a few typical days or experience selection, makes the input data more comprehensive and reliable, and lays a solid foundation for subsequent extreme scene identification and optimization.

[0129] In the step S2, the step of identifying the power flow out-of-limit scene and the unsafe scene in the operation data set comprises:

[0130] In step S21, the absolute value of the branch power flow is compared with the size of the rated capacity of the branch . If there exists a branch , that satisfies , wherein represents the absolute value, then the scene is included in the power flow out-of-limit set .

[0131] In step S22, among the remaining scenes excluding the out-of-limit scenes, the scene that does not satisfy the N-1 principle is identified, and the identified scene is included in the unsafe set .

[0132] In step S23, an extreme candidate scene set is constructed, which represents the union set.

[0133] In the step S22, for each scene , the N-1 checking operation performed comprises:

[0134] ​Step S221, removing branches from the network topology , corresponding nodes , , constructing a new node admittance matrix ;

[0135] Step S222, solving the direct current flow equation under the condition of keeping the power injection of each node unchanged:

[0136] (7)

[0137] wherein, is a voltage phase angle vector of each node, is a node active power injection of the scenario .

[0138] Step S223, calculating the branch flow after removal:

[0139] (8)

[0140] wherein, is a branch flow after removal of the line in the scenario , is a branch susceptance, and are voltage phases of nodes and after removal of the line in the scenario , and are voltage phases of nodes and

[0141] after removal of the line in the scenario . S224, if there is a branch , satisfying , it is determined that the scenario does not satisfy the N-1 safety under the working condition of removing the line , and the scenario is included in the unsafe scenario set

[0142] .

[0143] In the step S3, the step of selecting a representative extreme scenario comprises:

[0144] Step S31, constructing a node feature vector and a node weight wherein, and are the active load and the active power output of new energy of the scenario under the node ; is the set of all nodes;

[0145] Step S32, set the number of clusters to , randomly select 1 scenario from the set of extreme candidate scenarios as the first center , and record the current center set = ;

[0146] Step S33, for any scenario , calculate the weighted Euclidean distance from the scenario to the nearest existing center , wherein, is the existing cluster center, is the weighted Euclidean distance; see equation (10).

[0147] S34, according to the probability , extract the next center from the set of extreme candidate scenarios , and add it to the current center set , until initial centers are obtained , wherein is the initial cluster center, is the weighted Euclidean distance from the scenario to the nearest selected cluster center, denotes the weighted Euclidean distance from any scenario in the set of extreme candidate scenarios to the nearest selected cluster center.

[0148] S35, assign each scenario to the nearest center to form a cluster , and in each cluster, calculate the sum of the weighted Euclidean distances between each sample and other samples, and take the sample with the smallest sum of weighted Euclidean distances as the new cluster center of the corresponding cluster.

[0149] S36, if all cluster centers no longer change or the algorithm reaches the preset maximum number of iterations, stop; otherwise, return to S35.

[0150] S37, the representative set of extreme scenarios is composed of the center samples of the centers at the time of convergence , wherein denotes the center sample in the cluster corresponding to the cluster center.

[0151] In step S31, the node weight formula is:

[0152] (9)

[0153] wherein, is the weight coefficient of the node with new energy access determined according to expert experience, and can be taken as 10.

[0154] In the step S33, the weighted Euclidean distance The calculation formula is:

[0155] (10)

[0156] wherein, , is the scene clustering center The normalized load and new energy component under the scene, , is the scene The normalized load and new energy component under the scene, and the formula is as follows

[0157] (11)

[0158] (12)

[0159] wherein, and are the active load and new energy output of the node under the scene , and are the mean values of the active load and new energy output of the node , and are the standard deviations of the active load and new energy output of the node .

[0160] In the extreme candidate scene identification method in the application, first, the branch flow limit scene and the scene not meeting the N-1 safety from the annual operation data are screened out to form an extreme candidate set; then, the K-Medoids++ clustering method of weighted Euclidean distance is used to select representative scenes from the candidate set to ensure covering multiple types of extreme operating states. This process can reduce redundant scenes while ensuring comprehensive coverage, avoid the deviation caused by experience selection and single index selection, and thus more efficiently and accurately identify the extreme operating conditions faced by the system.

[0161] In the step S4, the steps of constructing and solving the phase shifter site selection and capacity determination model include:

[0162] Step S41, considering the phase shifter investment cost, operation cost and the penalty cost caused by heavy load line and new energy curtailment, a target function is constructed:

[0163] (11)

[0164] wherein, represents the minimization of total cost, represents the phase shifter investment cost, represents the grid operation cost, represents the penalty cost.

[0165] Step S42, according to the total number of phase shifter restrictions, phase angle discretization selection, branch phase angle expression, upper and lower limits of phase shifter capacity, the phase shifter investment decision constraints are constructed.

[0166] Step S43, according to the node power balance, linearized AC power flow expression, new energy curtailment constraint and linearized expression of branch power flow absolute value, as well as the conventional power flow upper and lower limit constraints, reference bus constraints and phase angle difference constraints, the grid operation constraints are constructed.

[0167] Step S44, the established model is solved by using a mixed integer programming solver to obtain the installation location and installation capacity of the phase shifter.

[0168] In the step S41, the calculation formula of each cost is:

[0169] (12)

[0170] wherein, is the annual fixed investment cost per unit number, is the annual cost coefficient per unit capacity, is the total number of branches,

[0171] branch phase shifter capacity, is the total number of branches;

[0172] (13)

[0173] wherein, is the generation cost, is the simulation network loss cost, is the total number of scenarios, is the weight of scenario , and are the total generation cost coefficient and network loss cost coefficient of scenario , is the total number of generators, is the total number of segments, is the segment cost coefficient, It is a generator set Segmented m-output It is a side road The resistance, It is a scene Next node To the node The positive trend.

[0174] (14)

[0175] (15)

[0176] in, and Scenes Reduce the cost factor for power curtailment and the load factor penalty factor. It is a scene The amount of renewable energy power that has been wasted It is a side road In the scene The portion exceeding the safe load rate threshold, The overload threshold is set to 0.85.

[0177] In step S42, the formulas for limiting the total number of phase shifters, selecting the phase shift angle discretization, expressing the phase shift angle of each branch, and setting the upper and lower limits of the phase shifter capacity are as follows:

[0178] (16)

[0179] in, ∈{0,1} represents a branch Whether to install a phase shifter It is the number of phase shifters. and These are the minimum and maximum number of phase shifters.

[0180] (17)

[0181] in, ∈{0,1} represents a branch. In the scene Should the first option be selected? A discrete phase shift angle. If a phase shifter is installed on each branch, then... =1, then it must be from the preset. Choose one angle from a set of discrete angles; if no phase shifter is installed, arbitrary angle selection is not allowed.

[0182] (18)

[0183] in, is a preset angle candidate set, and the embodiment takes , formula (18) represents the branch The phase shift angle under the scene is determined by a discrete variable weighted.

[0184] (19)

[0185] wherein, is the capacity of the phase shifter installed on the branch , and is the maximum installed capacity of the phase shifter.

[0186] In the step S43, the formula of the node power balance, the linearized alternating current power flow expression, the new energy curtailment constraint and the linearized expression of the branch power flow absolute value is:

[0187] (20)

[0188] (21)

[0189] wherein, is the set of generators connected to the node , and is the active power output of the generator at the node under the scene , and is the new energy output of the node under the scene , and is the active load of the node , represents the active power flowing out of the node through the line, represents the active power flowing in, is the reactive power output of the generator at the node under the scene , and is the reactive load of the node , and is the reactive power flow from the node to the node under the scene , and is the reactive power flow from the node to the node under the scene .

[0190] (22)

[0191] (23)

[0192] wherein, and denote the voltage phase angle of the nodes under the scenario , the voltage amplitude and phase angle of the nodes and are the conductance and susceptance of the branches , Bsh,ij is the parallel susceptance, and are the voltage phase angles of the nodes under the scenario , is the phase shift angle of the branches under the scenario , denote the phase angle difference after considering the phase shift angle.

[0193] (24)

[0194] wherein, is the reduction of new energy output under the scenario , is the theoretical maximum value of new energy output.

[0195] (25)

[0196] wherein, is an auxiliary variable introduced.

[0197] The application establishes a phase shifter site selection and capacity optimization model facing representative extreme scenarios, comprehensively considers investment cost, operation cost, network loss and power curtailment penalty in the objective function, introduces phase shifter investment decision constraints and power grid operation constraints in the constraint conditions, can linearly solve the installation position and capacity of the phase shifter, and thus better improves the power flow regulation capability and new energy consumption level of the power grid.

[0198] Embodiment 2

[0199] A computer system comprising a memory, a processor and a computer program stored on the memory, the processor executing the computer program to implement the steps of the method as described in embodiment 1.

[0200] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block or blocks in the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0201] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0202] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flows and / or block or blocks in the block diagrams. Figure 1 one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams. ​ one or more functions specified in the flowchart or flows and / or block or blocks of the block diagrams.

[0203] The above only is the preferred embodiment of the present application, it should be pointed out that, for those skilled in the technology in the art, without departing from the technical principles of the present application, can also make a number of improvements and variations, these improvements and variations should also be considered as the protection scope of the present application.

Claims

1. A phase shifter addressing and calibrating method for extreme scenarios, characterized in that, The method comprises the following steps: Based on the production simulation data, an annual hourly operation data set is constructed; The operation data set is subjected to over-limit judgment of power flow and N-1 safety check, and over-limit scenarios and unsafe scenarios of power flow in the operation data set are identified to construct an extreme candidate scenario set; A representative extreme scenario is selected from the extreme candidate scenario set by using a K-Medoids++ algorithm based on weighted Euclidean distance; Based on the representative extreme scenario, a linearized phase shifter site selection and capacity optimization model is constructed under the conditions of various cost constraints, investment decision constraints and power grid operation constraints, and a mixed integer programming solver is used to solve the phase shifter site selection and capacity optimization model to obtain the site selection and capacity results.

2. The method of claim 1, wherein, The step of constructing the annual hourly operation data set comprises the following steps: Step S11: Input the various parameters required for the production simulation, including the node set. Branch road set Branch roads Resistance and reactance ( , ), branch road Rated capacity ,unit Piecewise linearized power generation cost function ,unit Climbing / descent constraints and ,node At any moment active load ,time Photovoltaic irradiance, time ambient temperature ,time wind speed ; Step S12, hourly calculation of renewable energy production, including photovoltaic production and wind production ; In step S13, a direct current power flow-based production simulation model is constructed, and a mixed integer programming solver is used for hourly calculation and solution to obtain unit output and power flow distribution data under all-year scenarios, thereby constructing the annual hourly operation data set.

3. The method of claim 2, wherein, In step S12, the photovoltaic output calculation formula is: （1） in, For efficiency, For component area, For temperature coefficient, For the cell temperature, It is a node The maximum active power output of the photovoltaic unit; The wind power output calculation formula is: （2） wherein, , , Vcut, Vrated, and Vcut are the cut-in, rated, and cut-out wind speeds, respectively, Prated is the rated power of the wind turbine.

4. The method of claim 2, wherein, In step S13, the objective function of the production simulation model is: （3） wherein, is the unit output, denotes optimizing the unit output such that the objective function is minimized; The constraint condition is: （4） （5） （6） wherein, is the admittance coefficient, , are the voltage phase angles of the nodes , the nodes at the time instant , is the new energy active power of the nodes at the time instant , is the voltage amplitude of the nodes at the time instant , is the lower limit of the node voltage, is the upper limit of the node voltage, and represent the ramp-up and ramp-down constraints of the units , i.e. the maximum power that can be increased and decreased between two adjacent time instants.

5. The method of claim 1, wherein, The step of identifying the over-limit scenarios and unsafe scenarios of power flow in the operation data set comprises the following steps: Step S21, compare the absolute value of the branch flow and the rated capacity of the branch If the branch ( , ) satisfies , where represents the absolute value, the scenario is included in the flow out-of-limit set ;​​ Step S22, among the remaining scenarios excluding the out-of-limit scenarios, identify scenarios that do not satisfy the N-1 principle, and include the identified scenarios into the unsafe set ; Step S23, constructing an extreme candidate scene set , denotes a union set.

6. The method of claim 5, wherein, In the step S22, for each scene The N-1 check operation performed includes: Step S221, removing a branch from the network topology corresponding node , building a new node admittance matrix ;​ In step S222, the direct current power flow equation is solved under the condition that the power injection of each node is kept unchanged: （7） wherein, is the voltage phase angle vector for each node, is the scenario active injection of the node; In step S223, the power flow of each branch after disconnection is calculated: （8） wherein is the voltage phase angle of the bus after the outage of the line , is the bus susceptance and are the voltage phase angles of the buses and after the outage of the line and , respectively. S224, if there is a branch ( , ) satisfying , determine the scenario . In the working condition of cutting off the line , N-1 safety is not satisfied, and the scenario is included in the unsafe scenario set .

7. The method of claim 1, wherein, The step of selecting the representative extreme scenario comprises the following steps: Step S31, constructing node feature vector and node weight wherein, and are the active load and new energy active output of the scene under node respectively; is the set of all nodes; Step S32, set the cluster number as , randomly select 1 scene as the first center in the extreme candidate scene set , , record the current center set = ; Step S33, for any scene , compute the weighted Euclidean distance of the scene to the nearest existing center where, is the existing cluster center, is the weighted Euclidean distance; Step S34, according to the probability , the next center is extracted from the set of extreme candidate scenes , and added to the current center set , until the number of initial centers is obtained , where is the th initial clustering center, is the weighted Euclidean distance from scene to the nearest selected clustering center at present, represents the weighted Euclidean distance from any scene in the set of extreme candidate scenes to the nearest selected clustering center at present; Step S35, assign each scene to the nearest center to form a cluster In each cluster, the sum of weighted Euclidean distances between each sample and other samples is calculated, and the sample with the minimum sum of weighted Euclidean distances is taken as a new cluster center of the corresponding cluster. In step S36, if all the cluster centers no longer change or the algorithm reaches a preset maximum iteration number, the process is stopped; otherwise, the process returns to step S35. Step S37, the central samples at the time of convergence constitute a representative extreme scene set wherein denotes the central sample in the cluster corresponding to the th cluster center.

8. The method of claim 7, wherein, In step S31, the node weight formula is: （9） wherein, is the weight coefficient of the node with new energy access determined according to expert experience.

9. The method of claim 7, wherein, The step S33, the weighted Euclidean distance The calculation formula is: （10） wherein, , is the cluster center of the scenario the normalized load and new energy component under the scenario, , is the cluster center of the scenario the normalized load and new energy component under the scenario, the formula is as follows (11) (12) wherein, and are the active loads and new energy outputs at the nodes in the scenario , and are the mean values of the active loads and new energy outputs at the nodes , and are the standard deviations of the active loads and new energy outputs at the nodes .

10. The method of claim 1, wherein, The step of constructing and solving the linearized phase shifter site selection and capacity optimization model comprises the following steps: In step S41, a target function is constructed by comprehensively considering the investment cost, operation cost and penalty cost caused by overloaded lines and new energy curtailment of the phase shifter: （11） wherein, represents the minimization of the total cost, represents the shifter investment cost, represents the grid operation cost, represents the penalty cost; In step S42, phase shifter investment decision constraints are constructed according to the total number limit of the phase shifter, phase angle discretization selection, branch phase angle expression and upper and lower limits of the phase shifter capacity; In step S43, power grid operation constraints are constructed according to the node power balance, linearized alternating current power flow expression, new energy curtailment quantity constraint and linearized expression of the absolute value of the branch power flow, as well as the conventional power flow upper and lower limit constraint, reference bus constraint and phase angle difference constraint; In step S44, a mixed integer programming solver is used to solve the constructed model to obtain the installation position and installation capacity of the phase shifter.

11. The method of claim 10, wherein, In step S41, the calculation formula of each cost is: （12） wherein, is the annualized fixed investment cost per unit number, is the annualized cost coefficient per unit capacity, is the annualized cost coefficient per unit number, branches shifter capacity, is the total number of branches; （13） wherein, is the generation cost, is the simulated network loss cost, is the total number of scenarios, is the weight of scenario , and is the total generation cost coefficient and network loss cost coefficient of scenario , is the total number of generators, is the total number of segments, is the segment cost coefficient, is the unit segment m output, is the resistance of branch , is the active power flow from node to node in scenario . （14） （15） wherein, and are the scenarios under which the penalty coefficient of the cost of curtailment and the penalty coefficient of the load rate, are the scenarios under which the new energy curtailment is, is the branch exceeding the safe load rate threshold in the scenario , is the overload threshold.

12. The method of claim 10, wherein, In step S42, the formula of the total number limit of the phase shifter, phase angle discretization selection, branch phase angle expression and upper and lower limits of the phase shifter capacity is: （16） wherein, ∈ {0,1}, denotes a branch whether a phase shifter is installed, is the number of phase shifters, and is the minimum and maximum number of phase shifters; （17） wherein, ∈ {0, 1}, represents a branch In the scene whether the first discrete phase shift angle is selected, if a phase shifter is installed in each branch, i.e. = 1, one angle must be selected from the preset discrete angles; if no phase shifter is installed, no angle is allowed to be selected; （18） wherein, is a pre-determined set of angle candidates, equation (18) represents the branch The phase shift angle under the scenario is determined by a discrete variable weighting. （19） wherein, is the capacity of the phase shifter installed on the branch is the capacity of the phase shifter installed on the branch is the maximum capacity of the phase shifter.

13. The method of claim 10, wherein, In step S43, the formula of the node power balance, linearized alternating current power flow expression, new energy curtailment quantity constraint and linearized expression of the absolute value of the branch power flow is: （20） （21） wherein is a set of generators connected at nodes , is a scenario under which generators deliver active power at nodes , is a scenario under which new energy sources deliver active power at nodes , is an active load at nodes , representing active power flowing out of nodes , represents active power flowing in, is a scenario under which generators deliver reactive power at nodes , is a reactive load at nodes , is a scenario under which reactive power flows from nodes into nodes , is a scenario under which reactive power flows from nodes out to nodes ; （22） （23） wherein and denote the phase angles of the voltages at the nodes of the scenario the voltage amplitudes and phase angles of the branches of the scenario and are the conductance and susceptance of the branch Bsh,ijis the parallel susceptance, and are the phase angles of the voltages at the nodes of the scenario is the phase shift angle of the branch of the scenario denotes the phase angle difference taking into account the phase shift angle;​​​​ （24） wherein, is the scenario a reduction in new energy output, is the theoretical maximum of new energy output; （25） wherein is an introduced auxiliary variable.

14. A computer system comprising a memory, a processor and a computer program stored on the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1-13.

Images