A cross-provincial and cross-regional power transaction plan scheduling optimization method and device based on redundant security constraint identification

Abstract

The application relates to the technical field of power dispatch, and discloses a cross-provincial and cross-regional power transaction plan dispatch optimization method and device based on redundant security constraint identification. The method obtains boundary conditions of a regional power grid dispatch plan, screens out non-redundant security constraints from all security constraints of the regional power grid dispatch problem, constructs a transaction plan feasible region model, solves the transaction plan feasible region model by using a vertex search method, obtains limit values of transaction plan components, takes the limit values as transmission power constraint limit values of the transaction plan components, and optimizes regional power grid transaction plan dispatch according to the transmission power constraint limit values of the transaction plan components. The method can improve the accuracy of transmission power limit values of the transaction plan components, guarantees the efficiency of regional power grid transaction plan feasible region calculation, optimizes regional power grid transaction plan dispatch, and improves the economic benefits of the regional power grid dispatch plan.

Description

Technical Field

[0001] This invention relates to the field of power dispatching technology, and in particular to a method and apparatus for optimizing the dispatching of inter-provincial power trading plans based on redundancy safety constraint identification. Background Technology

[0002] Due to differences in power resource endowments across provinces and regions, and the geographical mismatch between power resource distribution and load demand, inter-provincial power transmission and reception is highly necessary. However, the execution of inter-provincial agreed-upon power volumes and medium-to-long-term traded power volumes requires the prior formulation of day-ahead and real-time power dispatch plans to establish inter-provincial power transmission and reception plans, which then serve as the basis for implementation. Therefore, the formulation of inter-provincial power transmission and reception plans is a crucial component of the regional power grid dispatching agency's day-ahead and real-time dispatch planning.

[0003] Currently, before the formulation of regional power grid dispatch plans, the transmission power limits of transaction plan components are mostly set based on manual experience. However, this method often results in unreasonable transmission power limits for transaction plan components due to the inaccuracy of manual experience. This leads to the inability of the regional power grid's day-ahead dispatch plan (or regional power grid) calculation to directly obtain results that meet the power balance and grid security constraints of each province within the given calculation timeframe. Furthermore, existing methods for analyzing the feasible region of tie-line power require a complete consideration of the internal security constraints of the power grid during the optimization iteration process of each vertex. Therefore, each vertex search process requires solving a full-scale dispatch optimization problem. When the power grid size increases dramatically, especially the number of security constraints, the time consumed by each vertex search increases sharply, exceeding the time requirements for the formulation of the regional power grid's day-ahead dispatch plan. Summary of the Invention

[0004] This invention provides a method and apparatus for optimizing the scheduling of cross-provincial power trading plans based on redundancy safety constraint identification. It can improve the accuracy of the power transmission limit of the trading plan components, ensure the efficiency of the calculation of the feasible region of the regional power grid trading plan, optimize the scheduling of the regional power grid trading plan, and improve the economic benefits of the regional power grid scheduling plan.

[0005] To address the aforementioned technical problems, the first embodiment of the present invention provides a method for optimizing the scheduling of inter-provincial power trading plans based on redundancy security constraint identification, comprising:

[0006] Obtain the boundary conditions for the regional power grid dispatch plan;

[0007] Based on the preset redundancy security constraint identification and optimization model, non-redundancy security constraints of the regional power grid dispatching problem are screened from the regional power grid dispatching plan;

[0008] The selected non-redundant security constraints and the obtained boundary conditions are used to form the constraints of the feasible region of the transaction plan, and the feasible region model of the transaction plan is constructed.

[0009] A set of feasible solutions of the feasible domain model of the transaction plan is set as the initial vertex of the feasible domain of the transaction plan, and the vertex search method is used cyclically to search for new vertices in the feasible domain of the transaction plan to update the boundary range of the feasible domain of the transaction plan until the updated boundary range satisfies the preset loop stopping condition.

[0010] Based on the vertices of the current feasible domain of the transaction plan, the limit values ​​of the transaction plan components are obtained as the transmission power constraint limits of the transaction plan components, and the regional power grid transaction plan scheduling is optimized based on the transmission power constraint limits of the transaction plan components.

[0011] This invention obtains the boundary conditions of a regional power grid dispatch plan, filters out non-redundant security constraints from all security constraints in the regional power grid dispatch problem, constructs a feasible region model of the trading plan, solves the feasible region model using the vertex search method, obtains the limit values ​​of the trading plan components as the transmission power constraint limits of the trading plan components, and optimizes the regional power grid trading plan dispatch based on these transmission power constraint limits. This method avoids conflicts between manually set transmission power constraint limits of trading plan components and power balance constraints of each provincial power grid by constructing a feasible region model, thus improving the accuracy of the transmission power constraint limits of trading plan components. It also uses redundancy constraint identification technology to identify effective security constraints that directly affect the dispatch feasible region, ensuring the efficiency of calculating the feasible region of the regional power grid trading plan, optimizing the regional power grid trading plan dispatch, and improving the economic benefits of the regional power grid dispatch plan.

[0012] Furthermore, the step of selecting non-redundant security constraints for the regional power grid dispatching problem from the regional power grid dispatching plan based on the preset redundancy security constraint identification and optimization model specifically involves:

[0013] Obtain the load fluctuation range;

[0014] Using load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables, an indicator variable is introduced for each safety constraint to construct a redundant safety constraint identification and optimization model.

[0015] With the goal of minimizing the sum of indicator variables, the redundant safety constraint identification optimization model is solved iteratively, and the constraint conditions are updated according to the solved model results until all safety constraints in the constraint conditions are redundant safety constraints.

[0016] In the process of obtaining the redundancy safety constraint identification results, this invention also considers the load fluctuation range as a decision optimization variable, constructs a redundancy safety constraint identification optimization model, and solves the model. In each solution, non-redundant safety constraints are selected. The redundancy constraint identification results of this invention can accommodate the needs of scenarios where the load fluctuates within a certain range, making it more applicable.

[0017] Furthermore, the step of solving the redundant safety constraint identification optimization model and updating the constraint conditions based on the solved model results specifically involves:

[0018] After obtaining the solution results of the redundant safety constraint identification optimization model, it is checked whether there are indicator variables with a value of 0 in the redundant safety constraint identification optimization model.

[0019] If it exists, then the safety constraint corresponding to the indicator variable with a detected value of 0 is determined to be a non-redundant safety constraint;

[0020] The non-redundant safety constraints are deleted from the redundant safety constraint identification and optimization model, and the constraint conditions are updated.

[0021] In this invention, the indicator variable for non-redundant safety constraints is 0, while the indicator variable for redundant safety constraints is 1. Therefore, in the redundant safety constraint identification optimization model, it is possible to determine whether non-redundant safety constraints still exist in the model by detecting an indicator variable with an infinite number of values ​​of 0. If they exist, the safety constraint is deleted from the model, and the constraints in the model are updated. Using the indicator variable to distinguish between redundant and non-redundant safety constraints, non-redundant safety constraints can be easily and quickly screened out, facilitating the subsequent construction of the feasible domain model of the transaction plan.

[0022] Furthermore, the method of using vertex search to search for new vertices within the feasible region of the transaction plan to update the boundary range of the feasible region of the transaction plan specifically involves:

[0023] Based on the current vertices of the feasible region, obtain the coefficients of the hyperplane equations corresponding to each boundary of the feasible region;

[0024] Based on the coefficients of each hyperplane equation, construct an optimization model for the outward shift of the plane;

[0025] Search for new feasible domain vertices in the optimization model after the plane has been moved outward;

[0026] The boundary range of the feasible domain for forming a new transaction plan is determined based on the new vertices.

[0027] This invention utilizes a vertex search method. Based on the current vertices of the feasible region, the boundary hyperplane is sequentially translated outward to obtain new vertices of the feasible region and the boundary range of the new transaction plan feasible region. By translating the boundary of the feasible region, the true feasible region boundary is gradually approximated, resulting in a more accurate feasible region boundary.

[0028] Furthermore, the updated boundary range of the feasible region of the transaction plan satisfies a preset loop stopping condition, specifically:

[0029] Calculate the distance deviation between the new feasible domain vertices and the previous boundary plane;

[0030] Compare the distance deviation values ​​of all new feasible domain vertices with the set threshold;

[0031] If the distance deviation values ​​of all new feasible domain vertices are less than the set threshold, then the boundary range of the updated feasible domain of this transaction plan is determined to meet the preset loop stopping condition.

[0032] After updating the boundary of the feasible region of the transaction plan, this invention calculates the distance deviation between the new vertex and the feasible region before the update. This process continues until the distance deviation of all new vertices is less than a preset threshold. At this point, the current feasible region boundary is considered to be the closest to the true feasible region boundary, and the current feasible region is determined to meet the preset loop stopping condition. This allows the loop to stop searching for new vertices within the feasible region of the transaction plan and outputs the boundary result of the current feasible region. This can help dispatching operators better set the boundary conditions related to the transaction plan components before compiling the regional power grid dispatch plan.

[0033] Furthermore, the boundary conditions for obtaining the regional power grid dispatch plan are specifically as follows:

[0034] Obtain the transmission power constraints of inter-provincial tie lines, the output adjustment range constraints of generating units in each province, the positive and negative reserve constraints of each province considering the transmission power of tie lines, the power balance constraints of each province considering the transmission power of tie lines, and the limits of trading components and their matching constraints with physical components.

[0035] This invention provides a method for optimizing inter-provincial power trading plan scheduling based on redundancy safety constraint identification. By obtaining the boundary conditions of the regional power grid scheduling plan and filtering out non-redundant safety constraints from all safety constraints in the regional power grid scheduling problem, a feasible region model of the trading plan is constructed. The feasible region model is solved using the vertex search method to obtain the limit values ​​of the trading plan components, which serve as the transmission power constraint limits for these components. Based on these transmission power constraint limits, the regional power grid trading plan scheduling is optimized. This method avoids conflicts between manually set transmission power constraint limits for trading plan components and the power balance constraints of each provincial power grid by constructing a feasible region model, thus improving the accuracy of the transmission power constraint limits. By using redundancy constraint identification technology to identify effective safety constraints that directly affect the scheduling feasible region, the efficiency of calculating the feasible region of the regional power grid trading plan is ensured, thereby optimizing the regional power grid trading plan scheduling and improving the economic benefits of the regional power grid scheduling plan.

[0036] Accordingly, the present invention provides a cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification, comprising: an acquisition module, a filtering module, a construction module, a loop module, and an optimization module;

[0037] The acquisition module is used to acquire the boundary conditions of the regional power grid dispatch plan;

[0038] The filtering module is used to identify and optimize the optimization model based on the preset redundancy security constraints, and to filter out the non-redundancy security constraints of the regional power grid dispatching problem from the regional power grid dispatching plan.

[0039] The construction module is used to combine the selected non-redundant security constraints and the obtained boundary conditions to form the constraints of the feasible region of the transaction plan, and to construct the feasible region model of the transaction plan.

[0040] The loop module is used to set a set of feasible solutions of the feasible domain model of the transaction plan as the initial vertex of the feasible domain of the transaction plan, and to use the vertex search method to search for new vertices in the feasible domain of the transaction plan in a loop to update the boundary range of the feasible domain of the transaction plan until the updated boundary range satisfies the preset loop stopping condition.

[0041] The optimization module is used to obtain the limit value of the transaction plan component based on the vertex of the current transaction plan feasible domain, as the transaction plan component transmission power constraint limit value, and optimize the regional power grid transaction plan scheduling based on the transaction plan component transmission power constraint limit value.

[0042] Furthermore, the filtering module includes: a first acquisition unit, a construction unit, and a loop unit;

[0043] The first acquisition unit is used to acquire the load fluctuation range;

[0044] The construction unit is used to construct a redundant safety constraint identification and optimization model by introducing an indicator variable for each safety constraint, with load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables;

[0045] The loop unit is used to iteratively solve the redundant safety constraint identification optimization model with the goal of minimizing the sum of the indicator variables, and update the constraint conditions according to the solved model results until all the safety constraints in the constraint conditions are redundant safety constraints.

[0046] Furthermore, the loop module includes: a second acquisition unit, an optimization unit, a search unit, and an update unit;

[0047] The second acquisition unit is used to acquire the coefficients of the hyperplane equations corresponding to each boundary of the feasible region based on the current vertices of the feasible region;

[0048] The optimization unit is used to construct an optimal model for outward displacement of the plane based on the coefficients of each hyperplane equation;

[0049] The search unit is used to search for new feasible domain vertices in the optimization model after the plane has been moved outward;

[0050] The update unit is used to form the boundary range of a new feasible domain for a transaction plan based on the new vertices.

[0051] Furthermore, the loop module includes: a calculation unit, a comparison unit, and a determination unit;

[0052] The computing unit is used to calculate the distance deviation between the new feasible domain vertex and the previous boundary plane;

[0053] The comparison unit is used to compare the distance deviation values ​​of all new feasible domain vertices with a set threshold.

[0054] The determining unit is used to determine that the boundary range of the updated feasible domain of this transaction plan meets the preset loop stopping condition when the distance deviation value of all new feasible domain vertices is less than the set threshold.

[0055] This invention provides a cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification. The device improves the accuracy of the power transmission limit of the trading plan components by organically combining modules, ensures the efficiency of the feasible domain calculation of the regional power grid trading plan, optimizes the scheduling of the regional power grid trading plan, and improves the economic benefits of the regional power grid scheduling plan. Attached Figure Description

[0056] Figure 1This is a flowchart illustrating an embodiment of the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by the present invention.

[0057] Figure 2 A flowchart illustrating one embodiment of the feasible domain of the transaction plan provided by the present invention;

[0058] Figure 3 A flowchart illustrating an embodiment of the redundancy safety constraint identification method provided by the present invention;

[0059] Figure 4 A flowchart illustrating another embodiment of the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by the present invention;

[0060] Figure 5 This is a schematic diagram of one embodiment of the cross-provincial power trading plan scheduling optimization device based on redundancy safety constraint identification provided by the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] In actual regional power grid dispatching operations, operators typically set transmission power limits for each transaction plan component based on the requirements of point-to-grid and grid-to-grid transaction plan execution, as well as the system's operational status. If the coupling constraints between different variables are considered during the stage where operators set the limits for each transaction component, and different combinations of limit values ​​for the transaction plan components are obtained, then post-event corrections due to conflicts between transaction plan component limit constraints and constraints such as provincial power grid power balance and positive / negative reserves can be avoided.

[0063] Based on this, the present invention first selects basic parameters related to the feasible region calculation of inter-provincial power trading plans according to the boundary conditions of the regional power grid dispatching plan. Then, to avoid directly solving the ultra-large-scale regional power grid dispatching optimization problem, and to mitigate the impact of vertex search calculation on the feasible region of the trading plan, the unit output, DC tie line transmission power, trading plan transmission power, and node load in the regional power grid dispatching optimization problem are used as decision variables. Furthermore, an integer variable is introduced for each security constraint to construct a redundant security constraint identification model. This ensures the adaptability of the redundant security constraint identification results to different load levels and avoids considering security constraints that do not directly affect the dispatching constraint space in the regional power grid dispatching optimization problem. Finally, considering the non-redundant security constraints of the regional power grid dispatching problem, as well as the power supply and demand balance of each province, the transmission power limit of inter-provincial tie lines, the maximum and minimum adjustable output of units in each province, and the matching relationship constraints between trading plan components and physical components (including unit group output and inter-provincial tie line transmission power), a feasible region model of the trading plan components is constructed and solved using the vertex search method. This model serves as a reference for the transmission power constraint limit of the trading plan components.

[0064] like Figure 1 The diagram shown is a flowchart of an embodiment of the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by the present invention. The method includes steps 101 to 105, and the specific steps are as follows:

[0065] Step 101: Obtain the boundary conditions of the regional power grid dispatch plan.

[0066] In this embodiment of the invention, the boundary conditions for obtaining the regional power grid dispatch plan are specifically as follows:

[0067] Obtain the transmission power constraints of inter-provincial tie lines, the output adjustment range constraints of generating units in each province, the positive and negative reserve constraints of each province considering the transmission power of tie lines, the power balance constraints of each province considering the transmission power of tie lines, and the limits of trading components and their matching constraints with physical components.

[0068] As an example of an embodiment of the present invention, such as Figure 2 As shown, the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by this invention can be roughly divided into three modules. Module one involves obtaining the boundary conditions of the regional power grid's scheduling plan. Based on the boundary conditions established by the regional power grid's day-ahead and real-time scheduling plans, constraints related to the feasible domain of the cross-provincial trading plan are selected and set. Specifically, the selected input data includes:

[0069] (1) Matching relationship between transaction plan components and physical components:

[0070]

[0071] Where Q represents the number of matching relationships; Y i,t PT represents the transmission power of the i-th transaction component during time period t. j,t PG represents the transmission power of the j-th physical component during time period t; k,t N represents the power generation of the k-th generating unit during time period t; q M q and G q Let represent the set of transaction plan components, the set of tie lines, and the set of generator sets in the q-th matching relationship, respectively.

[0072] For a specific matching relationship, the set of tie lines or the set of generator sets may be an empty set, i.e. The indicated transaction plan component is equal to only a portion of the power of the tie line group, or only equal to the output of only a portion of the generator group.

[0073] (2) The maximum adjustable output PG of unit k in each provincial power grid during time period t. k,t,max and minimum adjustable output PG k,t,min .

[0074] (3) Predicted system load demand PL of each provincial power grid at time period t z,t Positive and negative reserve requirements The subscript z represents the z-th provincial power grid.

[0075] (4) Maximum transmission power limit PT for inter-provincial tie lines j，max and minimum limit PT j，min ;

[0076] (5) Maximum safety constraints PS for cross-sections within and between provincial networks c，max and minimum limit PS c，min .

[0077] Step 102: Based on the preset redundancy security constraint identification and optimization model, select the non-redundancy security constraints of the regional power grid dispatching problem from the regional power grid dispatching plan.

[0078] As an example of an embodiment of the present invention, such as Figure 2 As shown, the constraints to be considered in the regional power grid dispatching problem of this invention specifically include:

[0079] (1) Transmission power constraints of inter-provincial connection lines:

[0080]

[0081] in, and These represent the minimum and maximum transmission power of tie line j in province z, respectively. The maximum transmission power limit PT of the tie line can be obtained from the input of module one. j，max and minimum limit PT j，min In the process, the limit value related to the corresponding province z is selected for matching.

[0082] (2) Output adjustment range constraints for generating units in each province:

[0083]

[0084] In the formula, G z α represents the set of units in province z; k,t This indicates the preset start / stop status of unit k during time period t. The sum of the output of all generating units in province z during time period t should satisfy:

[0085]

[0086] (3) Considering the positive and negative reserve constraints of each province for tie-line transmission power:

[0087]

[0088] In the formula, M z PT represents the set of connecting lines for province z. j,t This represents the transmission power of tie line j during time period t (with received power being positive); PL z,t This represents the load demand of province z during time period t; and This represents the positive and negative reserve requirements of province z during time period t.

[0089] (4) Considering the power balance constraints of each province in terms of tie-line transmission power:

[0090]

[0091] (5) Cross-sectional safety constraints:

[0092]

[0093] Where G, M, and D represent the set of generating units, tie lines, and load nodes included in all provinces within the regional power grid; G k,c G j,c and G l,c These represent the active power sensitivity (i.e., power transfer distribution factor) of the unit, tie line, and load node to safety constraints, respectively.

[0094] (6) Transaction component limits and their matching constraints with physical components:

[0095]

[0096] Equations (1) to (7) above can be written in vector form to form a series of inequalities. The column vectors of generator output, tie-line transmission power, transaction plan transmission power, and node load during time period t are respectively represented as PG. t PT t Y t and PL t Therefore, for the scheduling constraints of equations (1) to (7), they can be rearranged into the following set of linear inequalities:

[0097] A G,t PG t +A t,t PT t +A Y,t Y t +A L,t L t ≤B t (Equation 8)

[0098] Wherein, vector B t A represents the column vector formed by the constraint limits of equations (2) to (7); G,t A T,t A Y,t and A L,t Let m represent the constant coefficient matrices related to unit output, tie-line transmission power, transaction plan transmission power, and node load in the constraints of equations (2) to (7), respectively. Let m represent the total number of constraints included in equation (8). B That is, B t m B A 1×1 dimensional column vector.

[0099] For regional power grid dispatching plans, the security constraints in equation (7) include both internal security constraints within each provincial grid and security constraints between provincial grids, resulting in a large number of constraints. In operational practice, the number of security constraints that may reach the limit or be overloaded is very limited. Most security constraints have a large margin, and the load rate of the security constraints is not high. Furthermore, the security constraints that may be overloaded or reach the limit differ in different dispatching periods. Therefore, the process of redundancy constraint identification needs to consider the impact of load differences in different time periods.

[0100] For the dispatch space of a regional power grid, if, under a specific operating state, a certain constraint will never reach its limit under the premise that all other safety constraints remain within their limits, then this constraint is a redundant constraint and does not need to be considered in the regional power grid dispatch optimization problem under that operating state. Therefore, in the optimization model, non-redundant constraints have a clear characteristic: if a constraint is non-redundant, then there exists at least one operating point at its limit (reaching the limit) such that other constraints will not exceed their limits (they may or may not reach the limit). Otherwise, if no point can be found where all other constraints do not exceed their limits, then the constraint is a redundant constraint, and whether or not the dispatch optimization model considers this constraint will not affect the solution result of the dispatch model.

[0101] Then assume that the number of safety constraints contained in (Equation 8) is m. c (m) c <m B Therefore, after introducing a 0-1 variable for each security constraint, in the regional power grid dispatching and operation optimization space represented by equation (8), the c-th security constraint can be extended as follows:

[0102]

[0103] Among them, w c,t This represents the 0-1 variable introduced for the c-th safety constraint; d c,t Let represent the non-negative continuous variable introduced for the c-th safety constraint; M represents a very large positive constant. If the c-th safety constraint reaches its limit, according to the constraint in equation (9), d c,t It must be 0, and is subject to Constraints, w c,t It must also be 0; conversely, if the c-th constraint does not reach the bound, then d c,t It must be greater than 0, and w c,t =1.

[0104] Therefore, with all safety constraint indicator variables w c,t To minimize the sum, PG t PT t Y t PL t All variables are used as optimization decision variables, and the following optimization model is constructed:

[0105]

[0106] In the model of equation (10), the first constraint (8) represents all the constraints of the regional power grid dispatch problem; the second constraint (9) represents a series of extended constraints introduced for the security constraints of the regional power grid; the third constraint represents the allowable fluctuation range of the nodes, PL. min,t and PL max,t These represent the maximum and minimum value vectors of the nodes, respectively.

[0107] By solving the model of equation (10), if a certain safety constraint c reaches the bound and satisfies all the constraints expressed in equation (8), then the indicator variable w of the corresponding c-th safety constraint... c,t =0, which is recorded as a non-redundant safety constraint. Furthermore, according to the solution of the model of equation (10), it is possible to obtain other constraints that reach the boundary at the same time as the constraint c.

[0108] Then, these identified non-redundant safety constraints are removed from constraint equation (9) of model (10), but these identified non-redundant safety constraints are still retained in constraint equation (8) of model (10). Solving model (10) again, we can further identify that "under the premise of satisfying all constraint conditions, there will still be safety constraints that reach the limit". This process is repeated until the indicator variable w for safety constraints in the solution of optimization model (10) is found. c,t When all values ​​are 1, all non-redundant safety constraints have been identified. In the remaining set of safety constraints, there is no longer a situation where "a certain safety constraint reaches the limit, and all constraints satisfying equation (8) do not reach the limit".

[0109] In this embodiment of the invention, the step of selecting non-redundant security constraints for the regional power grid dispatching problem from the regional power grid dispatching plan based on a preset redundancy security constraint identification and optimization model specifically involves:

[0110] Obtain the load fluctuation range;

[0111] Using load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables, an indicator variable is introduced for each safety constraint to construct a redundant safety constraint identification and optimization model.

[0112] With the goal of minimizing the sum of indicator variables, the redundant safety constraint identification optimization model is solved iteratively, and the constraint conditions are updated according to the solved model results until all safety constraints in the constraint conditions are redundant safety constraints.

[0113] In this embodiment of the invention, during the process of obtaining the redundancy safety constraint identification results, the load fluctuation range is also used as a decision optimization variable to construct a redundancy safety constraint identification optimization model and solve the model. In each solution, non-redundancy safety constraints are selected. The redundancy constraint identification results of this invention can be compatible with the scenario requirements of load fluctuation within a certain range and are more applicable.

[0114] In this embodiment of the invention, solving the redundancy safety constraint identification optimization model and updating the constraint conditions based on the solved model results specifically involves:

[0115] After obtaining the solution results of the redundant safety constraint identification optimization model, it is checked whether there are indicator variables with a value of 0 in the redundant safety constraint identification optimization model.

[0116] If it exists, then the safety constraint corresponding to the indicator variable with a detected value of 0 is determined to be a non-redundant safety constraint;

[0117] The non-redundant safety constraints are deleted from the redundant safety constraint identification and optimization model, and the constraint conditions are updated.

[0118] In this embodiment of the invention, the indicator variable for non-redundant safety constraints is 0, and the indicator variable for redundant safety constraints is 1. Therefore, in the redundant safety constraint identification optimization model, it can be determined whether there are still non-redundant safety constraints in the model by detecting an infinite number of indicator variables with a value of 0. If they exist, the safety constraint is deleted from the model, and the constraint conditions in the model are updated. Using the indicator variable to distinguish between redundant and non-redundant safety constraints, non-redundant safety constraints can be easily and quickly screened out, which facilitates the construction of the subsequent feasible domain model of the transaction plan.

[0119] As an example of an embodiment of the present invention, such as Figure 3 The diagram shown is a flowchart of an embodiment of the redundancy safety constraint identification method provided by the present invention. Module two of the cross-provincial power trading plan scheduling optimization method based on redundancy safety constraint identification provided by the present invention is the identification of redundancy safety constraints, and the specific steps are as follows:

[0120] (1) Determine the boundary conditions for regional power grid dispatch optimization and the range of load fluctuations:

[0121] Determine the node load vector PL t The fluctuation range and regional power grid dispatch optimization problem need to consider the constraints of Equation (8), and also need to consider all security constraints to construct the redundant security constraint identification model of Equation (10).

[0122] (2) Construction and solution of the redundancy safety constraint identification optimization model:

[0123] Solve the optimization model of equation (10) and select the corresponding indicator variable v. c,t =0 corresponds to the security constraints, and these security constraints are marked as non-redundant security constraints. In the constraint condition equation (9) of the model (10), these identified non-redundant security constraints are temporarily ignored, and the optimization model (10) is updated.

[0124] (3) Verification and iteration of redundant security constraints:

[0125] Based on the updated redundant safety constraint identification optimization model, re-solve the updated model and check whether a new safety constraint indicator variable v exists in the results. c,t =0. If it exists, then iteratively update the model for solving equation (10); until after a certain iteration update, the indicator variable v corresponding to the remaining safety constraint results in equation (9) in the result of the optimized model of equation (10) is 0. c,t All values ​​are 1, and the calculation ends. When the calculation ends, the remaining safety constraints in equation (9) are all redundant safety constraints.

[0126] Step 103: Combine the selected non-redundant security constraints and the obtained boundary conditions to form the constraints of the feasible region of the transaction plan, and construct the feasible region model of the transaction plan.

[0127] Step 104: Set a set of feasible solutions of the transaction plan feasible region model as the initial vertices of the transaction plan feasible region, and use the vertex search method to search for new vertices in the transaction plan feasible region in a loop to update the boundary range of the transaction plan feasible region until the updated boundary range satisfies the preset loop stopping condition.

[0128] In this embodiment of the invention, the step of using the vertex search method to search for new vertices within the feasible region of the transaction plan to update the boundary range of the feasible region of the transaction plan specifically involves:

[0129] Based on the current vertices of the feasible region, obtain the coefficients of the hyperplane equations corresponding to each boundary of the feasible region;

[0130] Based on the coefficients of each hyperplane equation, construct an optimization model for the outward shift of the plane;

[0131] Search for new feasible domain vertices in the optimization model after the plane has been moved outward;

[0132] The boundary range of the feasible domain for forming a new transaction plan is determined based on the new vertices.

[0133] In this embodiment of the invention, the vertex search method is used to sequentially translate the boundary hyperplane outward based on the current vertices of the feasible region. This can obtain new vertices of the feasible region and the boundary range of the new transaction plan feasible region. By translating the boundary of the feasible region, the actual feasible region boundary is gradually approached, resulting in a more accurate feasible region boundary.

[0134] In this embodiment of the invention, the boundary range of the updated feasible region of the transaction plan satisfies a preset loop stopping condition, specifically:

[0135] Calculate the distance deviation between the new feasible domain vertices and the previous boundary plane;

[0136] Compare the distance deviation values ​​of all new feasible domain vertices with the set threshold;

[0137] If the distance deviation values ​​of all new feasible domain vertices are less than the set threshold, then the boundary range of the updated feasible domain of this transaction plan is determined to meet the preset loop stopping condition.

[0138] In this embodiment of the invention, after the boundary of the feasible region of the transaction plan is updated, the distance deviation value between the new vertex and the feasible region before the update is calculated. This continues until the distance deviation value of all new vertices is less than a preset threshold. Then, the current feasible region boundary is considered to be the closest to the true feasible region boundary, and the current feasible region is determined to meet the preset loop stopping condition. The loop can then stop searching for new vertices within the feasible region of the transaction plan, and the boundary result of the current feasible region is output. This can help dispatching operators better set the boundary conditions related to the transaction plan components before compiling the regional power grid dispatching plan.

[0139] Step 105: Based on the vertices of the current feasible domain of the transaction plan, obtain the limit values ​​of the transaction plan components as the transmission power constraint limits of the transaction plan components, and optimize the regional power grid transaction plan scheduling based on the transmission power constraint limits of the transaction plan components.

[0140] As an example of an embodiment of the present invention, such as Figure 2 As shown, module three of the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by this invention is the modeling and analysis of the feasible domain of the trading plan.

[0141] First, based on the coupling relationship between the components of the transaction plan and the operation of each provincial power grid, and the results of security constraint redundancy identification, the constraints of the feasible region of the transaction plan are constructed, specifically including:

[0142] (1) Transmission power constraints of inter-provincial connection lines:

[0143]

[0144] (2) Output adjustment range constraints for generating units in each province:

[0145]

[0146]

[0147] (3) Considering the positive and negative reserve constraints of each province for tie-line transmission power:

[0148]

[0149] (4) Considering the power balance constraints of each province in terms of tie-line transmission power:

[0150]

[0151] (5) Transaction component limits and their matching constraints with physical components:

[0152]

[0153] (6) Non-redundant safety constraints:

[0154]

[0155] Among them, Ω c This represents the set of non-redundant safety constraints.

[0156] Based on the feasible domain of the regional power grid dispatch plan formed by equations (1) to (6) and (11), select the transaction plan component objects that need to be set with transmission power limit constraints, and construct the feasible domain of the transaction plan.

[0157] For the regional power grid's dispatch plan during time period t, the decision variables that need to be solved through the dispatch plan model include: the unit's output PG. k,t Transmission power PT of the tie line j,t And the transmission power Y of the transaction plan i,t Therefore, the scheduling feasibility region for time period t can be expressed as:

[0158]

[0159] in, This represents the output vector of each unit during time period t, which includes the unit output variable PG. k,t The number is denoted as m. G PT t This represents the inter-provincial tie-line transmission power vector during time period t, which includes the tie-line power variable PT. j,t The number is denoted as m. T ;Y t This represents the power vector of the transaction plan component transmission during time period t, which contains the transaction plan component transmission power traversal Y. i,t The number is denoted as m.Y .

[0160] For transaction plan components that require setting limit ranges (if m) Y Each transaction component must have a transaction limit set, so that the feasible region of the regional power grid dispatch plan for time period t has a feasible solution can be represented as:

[0161]

[0162] The transmission power limit of the i-th transaction plan component in time period t is set to 0 in sequence. Then, under the premise of satisfying the constraints (1) to (6) and (11), m is first determined. Y +1 initial running point is used as the initial vertex of the feasible region of the transaction plan. Then, the vertex search strategy is adopted to sequentially shift the boundary hyperplane outward under the premise of satisfying the regional power grid dispatching constraints. It is worth noting that during the vertex search (boundary hyperplane shifting) process, the redundant security constraints identified in step two are not considered to avoid directly solving the ultra-large-scale power grid dispatching optimization problem containing all security constraints.

[0163] For the fth v The process of translating the boundary hyperplane outward to search for new vertices can be represented by solving the following optimization model:

[0164]

[0165] in, to Indicates the fth v The coefficients of the equations for a boundary hyperplane, if the vertices of the initial feasible region fall exactly on this plane, then the values ​​of the corresponding transaction plan components are... or Substituting this into the objective function of equation (14), the resulting value is exactly 1. Since the initial feasible region of the transaction plan is a conservative characterization of the boundary of the real feasible region (i.e., the initial feasible region is embedded within the range of the real feasible region), a new vertex can be obtained through model optimization calculation using equation (14), denoted as... Then point Y fv, The objective function value of equation (14) corresponding to t must be greater than or equal to 1.

[0166] After translating each boundary hyperplane of the initial feasible region outward, the boundary range of the initial feasible region in the previous iteration is updated based on all the vertices obtained by the new search. The approximate index of the feasible region after each boundary update is calculated relative to the feasible region before the update. If the approximate index meets the threshold requirement, the boundary result of the feasible region obtained in the most recent round is output; otherwise, the updated boundary of the initial feasible region is recalculated iteratively according to the vertex search method.

[0167] Ultimately, the limit values ​​of the transaction plan components corresponding to the vertices of the feasible domain of the transaction plan are used as a reference for the transmission power constraint limits of the transaction plan components, assisting dispatch operators in better setting the boundary conditions related to the transaction plan components before compiling the regional power grid dispatch plan.

[0168] As an example of an embodiment of the present invention, such as Figure 4 The diagram illustrates the interaction between the three modules of the cross-provincial power trading plan scheduling optimization method based on redundancy security constraint identification provided by this invention. Module 1 acquires the boundary information of the day-ahead and real-time scheduling plan operation boundaries of the regional power grid, which serves as the output of Module 1 and the input of Module 2. Module 2 completes the identification of redundancy security constraints, using all constraint conditions (1) to (7) of the feasible domain of the trading plan for each time period and the identification results of redundancy security constraints as the output of Module 2 and the input of Module 3. Module 3 completes the modeling and solution of the feasible domain of the trading plan, using the vertices of the feasible domain of the trading plan for each time period and the transmission power values ​​of their corresponding trading plan components as the output of Module 3.

[0169] Correspondingly, such as Figure 5 As shown, the present invention also provides a cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification, including: an acquisition module 201, a filtering module 202, a construction module 203, a loop module 204, and an optimization module 205;

[0170] The acquisition module 201 is used to acquire the boundary conditions of the regional power grid dispatch plan;

[0171] The filtering module 202 is used to identify and optimize the optimization model based on the preset redundancy security constraints, and to filter out the non-redundancy security constraints of the regional power grid scheduling problem from the regional power grid scheduling plan.

[0172] The construction module 203 is used to combine the selected non-redundant security constraints and the obtained boundary conditions to form the constraints of the feasible region of the transaction plan, and to construct the feasible region model of the transaction plan.

[0173] The loop module 204 is used to set a set of feasible solutions of the transaction plan feasible domain model as the initial vertex of the transaction plan feasible domain, and to repeatedly use the vertex search method to search for new vertices in the transaction plan feasible domain to update the boundary range of the transaction plan feasible domain until the updated boundary range meets the preset loop stopping condition.

[0174] The optimization module 205 is used to obtain the limit value of the transaction plan component based on the vertex of the current transaction plan feasible domain, as the transaction plan component transmission power constraint limit value, and optimize the regional power grid transaction plan scheduling based on the transaction plan component transmission power constraint limit value.

[0175] Furthermore, the filtering module 202 includes: a first acquisition unit, a construction unit, and a loop unit;

[0176] The first acquisition unit is used to acquire the load fluctuation range;

[0177] The construction unit is used to construct a redundant safety constraint identification and optimization model by introducing an indicator variable for each safety constraint, with load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables;

[0178] The loop unit is used to iteratively solve the redundant safety constraint identification optimization model with the goal of minimizing the sum of the indicator variables, and update the constraint conditions according to the solved model results until all the safety constraints in the constraint conditions are redundant safety constraints.

[0179] Furthermore, the loop module 204 includes: a second acquisition unit, an optimization unit, a search unit, and an update unit;

[0180] The second acquisition unit is used to acquire the coefficients of the hyperplane equations corresponding to each boundary of the feasible region based on the current vertices of the feasible region.

[0181] The optimization unit is used to construct an optimal model for the outward shift of the hyperplane based on the coefficients of each hyperplane equation;

[0182] The search unit is used to search for new feasible domain vertices in the optimization model after the plane has been moved outward;

[0183] The update unit is used to compose the boundary range of the new transaction plan feasible domain based on the new vertices.

[0184] Furthermore, the loop module 204 includes: a calculation unit, a comparison unit, and a determination unit;

[0185] The computational unit is used to calculate the distance deviation between the new feasible domain vertices and the previous boundary plane;

[0186] The comparison unit is used to compare the distance deviation values ​​of all new feasible domain vertices with a set threshold;

[0187] The determination unit is used to determine that the boundary range of the updated feasible domain of this transaction plan meets the preset loop stopping condition when the distance deviation values ​​of all new feasible domain vertices are less than the set threshold.

[0188] In summary, the present invention provides a method and apparatus for optimizing inter-provincial power trading plan scheduling based on redundancy security constraint identification. This method obtains the boundary conditions of the regional power grid scheduling plan, filters out non-redundant security constraints from all security constraints in the regional power grid scheduling problem, constructs a feasible region model of the trading plan, solves the feasible region model using the vertex search method, obtains the limit values ​​of the trading plan components as transmission power constraint limits for the trading plan components, and optimizes the regional power grid trading plan scheduling based on these transmission power constraint limits. This method avoids conflicts between manually set transmission power constraint limits for trading plan components and the power balance constraints of each provincial power grid by constructing a feasible region model, thus improving the accuracy of the transmission power constraint limits for trading plan components. Furthermore, by using redundancy constraint identification technology to identify effective security constraints that directly affect the scheduling feasible region, it ensures the efficiency of calculating the feasible region of the regional power grid trading plan, optimizes the regional power grid trading plan scheduling, and improves the economic benefits of the regional power grid scheduling plan.

[0189] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing inter-provincial power trading scheduling based on redundancy security constraint identification, characterized in that, include: Obtain the boundary conditions for the regional power grid dispatch plan; Based on the preset redundancy security constraint identification and optimization model, non-redundancy security constraints of the regional power grid dispatching problem are screened from the regional power grid dispatching plan; The selected non-redundant security constraints and the obtained boundary conditions are used to form the constraints of the feasible region of the transaction plan, and the feasible region model of the transaction plan is constructed. A set of feasible solutions of the feasible domain model of the transaction plan is set as the initial vertex of the feasible domain of the transaction plan, and the vertex search method is used cyclically to search for new vertices in the feasible domain of the transaction plan to update the boundary range of the feasible domain of the transaction plan until the updated boundary range satisfies the preset loop stopping condition. Based on the vertices of the current feasible domain of the transaction plan, the limit values ​​of the transaction plan components are obtained as the transmission power constraint limits of the transaction plan components, and the regional power grid transaction plan scheduling is optimized based on the transmission power constraint limits of the transaction plan components. Specifically, the step of selecting non-redundant security constraints for the regional power grid dispatching problem from the regional power grid dispatching plan based on a preset redundancy security constraint identification and optimization model includes: Obtain the load fluctuation range; Using load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables, an indicator variable is introduced for each safety constraint to construct a redundant safety constraint identification and optimization model. With the goal of minimizing the sum of indicator variables, the redundant safety constraint identification optimization model is solved iteratively, and the constraint conditions are updated according to the solved model results until all safety constraints in the constraint conditions are redundant safety constraints.

2. The method for optimizing inter-provincial power trading scheduling based on redundancy security constraint identification according to claim 1, characterized in that, The process of solving the redundant safety constraint identification optimization model and updating the constraint conditions based on the solved model results is as follows: After obtaining the solution results of the redundant safety constraint identification optimization model, it is checked whether there are indicator variables with a value of 0 in the redundant safety constraint identification optimization model. If it exists, then the safety constraint corresponding to the indicator variable with a detected value of 0 is determined to be a non-redundant safety constraint; The non-redundant safety constraints are deleted from the redundant safety constraint identification and optimization model, and the constraint conditions are updated.

3. The method for optimizing inter-provincial power trading scheduling based on redundancy security constraint identification according to claim 1, characterized in that, The method of using vertex search to search for new vertices within the feasible region of the transaction plan in order to update the boundary range of the feasible region of the transaction plan is as follows: Based on the current vertices of the feasible region, obtain the coefficients of the hyperplane equations corresponding to each boundary of the feasible region; Based on the coefficients of each hyperplane equation, construct an optimization model for the outward shift of the plane; Search for new feasible domain vertices in the optimization model after the plane has been moved outward; The boundary range of the feasible domain for forming a new transaction plan is determined based on the new vertices.

4. The method for optimizing inter-provincial power trading plan scheduling based on redundancy security constraint identification according to claim 3, characterized in that, The updated boundary range of the feasible region of the transaction plan satisfies a preset loop stopping condition, specifically: Calculate the distance deviation between the new feasible domain vertices and the previous boundary plane; Compare the distance deviation values ​​of all new feasible domain vertices with the set threshold; If the distance deviation values ​​of all new feasible domain vertices are less than the set threshold, then the boundary range of the updated feasible domain of this transaction plan is determined to meet the preset loop stopping condition.

5. The method for optimizing inter-provincial power trading scheduling based on redundancy security constraint identification according to claim 1, characterized in that, The boundary conditions for obtaining the regional power grid dispatch plan are specifically as follows: Obtain the transmission power constraints of inter-provincial tie lines, the output adjustment range constraints of generating units in each province, the positive and negative reserve constraints of each province considering the transmission power of tie lines, the power balance constraints of each province considering the transmission power of tie lines, and the limits of trading components and their matching constraints with physical components.

6. A cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification, characterized in that, include: The modules include: acquisition module, filtering module, building module, looping module, and optimization module. The acquisition module is used to acquire the boundary conditions of the regional power grid dispatch plan; The filtering module is used to identify non-redundant security constraints of the regional power grid dispatching problem from the regional power grid dispatching plan based on a preset redundancy security constraint identification optimization model. The construction module is used to combine the selected non-redundant security constraints and the obtained boundary conditions to form the constraints of the feasible region of the transaction plan, and to construct the feasible region model of the transaction plan. The loop module is used to set a set of feasible solutions of the feasible domain model of the transaction plan as the initial vertex of the feasible domain of the transaction plan, and to use the vertex search method to search for new vertices in the feasible domain of the transaction plan in a loop to update the boundary range of the feasible domain of the transaction plan until the updated boundary range satisfies the preset loop stopping condition. The optimization module is used to obtain the limit value of the transaction plan component based on the vertex of the current transaction plan feasible domain, as the transmission power constraint limit of the transaction plan component, and optimize the regional power grid transaction plan scheduling based on the transmission power constraint limit of the transaction plan component. The filtering module includes: a first acquisition unit, a construction unit, and a loop unit; The first acquisition unit is used to acquire the load fluctuation range; The construction unit is used to construct a redundant safety constraint identification and optimization model by introducing an indicator variable for each safety constraint, with load fluctuation range, unit output, tie line transmission power and component transmission power as optimization variables; The loop unit is used to iteratively solve the redundant safety constraint identification optimization model with the goal of minimizing the sum of the indicator variables, and update the constraint conditions according to the solved model results until all the safety constraints in the constraint conditions are redundant safety constraints.

7. The cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification according to claim 6, characterized in that, The loop module includes: a second acquisition unit, an optimization unit, a search unit, and an update unit; The second acquisition unit is used to acquire the coefficients of the hyperplane equations corresponding to each boundary of the feasible region based on the current vertices of the feasible region; The optimization unit is used to construct an optimal model for outward displacement of the plane based on the coefficients of each hyperplane equation; The search unit is used to search for new feasible domain vertices in the optimization model after the plane has been moved outward; The update unit is used to form the boundary range of a new feasible domain for a transaction plan based on the new vertices.

8. The cross-provincial power trading plan scheduling optimization device based on redundancy security constraint identification according to claim 7, characterized in that, The loop module includes: a calculation unit, a comparison unit, and a determination unit; The computing unit is used to calculate the distance deviation between the new feasible domain vertex and the previous boundary plane; The comparison unit is used to compare the distance deviation values ​​of all new feasible domain vertices with a set threshold. The determining unit is used to determine that the boundary range of the updated feasible domain of this transaction plan meets the preset loop stopping condition when the distance deviation value of all new feasible domain vertices is less than the set threshold.

Images