Alternating optimization method and system for voltage safety multi-cycle reactive power optimization scheduling

Abstract

The invention relates to an alternating optimization method and system for voltage safety multi-cycle reactive power optimization scheduling. The method comprises the following steps: expressing an optimal coordination problem of static variable resources and dynamic variable resources of a power grid as an optimization model of a multi-cycle reactive power optimization problem; based on the optimization model, specifying an objective function which maximizes the dynamic reactive power reserve of the generator, and establishing corresponding constraint conditions; considering the problem that the number of constraints and variables changes along with accidental conditions and time periods, and introducing consensus control variables representing all contingency between basic conditions and t time periods and discrete control variables representing time coupling constraint limitation into the optimization model; and decomposing the optimization model in which the two control variables are introduced into a plurality of sub-problems, solving the solutions of the plurality of sub-problems, outputting and combining the solutions into the solution of the multi-cycle reactive power optimization problem, and carrying out optimization scheduling according to the solution of the multi-cycle reactive power optimization problem.

Description

Technical Field

[0001] The invention relates to an alternating optimization method and system for voltage safety multi-period reactive power optimization dispatching, belonging to the technical field of voltage reactive power control. Background Art

[0002] Voltage and reactive power control are essential to ensure the proper functioning of the power system in both normal and emergency situations. Reactive resources in the transmission system can be divided into two categories: dynamic and static. Dynamic reactive resources are automatically controlled and can quickly and continuously change their voltage output in response to voltage deviations, regardless of their terminal voltage level. Therefore, when the voltage tends to fall, these resources can increase their reactive output to prevent the voltage from falling further, which may otherwise cause partial or total voltage collapse. Dynamic reactive resources include synchronous generators / condensers and flexible AC transmission system equipment. Inverter-based generation resources, such as wind turbines (Type III and Type IV) and solar photovoltaic plants, can also provide dynamic reactive support and voltage control. Static reactive resources include shunt devices, such as capacitor banks and switchable reactors, and tap-changing transformers. The reactive output of shunt devices is proportional to the square of their terminal voltage. Therefore, when the terminal voltage drops, their reactive output drops rapidly, which greatly limits their ability to prevent voltage instability.

[0003] In order to ensure the safe operation of the transmission system, it is necessary to always maintain an appropriate combination of dynamic and static voltage resources. Since dynamic reactive resources have the ability to respond quickly, which is crucial in emergencies, more reactive power needs to be reserved from these resources. That is, in normal operation, the reactive power output of dynamic reactive resources should be minimized to ensure that there is enough dynamic reactive power to deal with emergencies. To increase reactive reserves, system operators can switch parallel equipment to a pre-emergency state to reduce the reactive output of dynamic equipment. In some cases, operators may need to bring more dynamic equipment online, for example, dispatching unqualified generators, to obtain the reactive power reserves required to maintain operational safety.

[0004] In practical applications, there are multiple ways to dispatch variable resources in the transmission network. In power systems with advanced voltage control schemes, including secondary and tertiary voltage regulation, a central controller distributes set points to all designated components with the goal of minimizing losses and control costs while maintaining the voltage of all buses at appropriate levels. Switch shunts and transformer taps are controlled so that synchronous generators retain their reactive power regulation capabilities to cope with transient emergencies.

[0005] The multi-period optimal reactive power dispatch (MP-ORPD) problem plays a vital role in power systems. The main challenge is that the problem scale grows exponentially with the increase in the number of critical emergencies and time periods. This complexity not only increases the computational difficulty, but also makes the solution process more time-consuming, especially in practical large-scale power systems. Therefore, it is urgent to develop efficient solution algorithms to minimize the computational time and improve the traceability of the problem. To address this challenge, some studies have proposed a variety of solution methods that draw on the solution strategies of multi-period or multi-scenario optimization problems similar to the MP-ORPD problem. For example, the Benders decomposition technique is widely used in the solution of such problems. By dividing the optimization problem into smaller sub-problems, the Benders decomposition technique can not only simplify the calculation process, but also effectively utilize parallel computing resources, thereby improving the solution efficiency.

[0006] Benders is used to solve linear programming problems with special block structures and complex variables, such as problems involving multiple scenarios that arise in stochastic programming. The same concept is generalized to nonlinear programming. However, if one of the subproblems is infeasible, the generalized Benders decomposition may not converge. Some scholars have proposed a rolling process method to deal with problems similar to MP-ORPD without considering any emergencies. However, as the time period increases, the traceability of the problem becomes questionable, which limits the number of time periods that can be included. Other scholars have proposed heuristic search methods, including evolutionary algorithms and particle swarm optimization methods. However, the parameter tuning of the heuristic search method has a great impact on the optimization performance. Summary of the invention

[0007] In order to solve the above problems existing in the prior art, the present invention proposes an alternating optimization method and system for voltage safety multi-cycle reactive power optimization scheduling.

[0008] The technical solution of the present invention is as follows:

[0009] On the one hand, the present invention proposes an alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling, comprising the following steps:

[0010] The optimal coordination problem between the static variable resources and the dynamic variable resources of the power grid is expressed as an optimization model of the multi-period reactive power optimization problem.

[0011] Based on the optimization model, the objective function to maximize the dynamic reactive power reserve of the generator is specified, and the corresponding constraints are established;

[0012] Considering the problem that the number of constraints and variables varies with the contingency situation and time period, the optimization model introduces the consensus control variables representing the relationship between the basic situation and all the contingencies in time period t, and the discrete control variables representing the constraints subject to time coupling.

[0013] The optimization model introducing two control variables is decomposed into multiple sub-problems, and the solutions of the multiple sub-problems are obtained. The outputs are merged into the solution of the multi-period reactive power optimization problem, and the optimal scheduling is performed according to the solution of the multi-period reactive power optimization problem.

[0014] As a preferred implementation, after introducing the consensus control variable representing the basic situation and all the contingencies in the time period t, and the discrete control variable representing the constraints of time coupling into the optimization model, the optimization model is expressed as follows:

[0015]

[0016] Subject to the following constraints:

[0017]

[0018] Among them, x is the set of system state variables, u is the set of control variables, C t = {0, 1, 2, ...} is the critical contingency index set for each time period t, C t =0 means a system without randomness; T is a set of time periods; φ c is the objective function of the optimal power flow; is the maximum number of switch actions allowed by controller i within the time range, and U is the set of control indicators corresponding to the two discrete variables of shunt susceptance and transformer tap ratio; is the vector of all control variables of the contingency c at time t; y t represents the consensus control variable between the base case and all contingencies at t, z t represents the discrete control variable subject to time coupling constraints; γ t is the control variable and consensus control variable y t α is the first penalty parameter for consistency, and α is the second penalty parameter for encouraging consensus control variables to be close to effective discrete controls.

[0019] As a preferred implementation, the process of decomposing the optimization model introducing two control variables into multiple sub-problems and obtaining solutions to the multiple sub-problems includes two stages, wherein:

[0020] In the first stage, the discrete control variables are fixed, the optimization model is decomposed into multiple sub-problems, and all time periods of the system state variable set, control variable set, and consensus control variable set are solved to ignore the time coupling constraints;

[0021] In the second stage, the system state variables, control variables, and consensus control variables are fixed and the discrete control variables are optimized to impose constraints on discrete control changes within the time horizon;

[0022] In the first stage, convergence is achieved by updating the first penalty parameter, and in the second stage, convergence is achieved by updating the second penalty parameter.

[0023] As a preferred implementation, the processing process of the first stage is specifically as follows:

[0024] Fixed discrete control variable z, the optimization model is decomposed into T sub-problems, and the sub-problems with time period t are as follows:

[0025]

[0026] obey:

[0027]

[0028] For every c∈C t ,untie:

[0029]

[0030] obey:

[0031]

[0032] Get the solution

[0033] Update consensus control variable y t :

[0034]

[0035] Update the first penalty parameter γ t :

[0036]

[0037] Among them, β>1, ρ is the control variable The consensus control variable y from the previous iteration t The score of deviation;

[0038] After each subproblem converges, the updated control variables, consensus control variables, and the values ​​of the first penalty parameter are returned.

[0039] As a preferred implementation, the processing process of the second stage is specifically as follows:

[0040] Initializing system state variables, control variables, consensus control variables, discrete control variables, a first penalty parameter, and a second penalty parameter, wherein the control variables, consensus control variables, and discrete control variables are initialized to a tentative control plan;

[0041] Get the values ​​of the consensus control variables and the first penalty parameter updated in the first phase;

[0042] Collect solutions to multiple subproblems and update the discrete control variables and the second penalty parameter as follows:

[0043] Calculate the deviation of the consensus control variable updated in this iteration from the discrete control variable calculated after the previous iteration;

[0044]

[0045] Among them, κ>1;

[0046] If the deviation ν>0, the discrete control variable is updated by the following formula:

[0047]

[0048] obey:

[0049]

[0050] Let the solution of t∈T be the discrete control variable z t ; The second penalty parameter is updated as follows:

[0051]

[0052] If the deviation ν≤0, keep the same discrete control variables and second penalty function as the previous iteration;

[0053] The second stage stops when the error is less than the preset value s, where the error is:

[0054] E(u,y,z):=max{E1(u,y),E2(y,z)};

[0055] in:

[0056]

[0057] Among them, m is the consensus control variable y t size.

[0058] On the other hand, the present invention also proposes an alternating optimization system for voltage safety multi-cycle reactive power optimization dispatch, comprising:

[0059] A model building module is used to express the optimal coordination problem of static variable resources and dynamic variable resources of the power grid as an optimization model of a multi-period reactive power optimization problem;

[0060] The target constraint setting module specifies the objective function that maximizes the dynamic reactive power reserve of the generator based on the optimization model and establishes corresponding constraint conditions;

[0061] Model optimization module, used to consider the problem that the number of constraints and variables varies with the contingency situation and time period, introduces the consensus control variables representing the relationship between the basic situation and all the contingencies in time period t, and the discrete control variables representing the constraints subject to time coupling constraints into the optimization model;

[0062] The control optimization module is used to decompose the optimization model introducing two control variables into multiple sub-problems, and obtain solutions to the multiple sub-problems, output and merge them into the solution of the multi-period reactive power optimization problem, and perform optimization scheduling according to the solution of the multi-period reactive power optimization problem.

[0063] As a preferred implementation, after introducing the consensus control variable representing the basic situation and all the contingencies in the time period t, and the discrete control variable representing the constraints of time coupling into the optimization model, the optimization model is expressed as follows:

[0064]

[0065] Subject to the following constraints:

[0066]

[0067] Where x is the set of system state variables, u is the set of control variables, C = {0, 1, 2, ...} is the set of critical contingency indicators for each time period t, t = 0 represents a system without contingency; T is the set of time periods; φ c is the objective function of the optimal power flow; is the maximum number of switch actions allowed by controller i within the time range, and U is the set of control indicators corresponding to the two discrete variables of shunt susceptance and transformer tap ratio; is the vector of all control variables of the contingency c at time t; y t represents the consensus control variable between the base case and all contingencies at t, z t represents the discrete control variable subject to time coupling constraints; γ t is the control variable and consensus control variable y t α is the first penalty parameter for consistency, and α is the second penalty parameter for encouraging consensus control variables to be close to effective discrete controls.

[0068] As a preferred implementation, the process of decomposing the optimization model introducing two control variables into multiple sub-problems and obtaining solutions to the multiple sub-problems includes two stages, wherein:

[0069] In the first stage, the discrete control variables are fixed, the optimization model is decomposed into multiple sub-problems, and all time periods of the system state variable set, control variable set, and consensus control variable set are solved to ignore the time coupling constraints;

[0070] In the second stage, the system state variables, control variables, and consensus control variables are fixed and the discrete control variables are optimized to impose constraints on discrete control changes within the time horizon;

[0071] In the first stage, convergence is achieved by updating the first penalty parameter, and in the second stage, convergence is achieved by updating the second penalty parameter.

[0072] As a preferred implementation, the processing process of the first stage is specifically as follows:

[0073] Fixed discrete control variable z, the optimization model is decomposed into T sub-problems, and the sub-problems with time period t are as follows:

[0074]

[0075] obey:

[0076]

[0077] For every c∈C t ,untie:

[0078]

[0079] obey:

[0080]

[0081] Get the solution

[0082] Update consensus control variable y t :

[0083]

[0084] Update the first penalty parameter γ t :

[0085]

[0086] Among them, β>1, ρ is the control variable The consensus control variable y from the previous iteration t The score of deviation;

[0087] After each subproblem converges, the updated control variables, consensus control variables, and the values ​​of the first penalty parameter are returned.

[0088] As a preferred implementation, the processing process of the second stage is specifically as follows:

[0089] Initializing system state variables, control variables, consensus control variables, discrete control variables, a first penalty parameter, and a second penalty parameter, wherein the control variables, consensus control variables, and discrete control variables are initialized to a tentative control plan;

[0090] Get the values ​​of the consensus control variables and the first penalty parameter updated in the first phase;

[0091] Collect solutions to multiple subproblems and update the discrete control variables and the second penalty parameter as follows:

[0092] Calculate the deviation of the consensus control variable updated in this iteration from the discrete control variable calculated after the previous iteration;

[0093]

[0094] Among them, κ>1;

[0095] If the deviation ν>0, the discrete control variable is updated by the following formula:

[0096]

[0097] obey:

[0098]

[0099] Let the solution of t∈T be the discrete control variable z t ; The second penalty parameter is updated as follows:

[0100]

[0101] If the deviation ν≤0, keep the same discrete control variables and second penalty function as the previous iteration;

[0102] The second stage stops when the error is less than the preset value s, where the error is:

[0103] E(u,y,z):=max{E1(u,y),E2(y,z)};

[0104] in:

[0105]

[0106] Among them, m is the consensus control variable y t size.

[0107] On the other hand, the present invention also proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling as described in any embodiment of the present invention is implemented.

[0108] On the other hand, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling as described in any embodiment of the present invention.

[0109] Additional aspects and advantages of the present invention will be set forth in the following description, and some of them will be apparent from the description, or may be understood by practicing the present invention. In addition, the various aspects and advantages of the present invention may be realized and obtained by the method steps and combinations particularly pointed out in the appended claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0110] Figure 1 The figure is a schematic diagram of the method flow of the first embodiment of the present invention. DETAILED DESCRIPTION

[0111] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0112] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0113] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0114] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0115] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0116] Embodiment 1:

[0117] In order to overcome the shortcomings of the above algorithms, this embodiment proposes a new method that can handle the size of the MP-ORPD problem. The MP-ORPD problem is divided into a set of smaller sub-problems using a consensus-based decomposition technique, which can be solved independently and in parallel. The problem for each time period is solved using optimization problems with global variables (OPGV). The algorithm is extended to multi-period reactive scheduling problems with time coupling constraints and discrete control constraints. Discrete variables are treated as continuous variables, and penalty terms are added to the objective function to encourage continuous variables to approach valid discrete values ​​obtained through mixed-integer linear programming (MILP) subproblems.

[0118] See also Figure 1 The present embodiment provides an alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling, which specifically includes the following steps:

[0119] S100, expressing the optimal coordination problem between the static variable resources and the dynamic variable resources of the power grid as an optimization model of a multi-period reactive power optimization problem;

[0120] The goal of the day-ahead variable scheduling process is to dispatch reactive resources so that the grid has sufficient dynamic reactive reserves under the most critical emergencies. The inputs require substation-level active and reactive power forecasts, generation plans, network topology, critical emergencies and critical bus sets in different time periods, power outage plans, etc. This embodiment identifies critical emergencies and critical buses in the network through bus voltage sensitivity, and the output of the day-ahead variable scheduling planning algorithm is the optimal scheduling of static and dynamic variable resources. Specifically, the output includes set points for switch shunts, transformer taps, and regulated bus voltage values.

[0121] This embodiment expresses the optimal coordination of static variable resources and dynamic variable resources as the MP-ORPD problem. The control variables of this problem are the voltage amplitude of the voltage-stabilized bus, the susceptance of the switch shunt, and the value of the transformer tap ratio. The state variables are the voltage amplitude of the unstabilized bus, the voltage phase angle of each bus, and the reactive power of each generator. More specifically, the mathematical model of the problem is as follows:

[0122]

[0123] Formula (1) obeys formula (2):

[0124]

[0125] In the formula, x is the set of system state variables, u is the set of control variables, C t = {0, 1, 2, ...} is the critical contingency index set of each time period t, c = 0 means a system without contingency, T is the time period set, φ c is the objective function of the optimal power flow; is the maximum number of switching actions allowed by controller i within the time range, and U is the set of control indicators corresponding to the two discrete variables of shunt susceptance and transformer tap ratio. For illustration, is the vector of all control variables of the contingency c at time t. The discrete control variables are The continuous control variables are

[0126] S200, the objective function in the basic formula (1) can represent any objective function related to the OPF problem. The main goal of this embodiment is to maximize the dynamic reactive power reserve of the generator. Then, formulate the objective function:

[0127]

[0128] in:

[0129]

[0130] where |.| is the cardinality of the set, is the set of in-service branches (transmission lines, transformers, phase shifters), is the set of active generators in the network, is the critical busbar set, is the desired upper limit of the critical bus voltage, and Δv is a positive scalar. l ,a q and a k is the weight. is the average reactive power loss of the line. The term encourages the generator to minimize its reactive power. For the critical bus, the regularization term The voltage amplitude at the critical bus is forced to approach the voltage upper limit. By increasing the voltage at the base case operating point, the reactive power margin at the critical bus is increased. This practical heuristic method encourages capacitor banks (if any) near the critical bus to be turned on to provide reactive power to the critical bus load, increase the voltage at the critical bus, and thus improve the reactive power margin.

[0131] The constraints (2b) of the problem include equality constraints, which represent the power balance equations for each contingency in each time period; and inequality constraints, which represent the generator reactive power limits, thermal load limits, and voltage amplitude limits. The coupling constraints (2c) enforce consensus on the control variables in the same time period. In the MP-ORPD model, static reactive devices are shunt capacitors / reactors and transformer tap changers that are remotely controlled by the system operator. These devices cannot be operated in automatic control mode in response to voltage changes on the monitored bus. Therefore, if they are to be used for post-accident control, the operator needs to switch them immediately after the accident. Most reliability standards specify that manual intervention by the system operator should not be considered in voltage stability assessment and improvement, because manual remote control may not be fast enough to avoid voltage instability. Therefore, in the MP-ORPD approach, shunt devices and transformer tap changers are considered as pre-emergency control actions and are therefore fixed for any considered emergency situation. Therefore, the optimization problem requires finding preventive control measures that guarantee system safety in the event of an emergency.

[0132] The temporal coupling constraint (2d) coordinates the operation of static variable resources to limit the number of set point changes within the time range. The MP-ORPD problem (1) has a special block structure, and there are coupling constraints between different blocks. If the coupling constraints are removed, the blocks are decoupled.

[0133] S300. It can be seen from the optimization model (1) that the number of constraints and variables increases with the increase of the contingency situation and time period. Therefore, the scale of the optimization problem increases significantly, and the computational cost of its solution becomes expensive. This embodiment considers the problem that the number of constraints and variables varies with the contingency situation and time period, and introduces a consensus control variable representing the relationship between the basic situation and all the contingencies in the t time period and a discrete control variable representing the time coupling constraint into the optimization model.

[0134] S400, decomposing the optimization model introducing two control variables into multiple sub-problems, and obtaining solutions to the multiple sub-problems, outputting and merging the solutions to the multi-period reactive power optimization problem, and performing optimization scheduling according to the solutions to the multi-period reactive power optimization problem.

[0135] In one embodiment, the consensus control variable representing the relationship between the basic situation and all the contingencies in the time period t and the discrete control variable representing the constraints of time coupling are introduced into the optimization model. After adding the new variables, the optimization problem is as follows:

[0136]

[0137] Among them, y t represents the consensus control variable between the base case and all contingencies at t, z trepresents a discrete control variable subject to time-coupled constraints.

[0138] This embodiment replaces constraints (6b) and (6c) by using a penalty function. Formula (3) can be restated as:

[0139]

[0140] Subject to the following constraints:

[0141]

[0142] Among them, γ t is the control variable and consensus control variable y t The first penalty parameter is the consistency between the two, and α is the second penalty parameter that encourages the consensus control variable to be close to the effective discrete control. This discrete control is feasible with respect to the time coupling constraint. As mentioned earlier, the discrete variables are treated as continuous variables. The 1-norm and 2-norm in the objective function play different roles. The 2-norm in (7) encourages the control variable at time t to be Close to consensus control variable y t 1 norm encourages consensus control variable y t Discrete control variables z that approach time coupling constraints t Penalty terms 1 and 2 together penalize discrete control variables that are far from valid discrete values. This greatly reduces the impact of final rounding to improve solution accuracy.

[0143] In one embodiment, the process of decomposing the optimization model introducing two control variables into multiple sub-problems and obtaining solutions to the multiple sub-problems includes two stages, wherein:

[0144] When the variable zt is fixed, problem (7) can be decomposed into T subproblems. If the consensus control variable y is fixed, then each of these subproblems can be decomposed into |C t | subproblem. This embodiment solves this problem in two stages. The first stage (internal algorithm) ignores the time coupling constraint by fixing the discrete variable z and solving for all time periods with respect to x, u, and y. Then, the second stage (external algorithm) fixes x, u, and y and optimizes z to impose constraints on discrete control changes within the time range. The internal and external algorithms update the penalty parameter γ t and α to achieve convergence.

[0145] In one embodiment, the processing process of the first stage is as follows:

[0146] Fixed discrete control variable z, the optimization model is decomposed into T sub-problems, and the sub-problems with time period t are as follows:

[0147]

[0148] obey:

[0149]

[0150] The problem (8) for time period t is called the optimization problem with global variables (OPGV). The global variable here is consensus control. This can be decomposed into |Ct| subproblems and solved approximately once by the following steps:

[0151] 1) For every c∈C t ,untie:

[0152]

[0153] obey:

[0154]

[0155] Get the solution

[0156] 2) Update the consensus control variable y t :

[0157]

[0158] 3) Update the first penalty parameter γ t :

[0159]

[0160] Among them, β>1, ρ is the control variable The consensus control variable y from the previous iteration t Deviation score:

[0161]

[0162] The optimization algorithm is a fast local convergence decomposition algorithm for smooth problems. The internal algorithm ensures consistent control between the base case and all the contingencies in time period t. The internal algorithm can be viewed as a variable z t and penalty parameters α and γ t The function F returns the updated control variable, consensus control variable and the value of the first penalty parameter, that is, in,

[0163] In one embodiment, the processing process of the second stage is as follows:

[0164] The temporal coupling constraint (6d) limits the switching behavior of discrete variable resources within a time range. For example, the system operator may allow the shunt capacitor to be switched on or off only twice a day. The external algorithm updates the discrete control variable z t The value of the second penalty parameter α satisfies the time coupling constraint. The external algorithm first initializes the variables and the penalty parameter γ t and α. The variables u, y, and z are initialized to some tentative control plans. Then, the internal algorithm is called to approximately solve the T subproblems (8) to update and α t , which can be performed in parallel to reduce computation time. The external algorithm collects the solutions to the T subproblems and updates the discrete variables z t and α are as follows:

[0165] 1) Calculate the consensus control variables updated in this iteration From the discrete control variable z t Deviation calculated to the previous iteration:

[0166]

[0167] Among them, κ>1.

[0168] 2) If the deviation ν > 0, the discrete control variables are updated by solving the following simple MILP problem:

[0169]

[0170] obey:

[0171]

[0172] Let the solution of t∈T be; the second penalty parameter is updated as follows:

[0173]

[0174] 3) If the deviation ν≤0, keep the same discrete control variables and second penalty function as the previous iteration;

[0175] Variable z t are the target values ​​of discrete control because they satisfy the time coupling constraints and are close to the consensus control variables y t .for Continuous control in the consensus control variable y t is the target value for time period t. Therefore, the second stage proposed in this embodiment stops when the error is less than the preset value s, where the error is:

[0176]

[0177] in:

[0178]

[0179] Among them, m is the consensus control variable y t The optimal solution of the MP-ORPD problem is: t,i , For continuous control In Algorithm 1, for simplicity, it is assumed that in the next iteration, u, y, and γ become the values ​​of u, y, and γ, respectively.

[0180] Based on this embodiment, an optimization method for dispatching reactive resources and improving the voltage security of the transmission system is proposed. This is achieved by maximizing the dynamic reactive reserve and improving the reactive margin of key buses. The algorithm achieves these goals by coordinating static and dynamic variable resources while considering the most critical contingency in the optimization problem. The reactive power generation of the generator is minimized and the dynamic reactive reserve is maximized.

[0181] Embodiment 2:

[0182] This embodiment proposes an alternating optimization system for voltage safety multi-period reactive power optimization scheduling, including:

[0183] A model building module, used to express the optimal coordination problem of the static variable resources and the dynamic variable resources of the power grid as an optimization model of the multi-period reactive power optimization problem; this module is used to implement the function of step S100 in the first embodiment, which will not be described in detail here;

[0184] A target constraint setting module specifies an objective function for maximizing the dynamic reactive power reserve of the generator based on the optimization model, and establishes corresponding constraint conditions; this module is used to implement the function of step S200 in the first embodiment, and will not be described in detail here;

[0185] A model optimization module is used to consider the problem that the number of constraints and variables varies with the contingency situation and time period, and introduces a consensus control variable representing the relationship between the basic situation and all the contingencies in the time period t, and a discrete control variable representing the time coupling constraint into the optimization model; this module is used to implement the function of step S300 in the first embodiment, and will not be described in detail here;

[0186] The control optimization module is used to decompose the optimization model introducing two control variables into multiple sub-problems, and obtain solutions to the multiple sub-problems, output and merge them into the solution of the multi-period reactive power optimization problem, and perform optimization scheduling according to the solution of the multi-period reactive power optimization problem; this module is used to implement the function of step S400 in Example 1, which will not be repeated here.

[0187] Embodiment three:

[0188] This embodiment proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling as described in any embodiment of the present invention is implemented.

[0189] Embodiment 4:

[0190] This embodiment proposes a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling as described in any embodiment of the present invention is implemented.

[0191] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0192] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0193] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0194] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0195] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling, characterized in that: The following steps are involved: The optimal coordination problem between the static variable resources and the dynamic variable resources of the power grid is expressed as an optimization model of the multi-period reactive power optimization problem. Based on the optimization model, the objective function to maximize the dynamic reactive power reserve of the generator is specified, and the corresponding constraints are established; Considering the problem that the number of constraints and variables varies with the contingency situation and time period, the optimization model introduces the consensus control variables representing the relationship between the basic situation and all the contingencies in time period t, and the discrete control variables representing the constraints subject to time coupling. The optimization model introducing two control variables is decomposed into multiple sub-problems, and the solutions of the multiple sub-problems are obtained. The outputs are merged into the solution of the multi-period reactive power optimization problem, and the optimal scheduling is performed according to the solution of the multi-period reactive power optimization problem.

2. The alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling according to claim 1 is characterized in that: After introducing the consensus control variable representing the basic situation and all the contingencies in the t time period and the discrete control variable representing the time coupling constraint into the optimization model, the optimization model is expressed as follows: Subject to the following constraints: Among them, x is the set of system state variables, u is the set of control variables, C t = {0, 1, 2, ...} is the critical contingency index set of each time period t, C t＝0 represents a system without randomness; T is a set of time periods; φ c is the objective function of the optimal power flow; is the maximum number of switch actions allowed by controller i within the time range, and U is the set of control indicators corresponding to the two discrete variables of shunt susceptance and transformer tap ratio; is the vector of all control variables of the contingency c at time t; y t represents the consensus control variable between the base case and all contingencies at t, z t represents the discrete control variable subject to time coupling constraints; γ t is the control variable and consensus control variable y t α is the first penalty parameter for consistency, and α is the second penalty parameter for encouraging consensus control variables to be close to effective discrete controls.

3. The alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling according to claim 2 is characterized in that: The process of decomposing the optimization model with two control variables into multiple sub-problems and obtaining solutions to the multiple sub-problems includes two stages, wherein: In the first stage, the discrete control variables are fixed, the optimization model is decomposed into multiple sub-problems, and all time periods of the system state variable set, control variable set, and consensus control variable set are solved to ignore the time coupling constraints; In the second stage, the system state variables, control variables, and consensus control variables are fixed and the discrete control variables are optimized to impose constraints on discrete control changes within the time horizon; In the first stage, convergence is achieved by updating the first penalty parameter, and in the second stage, convergence is achieved by updating the second penalty parameter.

4. The alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling according to claim 3 is characterized in that: The processing process of the first stage is as follows: Fixed discrete control variable z, the optimization model is decomposed into T sub-problems, and the sub-problems with time period t are as follows: obey: For every c∈C t ,untie: obey: Get the solution Update consensus control variable y t : Update the first penalty parameter γ t : Among them, β>1, ρ is the control variable The consensus control variable y from the previous iteration t The score of deviation; After each subproblem converges, the updated control variables, consensus control variables, and the values ​​of the first penalty parameter are returned.

5. The alternating optimization method for voltage safety multi-cycle reactive power optimization scheduling according to claim 3 is characterized in that: The processing process of the second stage is as follows: Initializing system state variables, control variables, consensus control variables, discrete control variables, a first penalty parameter, and a second penalty parameter, wherein the control variables, consensus control variables, and discrete control variables are initialized to a tentative control plan; Get the values ​​of the consensus control variables and the first penalty parameter updated in the first phase; Collect solutions to multiple subproblems and update the discrete control variables and the second penalty parameter as follows: Calculate the deviation of the consensus control variable updated in this iteration from the discrete control variable calculated after the previous iteration; Among them, κ>1; If the deviation ν>0, the discrete control variable is updated by the following formula: obey: Let the solution of t∈T be the discrete control variable z t ; The second penalty parameter is updated as follows: If the deviation ν≤0, keep the same discrete control variables and second penalty function as the previous iteration; The second stage stops when the error is less than the preset value s, where the error is: E(u,y,z):=max{E1(u,y),E2(y,z)}; in: Among them, m is the consensus control variable y t size.

6. An alternating optimization system for voltage safety multi-cycle reactive power optimization dispatching, characterized in that: include: A model building module is used to express the optimal coordination problem of static variable resources and dynamic variable resources of the power grid as an optimization model of a multi-period reactive power optimization problem; The target constraint setting module specifies the objective function that maximizes the dynamic reactive power reserve of the generator based on the optimization model and establishes corresponding constraint conditions; Model optimization module, used to consider the problem that the number of constraints and variables varies with the contingency situation and time period, introduces the consensus control variables representing the relationship between the basic situation and all the contingencies in time period t, and the discrete control variables representing the constraints subject to time coupling constraints into the optimization model; The control optimization module is used to decompose the optimization model introducing two control variables into multiple sub-problems, and obtain solutions to the multiple sub-problems, output and merge them into the solution of the multi-period reactive power optimization problem, and perform optimization scheduling according to the solution of the multi-period reactive power optimization problem.

7. The alternating optimization system for voltage safety multi-cycle reactive power optimization dispatch according to claim 6 is characterized in that: After introducing the consensus control variable representing the basic situation and all the contingencies in the t time period and the discrete control variable representing the time coupling constraint into the optimization model, the optimization model is expressed as follows: Subject to the following constraints: Among them, x is the set of system state variables, u is the set of control variables, C t = {0, 1, 2, ...} is the critical contingency index set of each time period t, C t＝0 represents a system without randomness; T is a set of time periods; φ c is the objective function of the optimal power flow; is the maximum number of switch actions allowed by controller i within the time range, and U is the set of control indicators corresponding to the two discrete variables of shunt susceptance and transformer tap ratio; is the vector of all control variables of the contingency c at time t; y t represents the consensus control variable between the base case and all contingencies at t, z t represents the discrete control variable subject to time coupling constraints; γ t is the control variable and consensus control variable y t α is the first penalty parameter for consistency, and α is the second penalty parameter for encouraging consensus control variables to be close to effective discrete controls.

8. The alternating optimization system for voltage safety multi-cycle reactive power optimization dispatch according to claim 7 is characterized in that: The process of decomposing the optimization model with two control variables into multiple sub-problems and obtaining solutions to the multiple sub-problems includes two stages, wherein: In the first stage, the discrete control variables are fixed, the optimization model is decomposed into multiple sub-problems, and all time periods of the system state variable set, control variable set, and consensus control variable set are solved to ignore the time coupling constraints; In the second stage, the system state variables, control variables, and consensus control variables are fixed and the discrete control variables are optimized to impose constraints on discrete control changes within the time horizon; In the first stage, convergence is achieved by updating the first penalty parameter, and in the second stage, convergence is achieved by updating the second penalty parameter.

9. The alternating optimization system for voltage safety multi-cycle reactive power optimization dispatch according to claim 8 is characterized in that: The processing process of the first stage is as follows: Fixed discrete control variable z, the optimization model is decomposed into T sub-problems, and the sub-problems with time period t are as follows: obey: For every c∈C t ,untie: obey: Get the solution Update consensus control variable y t : Update the first penalty parameter γ t : Among them, β>1, ρ is the control variable The consensus control variable y from the previous iteration t The score of deviation; After each subproblem converges, the updated control variables, consensus control variables, and the values ​​of the first penalty parameter are returned.

10. The alternating optimization system for voltage safety multi-cycle reactive power optimization dispatching according to claim 8, characterized in that: The processing process of the second stage is as follows: Initializing system state variables, control variables, consensus control variables, discrete control variables, a first penalty parameter, and a second penalty parameter, wherein the control variables, consensus control variables, and discrete control variables are initialized to a tentative control plan; Get the values ​​of the consensus control variables and the first penalty parameter updated in the first phase; Collect solutions to multiple subproblems and update the discrete control variables and the second penalty parameter as follows: Calculate the deviation of the consensus control variable updated in this iteration from the discrete control variable calculated after the previous iteration; Among them, κ>1; If the deviation ν>0, the discrete control variable is updated by the following formula: obey: Let the solution of t∈T be the discrete control variable z t ; The second penalty parameter is updated as follows: If the deviation ν≤0, keep the same discrete control variables and second penalty function as the previous iteration; The second stage stops when the error is less than the preset value s, where the error is: E(u,y,z):=max{E1(u,y),E2(y,z)}; in: Among them, m is the size of the consensus control variable yt.

Images