WO2021077596A1

WO2021077596A1 - Group-based online optimization method for optimal power grid topology taking predicted accident sets into consideration

Info

Publication number: WO2021077596A1
Application number: PCT/CN2019/127465
Authority: WO
Inventors: 王蕾; 江晓东
Original assignee: 山东理工大学
Priority date: 2019-10-24
Filing date: 2019-12-23
Publication date: 2021-04-29
Also published as: CN110783913A; CN110783913B

Abstract

A group-based online optimization method for optimal power grid topology taking predicted accident sets into consideration. Under the constraint of the number of lines that are allowed to be connected and disconnected, a set of optimal connection and disconnection line schemes taking a mathematical model of the optimal connection and disconnection line, with an enhanced voltage stability, of predicted accident sets into consideration are found. The schemes satisfy the following conditions: maximizing a load margin of an expected ground-state power system after disconnection, such that load margins of all predicted accidents and the expected ground-state power system can satisfy threshold value requirements, and the expected ground-state power system can satisfy safe operation constraints after disconnection. The problem of online connection and disconnection line identification of power system voltage stability enhancement control is solved through four stages of evaluation, predicted accident grouping and predicted accident representative connection and disconnection line identification, expected ground-state power system connection and disconnection line identification, and verification, and finally, connection and disconnection line schemes satisfying a voltage stability load margin threshold value of an expected ground-state power system and predicted accidents sets are provided in order to enhance the static voltage stability of the power system.

Description

Group-based online optimization method for optimal power grid topology taking into account expected accident sets

Technical field

The invention relates to a group-based online optimization method for the optimal power grid topology taking into account anticipated accident sets, and belongs to the technical field of power system online voltage stability enhanced control.

Background technique

The voltage stability of the power system is one of the important indicators to maintain the normal and stable operation of the power system. With the increasing expansion of the power grid, the high-density and large-scale access of renewable energy, the increasing demand for power, the operation of the power system is complex and changeable, and the power system often runs near the stability boundary. At this time, once a failure occurs, the power system Stable operation will be destroyed, which can easily cause voltage instability in the power system, and even lead to large-scale power outages, resulting in huge economic losses. In order to enhance the voltage stability of the power system, commonly used control measures are the addition of reactive power compensation devices, transformer tap adjustment control, power generation plan rescheduling, load shedding, etc. A large number of simulations and practical applications have proved that these methods are very effective in improving the voltage stability of the power system, especially load shedding measures. When the power system is on the verge of voltage instability, removing part of the load can alleviate the power transmission of the power system. However, the implementation of this measure will cause the load to lose power, which violates the basic principle of continuous and reliable power supply of the power system.

The current power system is facing large-scale new energy access, and the traditional day-ahead control analysis results cannot adapt to the strong random characteristics of the current power system. A set of control analysis methods that can be applied online must be developed. The usual voltage stability control only considers the voltage stability of the ground-state power system, and does not consider the voltage stability of the expected accident set. Therefore, it is necessary to propose a voltage stability control measure and online calculation method that is more economical and considers the expected accident set. . Mathematically speaking, only applying the measures of interrupting the transmission line to enhance the voltage stability of the power system, its mathematical model is a nonlinear mixed integer programming problem. At this stage, there is no effective solution method, and the calculation is difficult. If the expected accident set of the expected ground state power system is taken into account, the problem is more difficult to solve. The main difficulties in solving the problem are: (1) The dimension of the integer variable increases: the disconnected line can be regarded as an integer variable, "0" It can indicate that the line is disconnected, and "1" can indicate that the line is running on the network; the predicted accident can also be regarded as an integer variable of the mathematical model of the problem, so the dimension of the integer variable increases, and the occurrence of the predicted accident makes nonlinear algebra The equation constraints (that is, the active power and reactive power balance equations of the system are expected to change after the accident), which also increases the difficulty of solving the problem; (2) The essential effect of nonlinear characteristics is large: First, the change of line parameters has a great impact on the voltage. The influence of stability margin is non-linear. The breaking of the line is the change of the line impedance parameter from the original value to infinity, but the voltage stability load margin cannot be estimated by the change of the parameter; in addition, even if it can be found to make a certain expected accident load margin The optimal breaking line that meets the threshold requirement, but it may also cause the load margin of other anticipated accidents to be severely reduced, and may even be lower than the threshold requirement.

In summary, in order to overcome the above-mentioned difficulties, reduce the amount of calculation for solving the problem, increase the calculation speed, and adapt to the online application of power system enhanced control, it is necessary to provide a fast and efficient way to account for the optimal set of accidents. Online optimization method of power grid topology.

Summary of the invention

In view of the shortcomings of the above methods, the present invention proposes a group-based online optimization method for the optimal grid topology taking into account the expected accident set, which can give the voltage stability load margin threshold that meets the expected ground state power system and the expected accident set. The circuit break scheme enhances the static voltage stability of the power system.

The technical solutions adopted by the present invention to solve its technical problems are:

On the one hand, the embodiment of the present invention provides a group-based online optimization method for the optimal grid topology taking into account the expected accident set. Under the constraint of the number of interrupted lines, the optimal enhanced voltage stability that takes into account the expected accident set is found. A set of optimal circuit breaker schemes based on the mathematical model of the circuit breaker. The optimal circuit breaker scheme satisfies the following conditions: maximizing the load margin of the expected ground state power system after the line is broken, and at the same time enabling all anticipated accidents and expectations The load margin of the ground-state power system meets the threshold requirements, and the ground-state power system is expected to meet the safe operation constraints after the disconnection.

As a possible implementation of this embodiment, the process of establishing the mathematical model of the optimal circuit breaker with enhanced voltage stability taking into account the expected accident set is specifically as follows:

Given the current power system status and data, short-term predicted load, power generation plan, grid maintenance plan, a set of anticipated accident sets, alternative breaking line plans and the maximum allowable number of breaking lines;

The objective function of the mathematical model of the optimal circuit breaker with enhanced voltage stability considering the expected accident set is:

The continuous power flow balance equation of the expected ground state power system after the transmission line is disconnected is:

The load margin limit requirements of the expected ground state power system after the transmission line is disconnected are:

λ _b ≥λ _th,b (3)

The safe operation constraints of the expected ground state power system after the transmission line is disconnected are:

It is envisaged that the continuous power flow balance equation of the accident system is:

The load margin limit requirements for all anticipated accident systems are:

The restriction conditions for the number of open circuits are:

NE(N _b -N)≤m (7)

In the formula, B is the set of power system nodes; C is the given set of predicted accidents; c _{i is the i-th} predicted accident in the set of predicted accidents, and c _i ∈ C;

And λ _b respectively represent the load margin of the predicted accident c _i and the expected ground state power system; λ _th,c , λ _th,b are the load margin limits of the predicted accident and the expected ground state power system _{respectively. N b} and N are the power transmission The expected base state power system power network after line breaking and the power network of the current power system; NE (N _b -N) is _{the difference between the grid topology N b} and N, that is, the circuit breaking plan, m represents the allowable opening the number of broken transmission _{_{line; V i, V i, min}} , V i, max represents the voltage amplitude of the node i, respectively, the lower and upper limits permissible voltage; S _{(i, j)} and S _{(i, j), max} respectively, Represents the power on the transmission line ij connected between nodes i and j and the maximum power allowed to flow.

As a possible implementation of this embodiment, the method includes the following steps:

Evaluation stage: According to the given data and expected power system data, the continuous power flow method is applied to analyze and calculate the voltage stability of the expected ground state power system; the fast load margin estimation method is used to analyze all expected accidents, and the estimated accidents are concentrated Predict the load margin of the accident;

Expected accident grouping and expected accidents represent the disconnection line identification stage: apply any clustering method to group all expected accidents, and select the expected accident with the lowest load margin as the representative in each group, where the load margin is lower than the threshold The representative of the expected accident is the representative of the serious expected accident, and then the best disconnecting line is identified for the representative of the serious expected accident, and the stage method is used to identify the effective disconnecting line. Finally, all the expected accident representatives are verified, and the results are sent to the next stage. ；

Expected base state power system disconnection line identification stage: the results of the previous stage are used as alternative disconnection lines, and the phase method is used to identify the disconnection lines that can maximize the expected base state power system voltage stability margin, and the identification results are sent Go to the next stage;

Verification stage: Perform a quick accident analysis on the solution of the previous stage to check whether the load margin of all expected accidents in the expected accident set meets the threshold requirements after these lines are disconnected; if the requirements are met, the solution is retained; otherwise the solution is deleted , Continue to perform rapid accident analysis, until all disconnected circuit solutions are all verified. As a possible implementation of this embodiment, the given data includes online data of the current power system obtained from the energy management system and the SCADA system; the expected power system data includes predicted load, power generation plan, and grid maintenance plan, Alternative disconnection of transmission lines.

As a possible implementation of this embodiment, the evaluation phase includes the following steps:

S1, the current power system online data obtained from the energy management system and the SCADA system, the network topology of the current power system, the predicted data of the expected power system and the expected accident set, input the set of single alternative breaking lines and the maximum allowable breaking The number of transmission lines m, set the counter k=1;

S2, apply the continuous power flow method to analyze the static voltage stability of the current power system and the expected base state power system, and obtain: PV curve, the static voltage stability load margin of the current power system and the expected base state power system, and the expected base state power system saddle node The voltage amplitude of the bifurcation point and the power of each branch, the left eigenvector corresponding to the zero eigenvalue of the Jacobian matrix;

S3: Analyze all anticipated accidents and judge the voltage stability of all anticipated accidents;

S4. Determine whether the load margin threshold of the expected ground state power system and the expected accident meets the following conditions:

S41: If the load margin of the expected base state power system meets the threshold requirement of formula (3), and the load margins of all anticipated accidents meet the threshold requirement of formula (6), then perform the expected base state power system open circuit identification stage;

S42: If the load margin of the expected ground state power system meets the threshold requirement of formula (3), but the load margin of the expected accident does not meet the threshold requirement of formula (6), then perform the expected accident grouping and the expected accident representative of the disconnected line identification stage;

S43: If the load margin of the expected base state power system does not meet the threshold requirement of formula (3), and the load margins of all anticipated accidents all meet the threshold requirement of formula (6), then perform the expected base state power system open circuit identification stage;

S44: If the load margin of the expected base state power system does not meet the threshold requirement of equation (3), and the load margin of an expected accident does not meet the threshold requirement of equation (6), then perform the expected accident grouping and the expected accident representative switch Line identification stage.

As a possible implementation of this embodiment, the expected accident grouping and the expected accident representative disconnected line identification stage includes the following steps:

S5, classify all anticipated accidents into groups, and determine the expected accident representatives of each anticipated accident group:

S51: Apply the fast load margin estimation method to estimate the load margin for all anticipated accidents;

S52: Perform the first grouping of all anticipated accidents according to the load margin distance index. The load margin distance index is defined as follows:

Where

Is the load margin of the anticipated accident c _i

And the expected accident c _j load margin

The absolute value of the difference;

S53,

According to the voltage distribution distance index, the groups obtained for the first time are grouped again. The voltage distribution distance index is defined as follows:

Where

Is the vector composed of the voltage amplitude of each node at the bifurcation point of the saddle node in the predicted accident c _i,

Is the sum of the absolute differences in the voltage amplitudes of the saddle node bifurcation points of the _{accidents c i} and c _j,

Is the voltage amplitude of the nth node at the bifurcation point of the saddle node in the predicted accident c _i;

When the expected accidents c _i and c _j meet the following conditions, the two expected accidents are classified into the same group:

Among them, ε _λ and ε _x are respectively the minimum values of the load margin distance index and the voltage distribution distance index of the predicted accident group;

S54: Select the expected accident with the lowest load margin in each group as a representative, and determine whether the expected accident representative meets the load margin threshold requirement of formula (6), and if it does not meet the load margin threshold requirement, it is defined as a serious expected accident representative;

S6, to identify the disconnected line on behalf of the expected accident:

S61: Apply the continuous power flow calculation method to accurately calculate the load margin of all the representatives of the serious anticipated accidents, and obtain: the voltage stability load margin of the representatives of the serious anticipated accidents, corresponding to the zero characteristic root of the Jacobian matrix at the bifurcation point of the saddle node The left eigenvector and each predicted accident represent the active power and reactive power of each branch at the bifurcation point of the saddle node;

S62. Calculate the change in the load margin of the _{representative c i} due to the expected accident caused by the interruption of the transmission line ij

The calculation method is as follows:

among them,

It is the non-zero left eigenvector corresponding to the zero eigenvalue of the Jacobian matrix at the bifurcation point of the saddle node where the accident representative c _{i is expected;}

Eigenvectors

Elements at the positions of the corresponding bus i active power balance equation, bus i reactive power balance equation, bus j active power balance equation, and bus j reactive power balance equation;

with

Are the active power and reactive power flowing from node i to node j before the line ij in the _{c i} system is broken by the expected accident;

with

They are the active power and reactive power flowing from node j to node i before line ij in the expected accident c _{i system is disconnected;}

S63. Use the weighted sensitivity index to evaluate the change in load margin after the line is disconnected. The calculation method of the weighted sensitivity index is as follows:

Δλ _ij ≥0 (15)

In the formula, C _{RC, W} is the representative set of serious anticipated accidents,

Is the weighted coefficient of the load margin of the expected accident representative c _i _{; Δλ ij} is the load margin change represented by all the severe expected accidents caused by the disconnection of the transmission line ij;

Calculate the weighted sensitivity one by one for all alternative disconnected lines and representatives of serious anticipated accidents, and send the disconnected lines with a weighted sensitivity greater than or equal to zero to the next step to continue calculation;

In S64, the weighted margin index is used to calculate and sort the alternative breaking lines selected in the previous step. The weighted margin calculation method is as follows:

In the formula, λ ^ij is the weighted margin;

_{The load margin of the representative c i} for the serious anticipated accident estimated by the load margin estimation method;

For step S63, all the disconnected lines selected and the representatives of serious anticipated accidents are calculated one by one weighted margin, and all alternative disconnected lines are sorted according to the weighted margin index according to the weighted margin value from large to small, and the sorting is performed The earlier alternative circuit breaker is sent to the next step;

S65. Apply the continuous power flow method to accurately calculate the load margin of the power system under each of the serious anticipated accident representatives in the sequence of the alternative interrupted transmission lines after the previous step. The calculation will enable all serious anticipated accidents to represent the load margin. The interrupted transmission lines with a degree greater than the threshold λ _th,c are retained until the first alternative interrupted transmission line does not meet the threshold requirements under a certain serious expected accident representative system, and then all the retained interrupted transmission lines The route solution is sent to the next step;

S66. Apply the continuous power flow method to further analyze and calculate the disconnected line solutions identified in the previous step, calculate their load margins under other expected accident representative systems after they are broken one by one, and eliminate those that do not meet the load margin threshold requirements. The circuit breaker solution, and the retained solution is sent to the expected ground state power system breaker circuit identification stage.

As a possible implementation of this embodiment, the disconnected line identification stage of the expected base state power system is the application of a staged identification method to identify the expected base state power system for disconnected lines, which specifically includes the following steps:

S71, scanning step: use sensitivity index to screen the alternative disconnection lines, and the sensitivity index calculation method is as follows:

Calculate the sensitivity one by one for the alternative breaking lines, and set the sensitivity to be greater than or equal to zero, that is

The interrupted transmission line is sent to the next step to continue calculation;

S72. Sorting step: use the load margin estimation method to estimate the load margin of the alternative circuit breakers screened out in the previous step, and sort according to the calculation results from large to small, and sort the top alternative breaking lines The line is sent to the next step to continue analysis and calculation;

S73, identification step: use the continuous power flow method to accurately calculate the load margin of the alternative circuit breakers sorted in the previous step, sort according to the calculated load margin value, and send the identification result to the verification stage to continue the analysis.

As a possible implementation of this embodiment, the verification phase includes the following steps:

S8, use the recognition result of the previous stage as the solution set, perform a fast predicted accident analysis, and verify whether the load margin of all predicted accidents meets the threshold requirement:

S81, extract a transmission line breaking plan from the solution set, generate a new grid topology, and send it to the next step;

S82, the system after the transmission line is disconnected is regarded as the network structure of a new expected base state power system, and the expected accident analysis and calculation of the power system after the interruption are performed according to the input data. If the load margin of all expected accidents meets the formula (6) If the threshold is required, the disconnected transmission line solution is retained, otherwise it is removed from the solution set;

S83: Judge whether all the solutions in the solution set have been verified, if yes, go to step S9, otherwise continue to perform step S81;

S9, if k=m, go to step S10; otherwise, set the counter k=k+1, and generate multiple disconnection line candidate sets at the same time, the generation method is as follows:

CA _k ＝A _k ×A ₁ (18)

In the formula, CA _k is an alternative breaking plan that allows k lines to be opened, A _k-1 and A ₁ are the recognition results when the number of disconnected lines is k-1 and 1, respectively, and formula (18) is expressed as The alternative breaking scheme that allows the breaking of k lines is the Cartesian product of the recognition results when the number of open lines is k-1 and one;

S10, output results, the output results include solutions for all effectively interrupted transmission lines, optimized load margins of the expected ground state power system, predicted failure set analysis results, and predicted accidents with minimum load margins.

On the other hand, the embodiment of the present invention provides a group-based online optimization method for the optimal grid topology taking into account expected accident sets, including the following steps:

Establish a mathematical model of the optimal circuit breaker for enhanced voltage stability that takes into account the set of anticipated accidents;

According to the real-time grid data given by the energy management system, the load data predicted based on historical data and the generator output plan, calculate the expected voltage stability margin value of the ground-state power system;

Use load margin estimation method to quickly estimate the voltage stability margin of all expected accidents in the expected accident set;

Use the characteristics of expected accident groups to divide all expected accidents into groups, and select representatives of expected accidents in each group;

According to whether the voltage stability margin of the expected ground state power system and the expected accident set meets the limit requirements, different calculations are performed according to the situation;

Identify the disconnected lines that enable all expected accident representatives to meet the expected accident set load margin limit;

Identify the open circuit that maximizes the expected base state power system load margin;

Perform analysis of anticipated accidents;

Output the result.

λ _b ≥λ _th,b (3)

The load margin limit requirements for all anticipated accident systems are:

The restriction conditions for the number of open circuits are:

NE(N _b -N)≤m (7)

And λ _b respectively represent the load margin of the predicted accident c _i and the expected ground state power system; λ _th,c and λ _th,b are the load margin limits of the predicted accident and the expected ground state power system _{respectively; N b} and N are respectively The expected ground state power network after the transmission line is disconnected and the current power network of the power system; NE (N _b -N) is _{the difference between the grid topology N b} and N, that is, the circuit break plan, and m represents the allowable number of breaking the transmission _{_{line; V i, V i, min}} , V i, max represents the voltage amplitude of the node i, respectively, the lower and upper limits permissible voltage; S _{(i, j)} and S _{(i, j), max} Respectively represent the power and the maximum power allowed to flow through the power transmission line ij connected between nodes i and j.

The technical solutions of the embodiments of the present invention may have the following beneficial effects:

The present invention is based on a group-based online optimization method for the optimal grid topology taking into account anticipated accident sets. The grid topology of the power system is optimized by the measure of breaking the running transmission line, and the online opening of the power system voltage stability enhancement control is solved. Identify the problem of broken lines, and finally give a circuit breaking scheme that meets the voltage stability load margin threshold of the expected ground-state power system and the expected accident set. At the same time, the expected ground-state power system after the line is disconnected meets the safety constraints of the power system operation, and the operation is dispatched Personnel can directly select one of the multiple circuit breaker schemes given for breaking operation, thereby enhancing the voltage stability of the power system, and this method does not require additional investment in power equipment, and has high economic efficiency and easy operation.

Another method of the present invention is based on the group-based online optimization method for the optimal grid topology considering the expected accident set. The method includes the following steps: establishing a mathematical model of the optimal breaking line for enhanced voltage stability that takes into account the expected accident set; Calculate the voltage stability margin value of the expected ground state power system (for example, the power system after 30 minutes) based on the real-time grid data, the load data predicted by historical data, and the generator output plan; use the load margin estimation method to quickly estimate the forecast The voltage stability margin of all expected accidents in the accident set; use the characteristics of expected accident groups to group all expected accidents, and select the expected accident representative in each group; according to the expected ground state power system and the voltage stability margin of the expected accident set meet the limit Perform different calculations according to the actual value requirements; identify the interrupted lines that can make all expected accident representatives meet the expected accident set load margin limit; identify the interrupted lines that can maximize the expected base state power system load margin; execute the expected accident Analysis; output results. The present invention makes full use of the group characteristics of predicted accidents on static voltage stability, avoids a large number of calculations of "dual nesting" for all predicted accidents to identify disconnected lines one by one, and overcomes its online calculation and online application. The present invention can provide a circuit breaker plan that meets the target requirements based on the number of allowed circuit breakers. The power system after the circuit breaks meets the static and safe operation requirements, and is especially suitable for large-scale power systems considering the expected accident set of power grid topology Optimized online calculation and analysis. According to the real-time power system status and data, short-term forecast data (including the predicted load, power generation plan, and maintenance plan of each node of the power system), and the expected accident set, the invention optimizes the power network structure by disconnecting the transmission line to ensure the expected accident set On the basis that the voltage stability margin of the expected ground-state power system meets the limit requirements, the static voltage stability margin of the expected ground-state power system is maximized. The invention makes full use of the group characteristics of anticipated accidents, reduces the amount of calculation that takes into account a large number of anticipated accidents to only need to consider representative anticipated accidents, greatly reduces the calculation difficulty of the original problem, increases the calculation speed, and can meet the requirements of online control. Real-time requirements. At the same time, the present invention can provide multiple groups of disconnecting circuit control schemes for operators, and the power system after each scheme can ensure that the safe operation requirements of the power system are met.

Description of the drawings:

Fig. 1 is a flow chart showing a group-based online optimization method of optimal power grid topology taking into account a set of anticipated accidents according to an exemplary embodiment;

Fig. 2 shows the overall architecture of a group-based online optimization method for optimal grid topology taking into account a set of anticipated accidents according to an exemplary embodiment;

Fig. 3 is a detailed flowchart of another online optimization method for group-based optimal grid topology taking into account expected accident sets according to an exemplary embodiment;

Figure 4 is a comparison diagram of the P-V curve of the ground state power system before and after the line 15-17 is disconnected and the severely anticipated accident system;

Figure 5 is a comparison diagram of load margins of all anticipated accidents before and after line 15-17 is disconnected.

Detailed ways

The present invention will be further described below in conjunction with the drawings and embodiments:

In order to clearly illustrate the technical characteristics of the present solution, the present invention will be described in detail below through specific implementations and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numerals and/or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not in itself indicate the relationship between the various embodiments and/or settings discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. The present invention omits descriptions of well-known components and processing techniques and processes to avoid unnecessarily limiting the present invention.

Example 1

The embodiment of the present invention provides a group-based online optimization method for the optimal grid topology taking into account the expected accident set. Under the constraint of the number of allowed interrupted lines, searching for the optimal interrupted line for enhanced voltage stability that takes into account the expected accident set A set of optimal circuit breaker schemes based on the mathematical model of, the optimal circuit breaker scheme satisfies the following conditions: maximizes the load margin of the expected ground state power system after the line is disconnected, and at the same time enables all expected accidents and expected base state power systems The load margin meets the threshold requirement, and the ground-state power system is expected to meet the safe operation constraints after disconnection.

λ _b ≥λ _th,b (3)

The load margin limit requirements for all anticipated accident systems are:

The restriction conditions for the number of open circuits are:

NE(N _b -N)≤m (7)

Fig. 1 is a flow chart showing a method for online optimization of the best power grid topology based on a group based on a set of anticipated accidents, according to an exemplary embodiment. As shown in FIG. 1, as a possible implementation manner of this embodiment, the method includes the following steps:

Verification phase: Perform a quick accident analysis on the solution of the disconnected line identification phase of the expected ground state power system, and check whether the load margin of all expected accidents in the expected accident set meets the threshold requirements after these lines are disconnected; if the requirements are met, keep this Solution; otherwise, delete the solution and continue to perform quick accident analysis until all disconnected circuit solutions are verified.

As a possible implementation of this embodiment, the given data includes online data of the current power system obtained from the energy management system and the SCADA system; the expected power system data includes predicted load, power generation plan, and grid maintenance plan, Alternative disconnection of transmission lines.

As shown in FIG. 2, as a possible implementation manner of this embodiment, the evaluation phase includes the following steps:

S42: If the load margin of the expected base state power system meets the threshold requirement of formula (3), but the load margin of the predicted accident does not meet the threshold requirement of formula (6), then perform the expected accident clustering and the expected accident representative disconnection line identification stage;

As shown in Figure 2, as a possible implementation of this embodiment, the expected accident grouping and the expected accident representative disconnected line identification stage includes the following steps:

Where

Is the load margin of the anticipated accident c _i

And the expected accident c _j load margin

The absolute value of the difference;

S53,

Where

S54: Select the expected accident with the lowest load margin in each group as a representative, and determine whether the expected accident representative meets the load margin threshold requirement of formula (6), if not, it is defined as a serious expected accident representative;

S6, to identify the disconnected line on behalf of the expected accident:

The calculation method is as follows:

among them,

Eigenvectors

with

Δλ _ij ≥0 (15)

In the formula, λ ^ij is the weighted margin;

S66. Apply the continuous power flow method to further analyze and calculate the disconnected line solutions identified in the previous step, calculate their load margins under all other expected accident representative systems after they are disconnected one by one, and exclude those that do not meet the load margin threshold requirements The circuit breaker is solved, and the retained solution is sent to the break circuit identification stage of the expected ground state power system.

As shown in FIG. 2, as a possible implementation of this embodiment, the expected base state power system disconnection line identification stage is to apply a staged identification method to the expected base state power system to perform disconnection line identification, which specifically includes the following steps:

among them,

Is the change in the expected base state power system load margin caused by the interruption of the transmission line ij;

ω _b is the non-zero left eigenvector corresponding to the zero eigenvalue of the Jacobian matrix at the bifurcation point of the saddle node of the expected ground state power system;

They are _{the elements of the eigenvector ω b} at the positions of the corresponding bus i active power balance equation, bus i reactive power balance equation, bus j active power balance equation, and bus j reactive power balance equation;

As shown in Figure 2, as a possible implementation of this embodiment, the verification phase includes the following steps:

CA _k ＝A _k ×A ₁ (18)

Example 2

The embodiment of the present invention provides a group-based online optimization method for the optimal grid topology taking into account expected accident sets, including the following steps:

Perform analysis of anticipated accidents;

Output the result.

The mathematical model of the optimal breaking circuit considering the expected accident set is as follows:

Given the current power system status and data, short-term predicted load, power generation plan, grid maintenance plan, a set of anticipated accident sets, alternative circuit breaker schemes and the maximum allowable number of circuit breakers, the enhanced voltage stability of the anticipated accident set is taken into account The mathematical model of the optimal breaking circuit is:

λ _b ≥λ _th,b (3)

NE(N _b -N)≤m (7)

Among them, B is a set of power system nodes; C is a given set of predicted accidents; c _{i is the i-th} predicted accident in the set of predicted accidents, and c _i ∈ C;

And λ _b respectively represent the load margin of the predicted accident c _i and the expected ground state power system; λ _th,c , λ _th,b are the load margin limits of the predicted accident and the expected ground state power system _{respectively; N b} and N are respectively It is expected that the power network after the transmission line of the ground state power system is interrupted (ie, the optimized power network topology) and the current power network of the power system (ie, the power network topology before optimization); NE(·) is mapped to the grid topology the difference (i.e., breaking line scheme), m represents a number to allow the breaking of transmission line, designated by the user between the _B and N, _{_{N; V i, V i, min}} , V i, max denote node i Voltage amplitude, allowable voltage lower limit and upper limit. S _{(i, j)} and S _{(i, j), max} respectively represent the power and the maximum power allowed to flow on the transmission line ij between nodes i and j.

In the above model, formula (1) is the objective function in the mathematical model, and its goal is to maximize the voltage stability of the expected ground state power system (that is, load margin λ _b ); formula (2) is the expectation after the transmission line is disconnected The continuous power flow balance equation of the ground state power system; equation (3) is the expected load margin limit requirement of the ground state power system after the transmission line is interrupted; equations (4a) and (4b) are the expected safe operation of the ground state power system after the transmission line is interrupted Constraints; Equation (5) is the continuous power flow balance equation of the expected accident system; Equation (6) is the load margin limit requirement of all expected accident systems; Equation (7) is the constraint condition for the number of lines allowed to be disconnected.

Equations (1)-(7) are the best breaking line mathematical model considering the expected accident set. This model describes the search for a set of optimal circuit breaking schemes under the constraint of the number of open circuit breakers, maximizing the load margin of the expected ground state power system after the line is broken, and enabling all expected accidents and expected ground state power at the same time The load margin of the system meets the threshold requirements, and the ground state power system is expected to meet the safe operation constraints after disconnection.

The overall architecture of the group-based online optimization method for the optimal grid topology taking into account the expected accident set is shown in Figure 2. The method generally includes four stages:

The first stage is the evaluation stage. This stage is based on given data (including online data of the current power system obtained from energy management systems and SCADA, etc.) and expected power system data (including predicted load, power generation plan and grid maintenance plan, alternative transmission lines, etc.) , Apply continuous power flow method to analyze and calculate the voltage stability of the expected ground state power system; apply fast load margin estimation method (for example, look-ahead margin estimation method) to analyze all expected accidents, and estimate all expected accidents. The load margin of the accident.

The second stage: the grouping of anticipated accidents and the identification of broken lines on behalf of anticipated accidents. At this stage, any clustering method is used to group all expected accidents, and the expected accident with the lowest load margin is selected as the representative in each group. The representative of the expected accident with the load margin lower than the threshold is the representative of the serious expected accident. Representatives of serious anticipated accidents shall identify the best disconnected lines, apply the staged method to identify effective disconnected lines, and finally verify all representatives of anticipated accidents, and send the results to the next stage.

The third stage: the identification stage of the open circuit of the expected ground state power system. In this stage, the results of the previous stage are used as alternative breaking lines, and the stage method is used to identify the breaking lines that can maximize the expected voltage stability margin of the ground-state power system, and the identification results are sent to the next stage.

The fourth stage: verification stage. In this stage, a quick accident analysis is performed on the solution of the third stage, and it is checked whether the load margin of all the expected accidents in the expected accident set meets the threshold requirement after these lines are disconnected respectively (that is, equation (6)). If it meets the requirements, keep the solution; otherwise, delete the solution and continue to perform quick accident analysis until all disconnected circuit solutions are verified.

The group-based online optimization method for the optimal grid topology taking into account the expected accident set specifically includes the following steps, as shown in Figure 3, the specific implementation steps are:

1. The online data of the current power system obtained from the energy management system and SCADA, the network topology of the current power system, and the forecast data of the expected power system (including load forecast data, the expected power system maintenance plan, and the expected power system generation plan ) And the expected accident set, enter a single set of alternative breaking lines, the maximum allowable number of interrupted transmission lines m, and set the counter k=1;

2. Apply the continuous power flow method to analyze the static voltage stability of the current power system and the expected base state power system, and obtain: i) PV curve, ii) the static voltage stability load margin of the current power system and the expected base state power system, iii) the expected base state The voltage amplitude at the bifurcation point of the saddle node of the power system and the power of each branch, the left eigenvector corresponding to the zero eigenroot of the Jacobian matrix, and other information;

3. Analyze all anticipated accidents and judge the voltage stability of all anticipated accidents;

4. Judging the satisfaction of the load margin threshold of the expected ground state power system and the expected accident, divided into the following four cases:

4.1 If the load margins of the expected ground-state power system and the expected accident meet their respective threshold requirements (Equations (3) and (6)), then step 7 is executed;

4.2 If the load margin of the expected ground state power system meets the threshold requirement (Equation (3)), but the load margin of the predicted accident does not meet the threshold requirement (Equation (6)), then step 5 is executed;

4.3 If the load margin of the expected base state power system does not meet the threshold requirement (Equation (3)), and the load margins of all anticipated accidents all meet the threshold requirement (Equation (6)), go to step 7;

4.4 If the load margins of the expected ground-state power system and the expected accidents do not meet their respective threshold requirements (Equations (3) and (6)), perform step 5;

5 Group all expected accidents into groups, and determine the expected accident representatives of each expected accident group:

5.1 Apply fast load margin estimation method (for example, look-ahead margin estimation method) to estimate the load margin of all anticipated accidents;

5.2 Perform the first grouping of all anticipated accidents according to the load margin distance index. The load margin distance index is defined as follows:

Where

Is the load margin of the anticipated accident c _i

And the expected accident c _j load margin

The absolute value of the difference.

5.3 According to the voltage distribution distance index, step 5.1

The groups obtained for the first time are grouped again, and the voltage distribution distance index is defined as follows:

Where

It is the voltage amplitude of the nth node at the bifurcation point of the saddle node in the _{predicted accident c i.}

Mathematically speaking, when the expected accidents c _i and c _j meet the following conditions, the two expected accidents are divided into the same group:

Among them, ε _λ and ε _x are the minimum values of the load margin distance index and the voltage distribution distance index of the predicted accident group, respectively.

5.4 Select the expected accident with the lowest load margin in each group as the representative, and judge whether the expected accident representative meets the load margin threshold requirement of formula (6), if not, it is defined as the severe expected accident representative.

6 Carry out the identification of the expected accident on behalf of the broken line.

6.1 Apply the continuous power flow calculation method to accurately calculate the load margin of all the representatives of the serious anticipated accidents, and obtain: i) the voltage stability load margins of the representatives of the serious anticipated accidents, ii) the zero characteristic of the Jacobian matrix at the bifurcation point of the saddle node The left eigenvector corresponding to the root, iii) each predicted accident represents the active power and reactive power of each branch at the bifurcation point of the saddle node;

6.2 Calculate the change in the load margin of the _{expected accident c i} caused by the interruption of the transmission line ij

The calculation method is as follows:

among them,

Eigenvectors

with

Respectively represent the expected accidents.. The active power and reactive power flowing from node j to node i in the system before line ij is disconnected.

6.2 The weighted sensitivity index is used to evaluate the change in load margin after the line is disconnected. The calculation method of the weighted sensitivity index is as follows:

Δλ _ij ≥0 (15)

In the formula, C _{RC, W} is the representative collection of serious expected accidents,

It is the weighting factor of the expected accident that represents the c _i load margin. Δλ _ij is the load margin variation represented by all serious expected accidents caused by the disconnection of the transmission line ij.

Calculate the weighted sensitivities one by one for all alternative disconnected lines and representatives of serious anticipated accidents, and send the disconnected lines whose weighted sensitivity is greater than or equal to zero (satisfying equation (14)) to the next step to continue calculation.

6.3 The weighted margin index is used to calculate and sort the alternative breaking lines selected in the previous step. The weighted margin calculation method is as follows:

In the formula, λ ^ij is the weighted margin;

The severe anticipated accidents estimated by the load margin estimation method (for example, the look-ahead method) represent the c _i load margin.

Calculate the weighted margins one by one for all the disconnected lines selected in step 6.2 and the representatives of serious anticipated accidents, and sort all alternative disconnected lines according to the weighted margin index from large to small according to the weighted margin index, and sort them The earlier alternative circuit breaker is sent to the next step.

6.4 Apply the continuous power flow method to accurately calculate the load margin of the power system under each representative representative of the serious expected accidents in order for the alternative interrupted transmission lines sorted in the previous step. The interrupted transmission lines with a margin greater than the threshold λ _th,c are retained until an alternative interrupted transmission line is encountered that does not meet the threshold requirements under a certain serious expected accident representative system, and then all the retained interrupted transmission lines The route solution is sent to the next step.

6.5 Apply the continuous power flow method to further analyze and calculate the disconnected line solutions identified in the previous step, calculate the load margins under all other expected accident representative systems after they are broken one by one, and eliminate the load margin threshold requirements that are not met The circuit breaker is solved, and the reserved solution is sent to the next step.

7 Application of staged identification method to identify the open circuit of the expected ground state power system

7.1 Scanning step: In this step, the sensitivity index is used to screen the alternative disconnection lines. The sensitivity index calculation method is as follows:

among them,

They are respectively _{the elements of the eigenvector ω b} in the corresponding position of the active power balance equation of bus i, the reactive power balance equation of bus i, the active power balance equation of bus j, and the reactive power balance equation of bus j.

Calculate the sensitivity one by one for the alternative breaking lines, and set the sensitivity to be greater than or equal to zero (ie,

) The interrupted transmission line is sent to the next step to continue the calculation.

7.2 Sorting step: In this step, the load margin estimation method (for example, the look-ahead method) is used to estimate the load margin of the candidate breaking lines selected in the previous step, and the calculation results are sorted from large to small, and Send the ranked alternative breaking lines to the next step to continue analysis and calculation.

7.3 Identification step: This step uses the continuous power flow method to accurately calculate the load margin of the alternative circuit breakers sorted in the previous step, sort according to the calculated load margin value, and send the identification result to the next step to continue the analysis.

8 Use the recognition result of the previous step as a solution set, perform a fast predicted accident analysis, and verify that the load margin thresholds of all predicted accidents are met.

8.1 Extract a circuit breaker plan from the solution set and send it to the next step.

8.2 The system after the transmission line is interrupted is regarded as the network structure of the expected ground state power system, and the expected accident analysis and calculation of the power system after the interruption are performed according to the input data. If the load margin of all expected accidents meets the threshold requirements (Equation (6)) ), the solution for the disconnected transmission line is retained, otherwise it is removed from the solution set.

8.3 Determine whether all the solutions in the solution set have been verified, if yes, go to step 9, otherwise continue to step 8.1.

9 If k=m, go to step 10; otherwise, set the counter k=k+1, and generate multiple disconnection line candidate sets at the same time. The generation method is as follows:

CA _k ＝A _k ×A ₁ (18)

In the formula, CA _k is an alternative breaking plan that allows k lines to be opened, A _k-1 and A ₁ are the recognition results when the number of disconnected lines is k-1 and 1, respectively, and formula (18) is expressed as The alternative breaking scheme that allows the breaking of k lines is the Cartesian product of the recognition results when the number of open lines is k-1 and one.

10 Output results, including solutions for all effectively interrupted transmission lines, optimized load margins of the expected ground state power system, predicted accident set analysis results, and predicted accidents with minimum load margins.

So far, the detailed calculation process of the online optimization method for the optimal grid topology based on the group-based and anticipated accident set according to the present invention is completed.

The method of the present invention adopts IEEE 118-node power system simulation and verification, and the data comes from IEEE standard calculation examples. In the simulation verification of single line breaking, node 69 is a balanced node, and the thermal limit of each line is limited to 500MVA. The 175 transmission lines (not including 9 radial lines) in the calculation example are all used as alternatives for breaking transmission. line. Forecast data: 39 loads in area 2 are increased by 20% on the basis of the current base state power system, and the generator output of node 1, node 4, and node 31 are used to adjust the output of the generator to meet the load increase, and their respective output is adjusted to the load increase 35%, 25% and 40% of the amount. It is estimated that there are 177 faults (consider the power failure of other single lines except generators and radiation lines). After calculation, the expected ground state power system load margin is λ _b =5.3519, and the calculation example sets λ _th,b = _th,c =3. After calculation, it is predicted that only the load margin of faults 30-38 in the accident does not meet the threshold requirement, and the load margin is 2.7833. Using the predictive accident grouping method proposed by the present invention, 16 predictive accident groups can be formed, and the grouping results are shown in Table 1.

Table 1 The expected accident clustering results of the IEEE 118-bus power system calculation example:

Applying the group-based online optimization method of the optimal grid topology based on the predicted accident set proposed by the present invention, the optimal breaking line is finally identified as 15-17. The expected ground-state power system load margin after breaking the line is 5.4399, the thermal stability limit margin is 0.1959, and the voltage limit margin is 0.8433. The P-V curve of the ground state power system before and after the line 15-17 is disconnected and the severely anticipated accident system is shown in Figure 4. The load margin comparison chart of all predicted accidents before and after the line 15-17 is disconnected is shown in Figure 5. After 15-17 breaks, the expected load margin of the ground-state power system is increased from 5.3519 to 5.4399 (an increase of 1.64%), and the load margin of the most serious anticipated accident is increased from 2.7833 to 3.5242 (an increase of 26.62%).

In the simulation verification of the breaking method of multiple lines, the predicted data is: the active power requirements of the 10 loads at nodes 50-59 are 13.6, 32, 27.2, 8, 180.8, 10.8, 134.4, 19.2, 27.2, 43.2 MW, and reactive power respectively. The requirements are 3.2, 16.14.4, 51.2, 35.2, 28.8, 4.8, 12.8 and 20.8 Mvar respectively. The output of the generators at node 10, node 36, and node 59 are adjusted to meet the load increment, and their output is adjusted to 173.76, 231.68 and 173.76MW respectively. After calculation, the expected load margin of the ground state power system is λ _b =2.1447, and the load margin of the most serious anticipated accident (that is, the line 53-54 out of operation) is 0.9981. This calculation example sets λ _th,b =2 and λ _th,c =1. Using the predictive accident grouping method proposed by the present invention, the identification results of a single disconnected line and multiple disconnected lines that meet the load margin threshold requirements are shown in Table 2 and Table 3. It can be seen that breaking a single transmission line 49-66 can increase the expected load margin of the ground-state power system from 2.1447 to 2.2687 (increased by 5.78%), and can increase the load margin of the most severe anticipated accident from 0.9981 to 1.0336 (increased) 3.56%). When two lines are allowed to open, the best open lines are 49-66 and 40-42, which can increase the load margin of the expected ground state power system and the most severe expected accident by 8.5% and 4.01%, respectively.

Table 2 Recognition results of a single broken line:

Table 3 Recognition results of multiple disconnected lines:

The above are only the preferred embodiments of the present invention. For those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications are also regarded as the present invention. The scope of protection of the invention.

Claims

A group-based online optimization method for the optimal power grid topology taking into account the expected accident set, which is characterized by finding the mathematics of the optimal interrupting line for the enhancement of voltage stability that takes into account the expected accident set under the constraint of the number of interrupted lines A set of optimal circuit breaker schemes of the model. The optimal circuit breaker scheme satisfies the following conditions: maximizes the load margin of the expected ground state power system after the line is broken, and at the same time enables all expected accidents and the expected base state power system load The margin meets the threshold requirement, and the ground-state power system is expected to meet the safe operation constraints after the line is disconnected.
The group-based online optimization method for the optimal power grid topology taking into account the expected accident set according to claim 1, wherein the mathematical model of the optimal interrupting line for the enhanced voltage stability that takes into account the expected accident set is specifically established. for:

Given the current power system status and data, short-term predicted load, power generation plan, grid maintenance plan, a set of anticipated accident sets, alternative breaking line plans and the maximum allowable number of breaking lines;

The objective function of the mathematical model of the optimal circuit breaker with enhanced voltage stability considering the expected accident set is:

The continuous power flow balance equation of the expected ground state power system after the transmission line is disconnected is:

The load margin limit requirements of the expected ground state power system after the transmission line is disconnected are:

λ b ≥λ th,b (3)

The safe operation constraints of the expected ground state power system after the transmission line is disconnected are:

It is envisaged that the continuous power flow balance equation of the accident system is:

The load margin limit requirements for all anticipated accident systems are:

The restriction conditions for the number of open circuits are:

NE(N b -N)≤m (7)

In the formula, B is the set of power system nodes; C is the given set of predicted accidents; c i is the i-th predicted accident in the set of predicted accidents, and c i ∈ C;
And λ b respectively represent the load margin of the predicted accident c i and the expected ground state power system; λ th,c and λ th,b are the load margin limits of the predicted accident and the expected ground state power system respectively; N b and N are respectively The expected ground state power network after the transmission line is disconnected and the current power network of the power system; NE (N b -N) is the difference between the grid topology N b and N, that is, the circuit break plan, and m represents the allowable number of breaking the transmission line; V i, V i, min , V i, max represents the voltage amplitude of the node i, respectively, the lower and upper limits permissible voltage; S (i, j) and S (i, j), max Respectively represent the power and the maximum power allowed to flow through the power transmission line ij connected between nodes i and j.
The group-based online optimization method of the optimal grid topology taking into account expected accident sets according to claim 2, wherein the method comprises the following steps:

Evaluation stage: According to the given data and expected power system data, the continuous power flow method is applied to analyze and calculate the voltage stability of the expected ground state power system; the fast load margin estimation method is used to analyze all expected accidents, and the estimated accidents are concentrated Predict the load margin of the accident;

Expected accident grouping and expected accidents represent the disconnection line identification stage: apply any clustering method to group all expected accidents, and select the expected accident with the lowest load margin as the representative in each group, where the load margin is lower than the threshold The representative of the expected accident is the representative of the serious expected accident, and then the best disconnecting line is identified for the representative of the serious expected accident, and the stage method is used to identify the effective disconnecting line. Finally, all the expected accident representatives are verified, and the results are sent to the next stage. ；

Expected base state power system disconnection line identification stage: the results of the previous stage are used as alternative disconnection lines, and the phase method is used to identify the disconnection lines that can maximize the expected base state power system voltage stability margin, and the identification results are sent Go to the next stage;

Verification stage: Perform a quick accident analysis on the solution of the previous stage to check whether the load margin of all expected accidents in the expected accident set meets the threshold requirements after these lines are disconnected; if the requirements are met, the solution is retained; otherwise the solution is deleted , Continue to perform rapid accident analysis, until all disconnected circuit solutions are all verified.
The group-based online optimization method for the optimal grid topology taking into account expected accident sets according to claim 3, wherein the given data includes online data of the current power system obtained from an energy management system and a SCADA system; The expected power system data includes predicted load, power generation plan and grid maintenance plan, and alternative disconnection of transmission lines.
According to claim 4, the group-based online optimization method for the optimal power grid topology taking into account expected accident sets, wherein the evaluation stage includes the following steps:

S1, the current power system online data obtained from the energy management system and the SCADA system, the network topology of the current power system, the predicted data of the expected power system and the expected accident set, input the set of single alternative breaking lines and the maximum allowable breaking The number of transmission lines m, set the counter k=1;

S2, apply the continuous power flow method to analyze the static voltage stability of the current power system and the expected base state power system, and obtain: PV curve, the static voltage stability load margin of the current power system and the expected base state power system, and the expected base state power system saddle node The voltage amplitude of the bifurcation point and the power of each branch, the left eigenvector corresponding to the zero eigenvalue of the Jacobian matrix;

S3: Analyze all anticipated accidents and judge the voltage stability of all anticipated accidents;

S4. Determine whether the load margin threshold of the expected ground state power system and the expected accident meets the following conditions:

S41: If the load margin of the expected base state power system meets the threshold requirement of formula (3), and the load margins of all anticipated accidents meet the threshold requirement of formula (6), then perform the expected base state power system open circuit identification stage;

S42: If the load margin of the expected base state power system meets the threshold requirement of formula (3), but the load margin of the predicted accident does not meet the threshold requirement of formula (6), then perform the expected accident clustering and the expected accident representative disconnection line identification stage;

S43: If the load margin of the expected base state power system does not meet the threshold requirement of formula (3), and the load margins of all anticipated accidents all meet the threshold requirement of formula (6), then perform the expected base state power system open circuit identification stage;

S44: If the load margin of the expected base state power system does not meet the threshold requirement of equation (3), and the load margin of an expected accident does not meet the threshold requirement of equation (6), then perform the expected accident grouping and the expected accident representative switch Line identification stage.
The group-based online optimization method for the optimal power grid topology taking into account expected accident sets according to claim 5, characterized in that the expected accident grouping and the expected accident representative disconnection line identification stage includes the following steps:

S5, classify all anticipated accidents into groups, and determine the expected accident representatives of each anticipated accident group:

S51: Apply the fast load margin estimation method to estimate the load margin for all anticipated accidents;

S52: Perform the first grouping of all anticipated accidents according to the load margin distance index. The load margin distance index is defined as follows:

Where
Is the load margin of the anticipated accident c i
And the expected accident c j load margin
The absolute value of the difference;

In S53, each group obtained for the first time is grouped again according to the voltage distribution distance index, and the voltage distribution distance index is defined as follows:

Where
Is the vector composed of the voltage amplitude of each node at the bifurcation point of the saddle node in the predicted accident c i,
Is the sum of the absolute differences in the voltage amplitudes of the saddle node bifurcation points of the accidents c i and c j,
Is the voltage amplitude of the nth node at the bifurcation point of the saddle node in the predicted accident c i;

When the expected accidents c i and c j meet the following conditions, the two expected accidents are classified into the same group:

Among them, ε λ and ε x are respectively the minimum values of the load margin distance index and the voltage distribution distance index of the predicted accident group;

S54: Select the expected accident with the lowest load margin in each group as a representative, and determine whether the expected accident representative meets the load margin threshold requirement of formula (6), and if it does not meet the load margin threshold requirement, it is defined as a serious expected accident representative;

S6, to identify the disconnected line on behalf of the expected accident:

S61: Apply the continuous power flow calculation method to accurately calculate the load margin of all the representatives of the serious anticipated accidents, and obtain: the voltage stability load margin of the representatives of the serious anticipated accidents, corresponding to the zero characteristic root of the Jacobian matrix at the bifurcation point of the saddle node The left eigenvector and each predicted accident represent the active power and reactive power of each branch at the bifurcation point of the saddle node;

S62. Calculate the change in the load margin of the representative c i due to the expected accident caused by the interruption of the transmission line ij
The calculation method is as follows:

among them,
It is the non-zero left eigenvector corresponding to the zero eigenvalue of the Jacobian matrix at the bifurcation point of the saddle node where the accident representative c i is expected;
Eigenvectors
Elements at the positions of the corresponding bus i active power balance equation, bus i reactive power balance equation, bus j active power balance equation, and bus j reactive power balance equation;
with
Are the active power and reactive power flowing from node i to node j before the line ij in the c i system is broken by the expected accident;
with
They are the active power and reactive power flowing from node j to node i before line ij in the expected accident c i system is disconnected;

S63. Use the weighted sensitivity index to evaluate the change in load margin after the line is disconnected. The calculation method of the weighted sensitivity index is as follows:

Δλ ij ≥0 (15)

In the formula, C RC, W is the representative set of serious anticipated accidents,
Is the weighted coefficient of the load margin of the expected accident representative c i ; Δλ ij is the load margin change represented by all the severe expected accidents caused by the disconnection of the transmission line ij;

Calculate the weighted sensitivity one by one for all alternative disconnected lines and representatives of serious anticipated accidents, and send the disconnected lines with a weighted sensitivity greater than or equal to zero to the next step to continue calculation;

In S64, the weighted margin index is used to calculate and sort the alternative breaking lines selected in the previous step. The weighted margin calculation method is as follows:

In the formula, λ ij is the weighted margin;
The load margin of the representative c i for the serious anticipated accident estimated by the load margin estimation method;

For step S63, all the disconnected lines selected and the representatives of serious anticipated accidents are calculated one by one weighted margin, and all alternative disconnected lines are sorted according to the weighted margin index according to the weighted margin value from large to small, and the sorting is performed The earlier alternative circuit breaker is sent to the next step;

S65. Apply the continuous power flow method to accurately calculate the load margin of the power system under each of the serious anticipated accident representatives in the sequence of the alternative interrupted transmission lines after the previous step. The calculation will enable all serious anticipated accidents to represent the load margin. The interrupted transmission lines with a degree greater than the threshold λ th,c are retained until the first alternative interrupted transmission line does not meet the threshold requirements under a certain serious expected accident representative system, and then all the retained interrupted transmission lines The route solution is sent to the next step;

S66. Apply the continuous power flow method to further analyze and calculate the disconnected line solutions identified in the previous step, calculate their load margins under other expected accident representative systems after they are broken one by one, and eliminate those that do not meet the load margin threshold requirements. The circuit breaker solution, and the retained solution is sent to the expected ground state power system breaker circuit identification stage.
The group-based online optimization method for the optimal grid topology taking into account expected accident sets according to claim 6, characterized in that the expected base state power system disconnection line identification stage is an application stage identification method for the expected base state power system Disconnected line identification, specifically including the following steps:

S71, scanning step: use sensitivity index to screen the alternative disconnection lines, and the sensitivity index calculation method is as follows:

Calculate the sensitivity one by one for the alternative breaking lines, and set the sensitivity to be greater than or equal to zero, that is
The interrupted transmission line is sent to the next step to continue calculation;

S72. Sorting step: use the load margin estimation method to estimate the load margin of the alternative circuit breakers screened out in the previous step, and sort according to the calculation results from large to small, and sort the top alternative breaking lines The line is sent to the next step to continue analysis and calculation;

S73, identification step: use the continuous power flow method to accurately calculate the load margin of the alternative circuit breakers sorted in the previous step, sort according to the calculated load margin value, and send the identification result to the verification stage to continue the analysis.
The group-based online optimization method for the optimal grid topology taking into account anticipated accident sets according to claim 7, wherein the verification phase includes the following steps:

S8, use the recognition result of the previous stage as the solution set, perform a fast predicted accident analysis, and verify whether the load margin of all predicted accidents meets the threshold requirement:

S81, extract a transmission line breaking plan from the solution set, generate a new grid topology, and send it to the next step;

S82, the system after the transmission line is disconnected is regarded as the network structure of a new expected base state power system, and the expected accident analysis and calculation of the power system after the interruption are performed according to the input data. If the load margin of all expected accidents meets the formula (6) If the threshold is required, the disconnected transmission line solution is retained, otherwise it is removed from the solution set;

S83: Judge whether all the solutions in the solution set have been verified, if yes, go to step S9, otherwise continue to perform step S81;

S9, if k=m, go to step S10; otherwise, set the counter k=k+1, and generate multiple disconnection line candidate sets at the same time, the generation method is as follows:

CA k ＝A k ×A 1 (18)

In the formula, CA k is an alternative breaking plan that allows k lines to be opened, A k-1 and A 1 are the recognition results when the number of disconnected lines is k-1 and 1, respectively, and formula (18) is expressed as The alternative breaking scheme that allows the breaking of k lines is the Cartesian product of the recognition results when the number of open lines is k-1 and one;

S10, output results, the output results include solutions for all effectively interrupted transmission lines, optimized load margins of the expected ground state power system, predicted failure set analysis results, and predicted accidents with minimum load margins.
A group-based online optimization method for the optimal grid topology taking into account the set of anticipated accidents, which is characterized by including the following steps:

Establish a mathematical model of the optimal circuit breaker for enhanced voltage stability that takes into account the set of anticipated accidents;

According to the real-time grid data given by the energy management system, the load data predicted based on historical data and the generator output plan, calculate the expected voltage stability margin value of the ground-state power system;

Use load margin estimation method to quickly estimate the voltage stability margin of all expected accidents in the expected accident set;

Use the characteristics of expected accident groups to divide all expected accidents into groups, and select representatives of expected accidents in each group;

According to whether the voltage stability margin of the expected ground state power system and the expected accident set meets the limit requirements, different calculations are performed according to the situation;

Identify the disconnected lines that enable all expected accident representatives to meet the expected accident set load margin limit;

Identify the open circuit that maximizes the expected base state power system load margin;

Perform analysis of anticipated accidents;

Output the result.
The group-based online optimization method for the optimal power grid topology taking into account the expected accident set according to claim 9, characterized in that the mathematical model of the optimal interrupting line for the enhanced voltage stability that takes into account the expected accident set is established. Specifically:

Given the current power system status and data, short-term predicted load, power generation plan, grid maintenance plan, a set of anticipated accident sets, alternative breaking line plans and the maximum allowable number of breaking lines;

The objective function of the mathematical model of the optimal circuit breaker with enhanced voltage stability considering the expected accident set is:

The continuous power flow balance equation of the expected ground state power system after the transmission line is disconnected is:

The load margin limit requirements of the expected ground state power system after the transmission line is disconnected are:

λ b ≥λ th,b (3)

The safe operation constraints of the expected ground state power system after the transmission line is disconnected are:

It is envisaged that the continuous power flow balance equation of the accident system is:

The load margin limit requirements for all anticipated accident systems are:

The restriction conditions for the number of open circuits are:

NE(N b -N)≤m (7)

In the formula, B is the set of power system nodes; C is the given set of predicted accidents; c i is the i-th predicted accident in the set of predicted accidents, and c i ∈ C;
And λ b respectively represent the load margin of the predicted accident c i and the expected ground state power system; λ th,c and λ th,b are the load margin limits of the predicted accident and the expected ground state power system respectively; N b and N are respectively The expected ground state power network after the transmission line is disconnected and the current power network of the power system; NE (N b -N) is the difference between the grid topology N b and N, that is, the circuit break plan, and m represents the allowable number of breaking the transmission line; V i, V i, min , V i, max represents the voltage amplitude of the node i, respectively, the lower and upper limits permissible voltage; S (i, j) and S (i, j), max Respectively represent the power and the maximum power allowed to flow through the power transmission line ij connected between nodes i and j.