Port micro-grid power line switching and screening method

Abstract

The invention discloses a port micro-grid power line switching and screening method, which is suitable for a port micro-grid, and comprises the following steps: carrying out continuous power flow calculation on a target power system, obtaining the sensitivity of the power system at a saddle node bifurcation point, and reserving a cut-off line with positive sensitivity through screening; acquiring a load margin estimated value corresponding to each cut-off line for the cut-off line with positive sensitivity; according to the estimated value of the load margin, carrying out initial sorting on the cut-off lines with positive sensitivity; and based on a continuation power flow method, carrying out load margin analysis on the cut-off lines of the previous preset ranking in the initial ranking, and reranking the cut-off lines of the previous preset ranking in the initial ranking according to the load margin. Based on the method, continuous power flow calculation does not need to be carried out on all lines, the line switching calculation complexity is reduced, and the line switching screening workload is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of port microgrid power system design, transformation and application, and more specifically, to a method for screening the switching of port microgrid power lines. Background Technique

[0002] At present, the port island power system, as an off-grid microgrid, is different from the grid-connected power system and has the characteristics of a complex working environment, frequent switching of operating conditions, and a small grid capacity. The main task of the port power network is to provide power supply for the port area, that is, the port power system should be able to stably provide reliable power supply, meet the power demands of various equipment in the port area, and ensure the normal operation of the port area. The change of grid line switching affects the node voltage and the flow of bus power. Therefore, changing the line opening and closing state can curb the over-limit of the bus voltage in the grid to enhance system stability.

[0003] However, the current port island power system has the following problems:

[0004] (1) The port power system is a typical island power system. The switching of grid lines will surely affect the operation stability and power quality of the system. Reasonable planning can improve the grid voltage state. Nowadays, there are various methods for calculating the switching of power system lines, but most of the calculations are relatively complex;

[0005] (2) Most of the research on power system topology identification and optimization focuses on equipment to judge line and system states, and there are few optimization methods for calculating the impact of line switching on system load margin in the theoretical stage. When there are too many lines in the system, continuous power flow calculations need to be carried out for each switching condition of each line, and the calculation amount is relatively large.

[0006] Therefore, how to preliminarily screen all possible switching conditions to reduce the complexity of line switching calculation, so as to greatly reduce the workload of line switching screening, is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0007] In view of the above problems, the present invention provides a method for screening the switching of port microgrid power lines to solve at least some of the technical problems mentioned in the above background technique.

[0008] In order to achieve the above object, the present invention adopts the following technical scheme:

[0009] The present invention provides a method for screening the switching of port microgrid power lines, including the following steps:

[0010] S1. Perform continuous power flow calculation on the target power system to obtain the sensitivity at the saddle-node bifurcation point of the power system, and retain the opened lines with positive sensitivity through screening;

[0011] S2. For the opened lines with positive sensitivity, obtain the estimated load margin corresponding to each opened line; perform an initial sorting on the opened lines with positive sensitivity according to the estimated load margin;

[0012] S3. Based on the continuation power flow method, perform a load margin analysis on the opened lines in the top preset rankings in the initial sorting, and re-sort the opened lines in the top preset rankings in the initial sorting according to the load margin.

[0013] Further, in step S1, the performing a continuation power flow calculation on the target power system to obtain the sensitivity of the power system at the saddle-node bifurcation point specifically includes:

[0014] (1) Construct a parameterized continuation power flow equation for the target power system, expressed as:

[0015] 0 = f(x, p, λ) = f(x, p) - λb

[0016] where x represents the voltage state vector of the target power system; λ represents the load growth factor, and λ ∈ R 1 , R 1 represents the set of real numbers in one-dimensional space; p represents the opened line parameters in the target power system, including the line conductance, susceptance, and shunt capacitance of the opened line; b represents the changes in the actual power load, reactive power load demand, and actual power generation, and b ∈ R n ;

[0017] (2) Perform a Taylor series expansion on the continuation power flow equation at the saddle-node bifurcation point to obtain:

[0018]

[0019] where, represents the Jacobian matrix of the target power system at the saddle-node bifurcation point; represents the partial derivative matrix of the continuation power flow equation with respect to the opened line parameter p; represents the partial derivative matrix of the continuation power flow equation with respect to the load growth factor λ; Δx represents the change in the voltage state vector; Δp represents the change in the opened line parameter; Δλ represents the change in the load growth factor, that is, represents the sensitivity of the target power system at the saddle-node bifurcation point;

[0020] (3) Set the coordinates of the saddle-node bifurcation point as (x * , λ * ), where x * represents the voltage state vector corresponding to the bifurcation point; λ * represents the system load margin corresponding to the saddle-node bifurcation point; at the saddle-node bifurcation point (x * , λ* ) At this point, the Jacobian matrix is singular, and there exists a left eigenvector ω corresponding to the zero eigenvalue of a Jacobian matrix , thus satisfying:

[0021]

[0022] Multiply both sides of this formula on the left by the eigenvector corresponding to the zero eigenvalue, and we get:

[0023]

[0024] Finally, the sensitivity Δλ of the target power system at the saddle-node bifurcation point is obtained and expressed as:

[0025]

[0026] Furthermore, for a target power system with n load nodes and m branches, the partial derivative matrix is a matrix of order (2n 1 + n 2 ) * m; where, n 1 represents the number of PQ load nodes; n 2 represents the number of PV load nodes.

[0027] Furthermore, in step S2, for the opened lines with positive sensitivity, the load margin estimation value corresponding to each opened line is obtained, specifically:

[0028] (1) In the target power system, for the opened line connecting load node i and load node j, during the derivation process, for load node i, this opened line is denoted as the opened line i-j, and for load node j, this opened line is denoted as j-i;

[0029] For the opened line i-j, only the active power balance equations and reactive power balance equations of load nodes i and j are directly related to the opened line parameter p; the active power balance equations and reactive power balance equations of load nodes i and j are expressed as:

[0030]

[0031] where, ΔP i represents the active power flowing into load node i after the opening of the opened line i-j; ΔQ i represents the reactive power flowing into load node i after the opening of the opened line i-j; ΔP j represents the active power flowing into load node j after the opening of the opened line i-j; ΔQ jDenotes the reactive power flowing from the interrupted line i-j to the load node j after interruption; P 0i Denotes the active power flowing from the interrupted line i-j to the load node i before interruption; Q 0i Denotes the reactive power flowing from the interrupted line i-j to the load node i before interruption; P 0j Denotes the active power flowing from the interrupted line i-j to the load node j before interruption; Q 0j Denotes the reactive power flowing from the interrupted line i-j to the load node j before interruption; n represents the total number of load nodes; V i * Denotes the voltage value of the load node i; Denotes the voltage value of the load node j; Denotes the voltage phase angle difference of the load node i with respect to the load node j; Denotes the voltage phase angle difference of the load node j with respect to the load node i; G ij Denotes the real part of the network admittance between the load nodes i and j; B ij Denotes the imaginary part of the network admittance between the load nodes i and j; G ji Denotes the real part of the network admittance between the load nodes j and i; B ji Denotes the imaginary part of the network admittance between the load nodes j and i; λ * Denotes the system load margin corresponding to the saddle-node bifurcation point; ΔP Gi Denotes the variation trend of the generator injection at the load node i; ΔP Li Denotes the variation trend of the active power at the load node i; ΔQ Li Denotes the variation trend of the reactive power at the load node i; ΔP Gj Denotes the variation trend of the generator injection at the load node j; ΔP Lj Denotes the variation trend of the active power at the load node j; ΔQ Lj Denotes the variation trend of the reactive power at the load node j;

[0032] (2) Differentiate the active power balance equation and the reactive power balance equation of the load nodes i and j with respect to the parameters of the interrupted line i-j, respectively, then we have:

[0033]

[0034] Similarly, we can obtain:

[0035]

[0036] Among them, p ij Denotes the parameters of the interrupted line i-j, including the line conductance, susceptance and shunt capacitance of the interrupted line i-j; Δp ijIndicates the change in the parameters of the interrupted line i-j; g ij Indicates the branch conductance of the interrupted line i-j; b ij Indicates the branch susceptance of the interrupted line i-j; b ij0 Indicates half of the shunt susceptance to ground of the interrupted line i-j; P ij Indicates the active power flowing from the load node i to the load node j in the interrupted line i-j; Q ij Indicates the reactive power flowing from the load node i to the load node j in the interrupted line i-j; p ji Indicates the parameters of the interrupted line j-i, including the line conductance, susceptance and shunt capacitance of the interrupted line j-i; Δp ji Indicates the change in the parameters of the interrupted line j-i; P ji Indicates the active power flowing from the load node j to the load node i in the interrupted line j-i; Q ji Indicates the reactive power flowing from the load node j to the load node i in the interrupted line j-i;

[0037] (3) Define the vector The change in the load margin caused by the interruption of the transmission line between the load node i and the load node j is obtained as follows:

[0038]

[0039] Among them, Δλ ij Indicates the change in the load margin corresponding to the interrupted line i-j; Indicates the vector The element representing the active power balance equation of the load node i in; Indicates the vector The element representing the reactive power balance equation of the load node i in; Indicates the vector The element representing the active power balance equation of the load node j in; Indicates the vector The element representing the reactive power balance equation of the load node j in.

[0040] Furthermore, in step S3, based on the continuous power flow method, the load margin analysis is performed on the interrupted lines in the top preset ranking in the initial ranking, specifically including:

[0041] For the interrupted lines in the top preset ranking in the initial ranking, obtain the first power flow solution (V i,1 , λ 1 ) and the second power flow solution (V i,2 , λ 2 ); among them, V i,1 Indicates the voltage value of the grid bus under the first operating condition; λ 1Indicates the load margin under the first operating condition; V i,2 Indicates the voltage value of the power grid bus under the second operating condition; λ 2 Indicates the load margin under the second operating condition;

[0042] Among the positions corresponding to the first power flow solution (V i,1 , λ 1 ) and the second power flow solution (V i,2 , λ 2 ), taking the load node with the largest voltage magnitude change as a reference, perform P-V curve fitting to obtain the load margin of the corresponding opened line.

[0043] Furthermore, taking the load node with the largest voltage magnitude change as a reference, performing P-V curve fitting to obtain the load margin of the corresponding opened line specifically includes:

[0044] (1) Obtain the load node i with the largest voltage magnitude change, expressed as:

[0045]

[0046] where ΔV i Indicates the voltage change of the power grid bus under the first and second operating conditions;

[0047] (2) Substitute the first power flow solution (V i,1 , λ 1 ) and the second power flow solution (V i,2 , λ 2 ) into the formula of load node i to obtain:

[0048]

[0049] where Indicates the derivative of V i,2 ;

[0050] Further obtain the first coefficient α, the second coefficient β, and the third coefficient γ;

[0051] (3) According to the first coefficient α, the second coefficient β, and the third coefficient γ, calculate the estimated load margin λ * of the corresponding opened line, expressed as:

[0052] λ * = α - β 2 / 4γ

[0053] (4) Judge the estimated load margin λ * :

[0054] If the estimated load margin λ * and λ in the second power flow solution 2If the relationship conforms to the preset relational formula, then the estimated load margin λ * is used as the final load margin;

[0055] If the estimated load margin λ * and the λ 2 in the second power flow solution do not conform to the preset relational formula, it indicates that the distance between the second power flow solution and the saddle-node bifurcation point is farther than the threshold. At this time, the second power flow solution is used as the new first power flow solution, and a new second power flow solution is obtained again; and according to the new first power flow solution and the new second power flow solution, a new estimated load margin is recalculated until the relationship between the new estimated load margin and the λ 2 in the new second power flow solution conforms to the preset relational formula, and the new estimated load margin is used as the final load margin.

[0056] Furthermore, the preset relational formula is expressed as:

[0057]

[0058] where ε represents a real number.

[0059] Furthermore, the obtaining of the new second power flow solution specifically includes:

[0060] 1) According to the estimated load margin λ * and the λ 2 in the second power flow solution, a new load growth coefficient change amount is obtained, expressed as:

[0061] Δλ = μ(λ * - λ 2 )

[0062] where μ represents a step control coefficient, and μ < 1;

[0063] 2) By taking the derivative of the voltage state vector x through the continuation power flow equation, we obtain:

[0064]

[0065] 3) Combining the new load growth coefficient change amount formula with the derivative formula of the voltage state vector x obtained from the continuation power flow equation, a new second power flow solution is obtained, expressed as:

[0066]

[0067] where, represents the new second power flow solution; x 2 represents the second power flow solution.

[0068] Through the above technical solutions, it can be seen that compared with the prior art, the present invention discloses a method for screening the switching of power lines in a port microgrid, which has the following beneficial effects:

[0069] The present invention takes into account the impact of the switching of power lines in a port microgrid on the stability of the power system, and designs a simple screening method for calculating the impact of line switching on the steady-state stability of the system. First, the opened lines are preliminarily screened based on sensitivity; then, an initial ranking is performed based on the estimated load margin value, and the opened lines in the top preset ranking in the initial ranking are selected to achieve secondary screening; finally, the load margin of the opened lines remaining after the secondary screening is calculated based on the continuous power flow method for final ranking, and the one with the highest priority in the ranking is the optimal opened line. Based on the method provided by the present invention, it is not necessary to perform continuous power flow calculations on all lines, reducing the computational complexity of line switching and greatly reducing the workload of line switching screening.

[0070] The technical solution of the present invention will be further described in detail below through the accompanying drawings and embodiments. Description of the Drawings

[0071] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative efforts.

[0072] Figure 1 Schematic diagram of the flow of the power line switching screening method for the port microgrid provided by the embodiment of the present invention.

[0073] Figure 2 Schematic diagram of the composition of the port area power system provided by the embodiment of the present invention.

[0074] Figure 3 Schematic diagram of the ETAP model of the port area power system provided by the embodiment of the present invention.

[0075] Figure 4 Schematic diagram of the P-V curve provided by the embodiment of the present invention.

[0076] Figure 5 Schematic diagram of the comparison of the system P-V curves obtained based on continuous power flow calculations provided by the embodiment of the present invention. Detailed Embodiments

[0077] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0078] An embodiment of the present invention discloses a method for screening the switching of power lines in a port microgrid, as Figure 1 shown, including the following steps:

[0079] S1. Perform a continuous power flow calculation on the target power system to obtain the sensitivity at the saddle-node bifurcation point of the power system, and retain the open-circuit lines with positive sensitivity through screening;

[0080] S2. For the open-circuit lines with positive sensitivity, obtain the estimated load margin value corresponding to each open-circuit line; perform an initial sorting on the open-circuit lines with positive sensitivity according to the estimated load margin value;

[0081] S3. Based on the continuous power flow method, perform a load margin analysis on the open-circuit lines in the top preset ranking in the initial sorting, and re-sort the open-circuit lines in the top preset ranking in the initial sorting according to the load margin.

[0082] In the embodiment of the present invention, first, the open-circuit lines in the target power system are screened based on sensitivity in step S1, which can greatly reduce the subsequent workload. In step S2, an initial sorting is performed based on the estimated load margin value. Since the estimated load margin value is obtained in this stage, there may be errors in its sorting. Therefore, in step S3, the most accurate continuous power flow method is used to recalculate and sort the open-circuit lines in the top preset ranking in the initial sorting again. Since the calculation amount of the continuous power flow method is too large, after two screenings through step S1 and step S2 in the embodiment of the present invention, the continuous power flow method is used in step S3 for the final calculation, thereby reducing the complexity of the line switching calculation, achieving a significant reduction in the workload of line switching screening, and contributing to the goal of efficiently and conveniently improving the static stability of the off-grid port area system.

[0083] Next, each of the above steps will be described in detail.

[0084] In the above step S1, to efficiently judge the specific impact of the disconnection of each open-circuit line on the static stability of the target power system, this step derives a load margin evaluation method based on sensitivity analysis to achieve a rapid preliminary screening of the line disconnection behavior. Specifically, perform a continuous power flow calculation on the target power system to obtain the sensitivity at the saddle-node bifurcation point of the power system, and retain the open-circuit lines with positive sensitivity through screening;

[0085] S11. Perform a continuous power flow calculation on the target power system to obtain the sensitivity at the saddle-node bifurcation point of the power system, specifically including:

[0086] (1) Construct a parameterized continuous power flow equation for the target power system, expressed as:

[0087] 0 = f(x, p, λ) = f(x, p) - λb (1)

[0088] Where x represents the voltage state vector of the target power system; λ is a variable continuous parameter, representing the load growth factor here, and λ ∈ R 1 , R 1 represents the set of real numbers in one-dimensional space; p represents the open-circuit line parameters in the target power system, including the line conductance, susceptance, and shunt capacitance to ground of the open-circuit line; b represents the changes in the actual active power load and reactive power load demands, as well as the change in the actual power generation, and b ∈ R n ;

[0089] (2) Perform a Taylor series expansion of the continuous power flow equation (i.e., Equation (1)) at the saddle-node bifurcation point (SNB) to obtain:

[0090]

[0091] Where represents the Jacobian matrix of the target power system at the saddle-node bifurcation point; represents the partial derivative matrix of the continuous power flow equation with respect to the open-circuit line parameter p; represents the partial derivative matrix of the continuous power flow equation with respect to the load growth factor λ; Δx represents the change in the voltage state vector; Δp represents the change in the open-circuit line parameter; Δλ represents the change in the load growth factor, that is, the sensitivity of the target power system at the saddle-node bifurcation point;

[0092] (3) Set the coordinates of the saddle-node bifurcation point as (x * , λ * ), where x * represents the voltage state vector corresponding to the bifurcation point; λ * represents the system load margin corresponding to the saddle-node bifurcation point; at the saddle-node bifurcation point (x * , λ * ), the Jacobian matrix is singular, and there exists a left eigenvector ω corresponding to the zero eigenvalue of the Jacobian matrix , so it satisfies:

[0093]

[0094] Multiply both sides of this equation on the left by the eigenvector corresponding to the zero eigenvalue to obtain:

[0095]

[0096] Finally, obtain the sensitivity Δλ of the target power system at the saddle-node bifurcation point, expressed as:

[0097]

[0098] Among them, for a target power system with n load nodes and m branch lines, the partial derivative matrix is a matrix of order (2n 1 + n 2 ) * m; where n 1 represents the number of PQ load nodes; n 2 represents the number of PV load nodes.

[0099] S12. Retain the opened lines with positive sensitivity by screening; specifically, use a linear sensitivity method to scan all alternative opened lines. If the sensitivity is positive after opening, retain the alternative opened line; otherwise, screen it out.

[0100] In step S1 above, the opened line conditions are screened through sensitivity analysis, and the opened lines with positive sensitivity are retained, that is, the lines that may increase the load margin after being cut are screened out. Although sensitivity analysis can make a preliminary qualitative judgment on the candidate cut lines, it cannot provide a quantitative analysis of the cut operation. Therefore, in the embodiment of the present invention, in step S2, the opened lines retained by screening are initially sorted by obtaining the load margin estimation value.

[0101] In step S2 above, for the opened lines with positive sensitivity, obtain the load margin estimation value corresponding to each opened line; initially sort the opened lines with positive sensitivity according to the load margin estimation value;

[0102] S21. For the opened lines with positive sensitivity, obtain the load margin estimation value corresponding to each opened line, specifically:

[0103] (1) In the target power system, for the opened line connecting load node i and load node j, during the derivation process, for load node i, this opened line is denoted as the opened line i - j, and for load node j, this opened line is denoted as j - i;

[0104] For the opened line i - j, only the active power balance equations and reactive power balance equations of load nodes i and j are directly related to the opened line parameter p; therefore, the above partial derivative matrix has only 4 non-zero elements, that is and Based on this, the active power balance equations and reactive power balance equations of load nodes i and j are expressed as:

[0105]

[0106] where ΔP iIt represents the active power flowing to the load node i after the disconnected line ij is disconnected; ΔQ i It represents the reactive power flowing to the load node i after the disconnected line ij is disconnected; ΔP j It represents the active power flowing to load node j after the disconnected line ij is disconnected; ΔQ j P represents the reactive power flowing to load node j after disconnection of line ij; 0i It represents the active power flowing to the load node i before the disconnected line ij is disconnected; Q 0i P represents the reactive power flowing to load node i before disconnecting line ij; 0j It represents the active power flowing to load node j before disconnecting line ij; Q 0j represents the reactive power flowing to load node j before disconnection of line ij; n represents a total of n load nodes; V i * represents the voltage value of load node i; represents the voltage value of load node j; Represents the voltage phase angle difference between load node i and load node j; G represents the voltage phase angle difference of load node j to load node i; ij represents the real part of the network admittance between load node i and load node j; B ij represents the imaginary part of the network admittance between load node i and load node j; G ji represents the real part of the network admittance between load node j and load node i; B ji represents the imaginary part of the network admittance between load node j and load node i; * represents the system load margin corresponding to the saddle node bifurcation point; ΔP Gi Indicates the generator injection change trend of load node i; ΔP Li Indicates the active power change trend of load node i; ΔQ Li Indicates the reactive power change trend of load node i; ΔP Gj Indicates the generator injection change trend of load node j; ΔP Lj Indicates the active power change trend of load node j; ΔQ Lj Indicates the reactive power change trend of load node j;

[0107] (2) The active power balance equation and reactive power balance equation of load node i and load node j, i.e., the above formulas (6a)-(6d), are respectively differentiated with respect to the parameters of the disconnected line ij, and we have:

[0108]

[0109] Similarly, we can get:

[0110]

[0111] Among them, p ij Indicates the parameters of the disconnected line ij, including the line conductance, susceptance and ground accommodation of the disconnected line ij; Δp ij Indicates the change in the disconnected line ij parameter; g ij represents the branch conductance of the disconnected line ij; b ij Indicates the branch susceptance of the disconnected line ij; b ij0 P represents half of the earth susceptance of the disconnected line ij; ij It represents the active power flowing from load node i to load node j in the disconnected line ij; Q ij represents the reactive power flowing from load node i to load node j in the disconnected line ij; p ji Indicates the parameters of the disconnected line ji, including the line conductance, susceptance and ground accommodation of the disconnected line ji; Δp ji Indicates the change in the parameters of the disconnected line; P ji It represents the active power flowing from load node j to load node i in the disconnected line ji; Q ji represents the reactive power flowing from load node j to load node i in the disconnected line ji; according to the derivation of the above formulas (7a)-(7d), the load node i can be obtained and In terms of value, they are equal to P ij and Q ij ; Load node j and Equal to P in value ji and Q ji ;

[0112] (3) Define vector The parameters It is just an auxiliary parameter of the formula, which is only used to shorten the length of the formula to facilitate subsequent explanation. Based on the defined formula, the load margin change caused by the disconnection of the transmission line between load node i and load node j is obtained as follows:

[0113]

[0114] Among them, Δλ ij Indicates the load margin change corresponding to the disconnected line ij; Representation vector where represents the element of the active power balance equation of load node i; Representation vector represents the element of reactive power balance equation of load node i; Representation vector where represents the element of the active power balance equation of load node j; Representation vector represents the element of the reactive power balance equation of load node j;.

[0115] S22. Initially sort the disconnected lines with positive sensitivity according to the load margin estimation value.

[0116] The above step S2 is a sorting based on the load margin estimation value, which may not have a small error. Therefore, in the embodiment of the present invention, in step S3, the Look-ahead load margin method is also used to calculate the load margin of the disconnected line. The method is based on the fact that the PV curve near the saddle node bifurcation can be approximately regarded as a part of the quadratic curve, so as to calculate the load margin.

[0117] In the above step S3, based on the continuous power flow method, the load margin analysis is performed on the disconnected lines with the previous preset ranking in the initial sorting, and the disconnected lines with the previous preset ranking in the initial sorting are re-sorted according to the load margin;

[0118] S31. Based on the continuous power flow method, load margin analysis is performed on the disconnected lines with the first preset ranking in the initial sorting, specifically:

[0119] Different from the traditional curve fitting method, the look-ahead load margin calculation technology only uses two power flow solutions and the load margin derivative corresponding to the second power flow solution to estimate the load margin. In the embodiment of the present invention:

[0120] First, for the disconnected lines with the highest preset ranking in the initial sorting, the first power flow solution (V i,1 ,λ 1 ) and the second power flow solution (V i,2 ,λ 2 ), where V i,1 Indicates the voltage value of the power grid bus under the first working condition; 1 Indicates the load margin under the first working condition; V i,2 Indicates the voltage value of the power grid bus under the second working condition; 2 Indicates the load margin under the second working condition;

[0121] Secondly, in order to improve the fitting accuracy and considering the differences in the PV curves of each node, in the first power flow solution (V i,1 ,λ 1 ) and the second power flow solution (V i,2 ,λ 2 ) corresponding to the position, taking the load node with the largest voltage amplitude change as a reference, PV curve fitting is carried out to obtain the load margin of the corresponding disconnected line; specifically including:

[0122] (1) Obtain the load node i with the largest voltage amplitude change. The selection criteria of the load node i follow the formula (9):

[0123]

[0124] Where, ΔV i Indicates the voltage change of the power grid bus under the first working condition and the second working condition;

[0125] (2) Solve the first power flow (V i,1 ,λ 1 ) and the second power flow solution (V i,2 ,λ 2 ) is substituted into the load node i formula (i.e. the above formula (9)), and we get:

[0126]

[0127] in, Indicates V i,2 The derivative of

[0128] Further solving the above formula (10) can obtain the first coefficient α, the second coefficient β and the third coefficient γ; these three coefficients are auxiliary symbols for mathematical calculations and have no special meanings. They can be interpreted as the coefficients of the mathematical equation of the curve V-λ;

[0129] (3) Based on the first coefficient α, the second coefficient β and the third coefficient γ, calculate the estimated load margin λ of the corresponding disconnected line * , expressed as:

[0130] λ * =α-β 2 / 4γ (11)

[0131] (4) Estimated load margin λ * Make a judgment:

[0132] For the estimated load margin λ * and the second power flow solution 2 The relationship between the two is set as follows:

[0133]

[0134] Among them, ε represents a real number;

[0135] If the estimated load margin λ * and the second power flow solution 2 If the relationship between meets the preset relationship, the estimated load margin λ * As the final load margin;

[0136] If the estimated load margin λ * and the second power flow solution 2 If the relationship between does not conform to the preset relationship, it means that the distance between the second power flow solution and the saddle node bifurcation point is far greater than the threshold. At this time, the second power flow solution is used as the new first power flow solution, and the new second power flow solution is re-obtained; and the new estimated load margin is recalculated according to the new first power flow solution and the new second power flow solution until the new estimated load margin is equal to λ in the new second power flow solution. 2 The relationship between the load margin and the load margin complies with the preset relationship, and the new estimated load margin is used as the final load margin;

[0137] The above-mentioned re-obtaining of a new second power flow solution specifically includes:

[0138] 1) Based on the estimated load margin λ * and the second power flow solution 2 , get the new load growth factor change, expressed as:

[0139] Δλ=μ(λ * -λ 2 ) (13)

[0140] Wherein, μ represents the step size control coefficient, and μ<1;

[0141] 2) Derivative the voltage state vector x through the continuous power flow equation to obtain:

[0142]

[0143] 3) Combine the new load growth coefficient change formula with the derivative formula of the continuous power flow equation to the voltage state vector x, that is, substitute the above formulas (13) and (14) into the following formula (15) to obtain the new second power flow solution, which is expressed as:

[0144]

[0145] in, represents the new second power flow solution; x 2 represents the second power flow solution.

[0146] S32. Reorder the disconnected lines with a preset ranking in the initial order according to the load margin to obtain a final priority order of the disconnected lines; wherein the disconnected line corresponding to the highest priority is the best disconnected line.

[0147] Next, a specific embodiment is used to verify the port microgrid power line switching screening method provided by the present invention.

[0148] 1. Introduction to target power generation system:

[0149] Taking the off-grid power generation system of a port area as an example, its components include generators, distribution boxes, loads, quay cranes, etc. The schematic diagram of the power system is as follows: Figure 2 shown. Figure 2 In the figure, LOAD represents the system load; SS1, SS2 and SS3 represent the three main power consumption areas of the power grid; SPARE represents the vacant interface; STS1-STS7 represent the seven quay cranes at the port. The core of the power system is the generator and the quay crane. The eight generators are not put into operation on a daily basis, but are put into operation according to the use of the quay crane. In addition, it is equipped with protection devices such as circuit breakers and fuses to protect the power generation equipment and various power consumption equipment.

[0150] 2. Build the corresponding single-line diagram model:

[0151] Figure 3 Built on ETAP power system software Figure 2 Single-line diagram model of the system. Figure 3 In the diagram, DG1 and DG3 represent generator sets; MSB-SS1-1 represents busbar SS1-1; MSB-SS2-2 represents busbar SS1-2; the same applies to the rest of MSB-XX; M1-M4 represent transformer busbars; 1TR1 represents transformer 1 under busbar SS1; 1TR2 represents transformer 2 under busbar SS1; the same applies to the rest of STS1-STS7 represent 7 quay cranes at the port. The steps to build the single-line diagram model are as follows:

[0152] (1) The port area power grid is divided into several structural levels from primary to secondary and from core to terminal according to the voltage level and purpose of the network;

[0153] (2) According to the drawings, select the appropriate structural level as the main frame of the model, i.e., the busbar;

[0154] (3) The equivalent power grid and equivalent load are used to represent the external network that transmits active power to the power grid and the load within the same area.

[0155] (4) Select a voltage level network of a specific scale and structural level and express it in the form of equivalent load. In practical applications, the network model can be expanded to a lower level according to needs. Following the partitioning principle, the network structure is subdivided into multiple areas, and the modeling of each area is processed in blocks.

[0156] (5) Each composite network is connected by transmission lines on the single-line diagram, and each component is edited within the composite network;

[0157] (6) Input the real data of the equipment into each component editor: each component includes busbar, cable, transformer, generator, load, capacitor, reactor, etc.;

[0158] (7) Improve the position and opening and closing status of the circuit breaker according to the drawings.

[0159] III. Description of the Effects of the Invention:

[0160] The port system has a total of 139 branches. The total system load is 7.098MW and 1.406Mvar, and the generator power generation is 7.263MW and 1.678Mvar. In the test, the upper and lower limits of each node voltage are set to 1.05pu and 0.95pu. Line switching operations are performed on all branches, and the bus SS3 is calculated using the new method of this patent and the traditional continuous power flow method (CPFLOW) to obtain a switching condition with the best effect of improving the system load margin.

[0161] The traditional method of analyzing line switching uses the continuous power flow method to calculate the bus voltage while changing the load of the selected bus to obtain the PV curve. If the load is increased to the load capacity limit and then reduced back, the PV curve of the entire power supply voltage can be traced. Figure 4 As shown, only the operating points above the critical point represent a stable system. Figure 4 In the PV curve shown, P is the total load of the system, and V is the bus voltage; Predictor is prediction, and Corrector is correction. Both are steps of the continuous power flow method. The continuous power flow method is a classic algorithm in the prior art, so it will not be explained again. As the system load P gradually increases, the bus can hardly bear such a large load, so the voltage gradually decreases until the saddle node bifurcation point SNB, and the system collapses. Therefore, the upper half of the curve is a stable state, and the lower half is an unstable state. CPFLOW uses a prediction-correction strategy to determine the solution trajectory of the power flow equation. The core idea is to start from the existing operating point, and as the load gradually increases, the power flow (that is, the operating point of the system) is continuously solved through prediction and correction operations until the voltage collapse point (SNB) is reached. Finally, a complete PV curve is drawn, and the power flow solution (stability reserve) under the critical load state can also be obtained.

[0162] The power system line switching screening method provided by the present invention is applied to analyze the port area power system, and the results are as follows:

[0163] 1. After step S1, the operating conditions with negative sensitivity are removed, and finally 28 switching operating conditions with positive sensitivity are obtained.

[0164] 2. According to the load margin estimation value of each working condition estimated by the method of step S2, the working conditions are ranked, and the top 5 pass and enter the third stage for CPFLOW calculation.

[0165] Table 1 shows the load margin and ranking obtained by the look-ahead method and the load margin and ranking calculated by the traditional CPFLOW method.

[0166] Table 1 Ranking of the top 5 disconnected lines

[0167]

[0168] The PV curves of the top five interruption conditions of the expected power system and load margin are as follows: Figure 5 As shown, the data table is shown in Table 1. According to the results, the following conclusions can be drawn:

[0169] 1. The load margin of the expected power system is 3.096MW. The first-ranked breaking condition is line 56, with a load margin of 3.118MW, an increase of 22kw; the second-ranked breaking condition is line 44, with a load margin of 3.111MW, an increase of 15kw; the third-ranked breaking condition is line 58, with a load margin of 3.103MW, an increase of 7kw.

[0170] 2. The load margins and rankings of the switching lines obtained using the look-ahead method are similar to those obtained using the traditional CPFLOW method, which further verifies the high accuracy of the look-ahead method.

[0171] In summary, with the method of the present invention, there is no need to perform CPFLOW calculations on all lines. Instead, sensitivity and Look-ahead methods can be used for preliminary screening. Engineering designers can effectively save time for calculating line switching conditions, thereby maintaining system voltage stability and ensuring port safety.

[0172] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0173] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for selecting and switching power lines in a port microgrid, characterized in that: The steps include: S1. Continuous power flow calculation is performed on the target power system to obtain the sensitivity of the power system at the saddle node bifurcation point, and the disconnected lines with positive sensitivity are retained by screening; S2. For the disconnected lines with positive sensitivity, obtain the load margin estimation value corresponding to each disconnected line; and perform initial sorting of the disconnected lines with positive sensitivity according to the load margin estimation value; S3. Based on the continuous power flow method, a load margin analysis is performed on the disconnected lines with a previous preset ranking in the initial sorting, and the disconnected lines with a previous preset ranking in the initial sorting are re-sorted according to the load margin.

2. A method for selecting and switching power lines of a port microgrid according to claim 1, characterized in that: In step S1, the continuous power flow calculation is performed on the target power system to obtain the sensitivity of the power system at the saddle node bifurcation point, which specifically includes: (1) Construct a parameterized continuous power flow equation for the target power system, which is expressed as: 0=f(x,p,λ)=f(x,p)-λb Where x represents the voltage state vector of the target power system; λ represents the load growth factor, and λ∈R 1 , R 1 represents a set of real numbers in one-dimensional space; p represents the disconnected line parameters in the target power system, including the line conductance, susceptance and ground accommodation of the disconnected line; b represents the changes in the actual power load and reactive power load demand, as well as the changes in the actual power generation, and b∈R n ; (2) Performing Taylor series expansion on the continuous power flow equation at the saddle node bifurcation point, we obtain: in, represents the Jacobian matrix of the target power system at the saddle node bifurcation point; represents the partial derivative matrix of the continuous power flow equation with respect to the disconnected line parameter p; represents the partial derivative matrix of the continuous power flow equation to the load growth coefficient λ; Δx represents the change of the voltage state vector; Δp represents the change of the disconnected line parameter; Δλ represents the change of the load growth coefficient, that is, the sensitivity of the target power system at the saddle node bifurcation point; (3) Set the coordinates of the saddle node bifurcation point to (x * ,λ * ), where x * represents the voltage state vector corresponding to the bifurcation point; λ * represents the system load margin corresponding to the saddle node bifurcation point; at the saddle node bifurcation point (x * ,λ * ), the Jacobian matrix There is a singularity, there is a corresponding Jacobian matrix The left eigenvector ω of the zero eigenvalue therefore satisfies: The formula is multiplied on both sides. The eigenvector corresponding to the zero eigenroot is obtained: Finally, the sensitivity Δλ of the target power system at the saddle node bifurcation point is obtained, which is expressed as:

3. A method for selecting and switching power lines of a port microgrid according to claim 2, characterized in that: For a target power system with n load nodes and m branches, the partial derivative matrix It is a (2n1+n2)*m-order matrix, where n1 represents the number of PQ load nodes and n2 represents the number of PV load nodes.

4. A method for selecting and switching power lines of a port microgrid according to claim 2, characterized in that: In step S2, for the disconnected lines with positive sensitivity, the load margin estimation value corresponding to each disconnected line is obtained, specifically: (1) In the target power system, for the disconnected line connecting the load node i and the load node j, in the derivation process, for the load node i, the disconnected line is recorded as disconnected line ij, and for the load node j, the disconnected line is recorded as ji; For the disconnected line ij, only the active power balance equation and reactive power balance equation of load node i and load node j have a direct relationship with the disconnected line parameter p; the active power balance equation and reactive power balance equation of load node i and load node j are expressed as: Among them, ΔP i It represents the active power flowing to the load node i after the disconnected line ij is disconnected; ΔQ i It represents the reactive power flowing to the load node i after the disconnected line ij is disconnected; ΔP j It represents the active power flowing to load node j after the disconnected line ij is disconnected; ΔQ j P represents the reactive power flowing to load node j after disconnection of line ij; 0i It represents the active power flowing to the load node i before the disconnected line ij is disconnected; Q 0i P represents the reactive power flowing to load node i before disconnecting line ij; 0j It represents the active power flowing to load node j before disconnecting line ij; Q 0j It represents the reactive power flowing to load node j before disconnection of the disconnected line ij; n represents a total of n load nodes; represents the voltage value of load node i; represents the voltage value of load node j; Represents the voltage phase angle difference between load node i and load node j; G represents the voltage phase angle difference of load node j to load node i; ij represents the real part of the network admittance between load node i and load node j; B ij represents the imaginary part of the network admittance between load node i and load node j; G ji represents the real part of the network admittance between load node j and load node i; B ji represents the imaginary part of the network admittance between load node j and load node i; * represents the system load margin corresponding to the saddle node bifurcation point; ΔP Gi Indicates the generator injection change trend of load node i; ΔP Li Indicates the active power change trend of load node i; ΔQ Li Indicates the reactive power change trend of load node i; ΔP Gj Indicates the generator injection change trend of load node j; ΔP Lj Indicates the active power change trend of load node j; ΔQ Lj Indicates the reactive power change trend of load node j; (2) The active power balance equation and reactive power balance equation of load node i and load node j are differentiated with respect to the parameters of disconnected line ij, respectively, and we have: Similarly, we can get: Among them, p ij Indicates the parameters of the disconnected line ij, including the line conductance, susceptance and ground accommodation of the disconnected line ij; Δp ij Indicates the change in the disconnected line ij parameter; g ij represents the branch conductance of the disconnected line ij; b ij Indicates the branch susceptance of the disconnected line ij; b ij0 P represents half of the earth susceptance of the disconnected line ij; ij It represents the active power flowing from load node i to load node j in the disconnected line ij; Q ij represents the reactive power flowing from load node i to load node j in the disconnected line ij; p ji Indicates the parameters of the disconnected line ji, including the line conductance, susceptance and ground accommodation of the disconnected line ji; Δp ji Indicates the change in the parameters of the disconnected line; P ji It represents the active power flowing from load node j to load node i in the disconnected line ji; Q ji It represents the reactive power flowing from load node j to load node i in the disconnected line ji; (3) Define vector The load margin change caused by the disconnection of the transmission line between load node i and load node j is obtained as: Among them, Δλ ij Indicates the load margin change corresponding to the disconnected line ij; Representation vector where represents the element of the active power balance equation of load node i; Representation vector represents the element of reactive power balance equation of load node i; Representation vector where represents the element of the active power balance equation of load node j; Representation vector represents the element of the reactive power balance equation of load node j.

5. A method for selecting and switching power lines of a port microgrid according to claim 2, characterized in that: In step S3, the load margin analysis of the disconnected lines with the first preset ranking in the initial sorting is performed based on the continuous power flow method, specifically including: For the disconnected lines with the highest preset ranking in the initial sorting, the first power flow solution (V i,1 ,λ1) and the second power flow solution (V i,2 ,λ2); where V i,1 represents the voltage value of the power grid bus under the first working condition; λ1 represents the load margin under the first working condition; V i,2 represents the voltage value of the power grid bus under the second working condition; λ2 represents the load margin under the second working condition; In the first power flow solution (V i,1 ,λ1) and the second power flow solution (V i,2 ,λ2), the load node with the largest voltage amplitude change is taken as a reference to carry out PV curve fitting to obtain the load margin of the corresponding disconnected line.

6. A method for selecting and switching power lines of a port microgrid according to claim 5, characterized in that: The PV curve fitting is performed by taking the load node with the largest voltage amplitude change as a reference to obtain the load margin of the corresponding disconnected line, specifically including: (1) Obtain the load node i with the largest voltage amplitude change, expressed as: Where, ΔV i Indicates the voltage change of the power grid bus under the first working condition and the second working condition; (2) Solve the first power flow (V i,1 ,λ1) and the second power flow solution (V i,2 ,λ2) into the load node i formula, we get: in, Indicates V i,2 The derivative of Further obtaining a first coefficient α, a second coefficient β and a third coefficient γ; (3) According to the first coefficient α, the second coefficient β and the third coefficient γ, the estimated load margin λ of the corresponding disconnected line is calculated * , expressed as: l * =a-b 2 / 4c (4) Estimated load margin λ * Make a judgment: If the estimated load margin λ * If the relationship between λ2 and the second power flow solution is consistent with the preset relationship, the estimated load margin λ * As the final load margin; If the estimated load margin λ * If the relationship with λ2 in the second power flow solution does not conform to the preset relationship, it indicates that the distance between the second power flow solution and the saddle node bifurcation point is far greater than the threshold. At this time, the second power flow solution is used as the new first power flow solution, and the new second power flow solution is re-obtained; and based on the new first power flow solution and the new second power flow solution, the new estimated load margin is recalculated until the relationship between the new estimated load margin and λ2 in the new second power flow solution conforms to the preset relationship, and the new estimated load margin is used as the final load margin.

7. A method for selecting and switching power lines of a port microgrid according to claim 6, characterized in that: The preset relationship is expressed as: Here, ε represents a real number.

8. A method for selecting and switching power lines of a port microgrid according to claim 6, characterized in that: The re-obtaining of the new second power flow solution specifically includes: 1) Based on the estimated load margin λ * The new load growth factor change is obtained by combining λ2 in the second power flow solution, which is expressed as: Δλ=μ(λ * -λ2) Wherein, μ represents the step size control coefficient, and μ<1; 2) Derivative the voltage state vector x through the continuous power flow equation to obtain: 3) Combining the new load growth coefficient change formula with the derivative formula of the continuous power flow equation to the voltage state vector x, a new second power flow solution is obtained, which is expressed as: in, represents the new second power flow solution; x2 represents the second power flow solution.

Images