Power distribution system intraday economic dispatching method and system considering power generation uncertainty

Abstract

The invention discloses a power distribution system intraday economic dispatching method and system considering power generation uncertainty, and the method comprises the steps: collecting real-time data of a power distribution system, and building a two-stage adaptive robust centralized intraday economic dispatching model; dividing the global model into a plurality of sub-regions according to the physical structure of the power distribution system, and establishing a two-stage adaptive robust distributed intraday economic dispatching model; and solving the two-stage adaptive robust distributed intraday economic dispatching model by using a fully distributed consistency ADMM algorithm. Through regional processing, the calculation burden of each sub-region is reduced, and the calculation efficiency of the whole system is improved. Due to the distributed calculation characteristic of the ADMM algorithm, scheduling optimization can be carried out on all the sub-regions in parallel, and the solving efficiency is greatly improved. The rapid convergence characteristic enables the system to respond to the scheduling requirement in a short time, and the stability and safety of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system optimization dispatching, and in particular to a method and system for intraday economic dispatching of a distribution system taking into account power generation uncertainty. Background Art

[0002] Due to the uncertainty of renewable energy generation in CPDS, there is often a deviation between its actual output and the predicted value, which causes the actual system operation point to deviate from the reference operation point determined by the intraday economic dispatch model based on the predicted value of renewable energy generation. In extreme scenarios with a large degree of deviation, high adjustment costs will be generated in the real-time power balance stage, seriously affecting the economic operation of CPDS.

[0003] To address this problem, this chapter proposes a two-stage adaptive robust distributed intraday economic dispatch model for CPDS to obtain the operating point that optimizes the total operating cost of CPDS in the two stages while considering the uncertainty of renewable energy output. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing intraday economic dispatch technology of distribution system has the problems of insufficient coping with power generation uncertainty, low calculation efficiency and inability to guarantee global optimality, and there are technical bottlenecks in achieving consistency and convergence.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for intraday economic dispatch of a distribution system considering power generation uncertainty, comprising:

[0007] Collect real-time data of the distribution system and establish a two-stage adaptive robust centralized intraday economic dispatch model;

[0008] According to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established.

[0009] The fully distributed consistent ADMM algorithm is used to solve the two-stage adaptive robust distributed intraday economic dispatch model.

[0010] As a preferred solution of the intraday economic dispatch method of the distribution system considering power generation uncertainty described in the present invention, wherein: the real-time data collected from the distribution system includes load data, wind power and photovoltaic power generation data, conventional power generation data, distribution system status data and uncertainty data;

[0011] Data is collected through smart meters, micro-phasor measurement units, distributed sensors and measurement equipment in wind and photovoltaic power plants.

[0012] As a preferred solution of the intraday economic dispatch method of the power distribution system considering power generation uncertainty described in the present invention, the establishment of a two-stage adaptive robust centralized intraday economic dispatch model includes a first-stage decision and a second-stage decision;

[0013] The first stage decision includes the first stage decision variable definition, the first stage objective function definition and the first stage constraint conditions;

[0014] The second stage decision includes the definition of the second stage decision variables, the definition of the second stage objective function and the second stage constraints;

[0015] The first stage decision variable definition y f The formula is:

[0016]

[0017] Where ν represents the set of all buses in the physical system, ξ represents the set of all distribution lines, G represents all power generation units, ν represents the square of the bus voltage amplitude of the decision variable, and p ij represents the active power on the directional distribution line i,j), q ij represents the reactive power on the directional distribution line (i, j), l ij represents the square of the current amplitude on the directional distribution line (i, j), p g Represents the active output of the generating unit, q g Indicates the reactive power output of the generating unit;

[0018] The definition of the objective function in the first stage includes: the decision-making goal of the first stage is based on the predicted value of wind power generation Minimize the power generation cost and wind curtailment cost of all power generation units. The formula is:

[0019]

[0020] Among them, c2, c1, c0 are the coefficients of the secondary power generation cost function of the generator, represents the penalty cost coefficient for wind curtailment; represents the set of wind turbines, ∑ j∈G represents the sum of all conventional generators j, c 2,j represents the quadratic cost coefficient of conventional generator j, c 1,j represents the primary cost coefficient of conventional generator j, c 0,j represents the fixed cost of conventional generator j, p g,j represents the power generated by conventional generator j, represents the power generation cost of conventional generator j, ∑ j∈Wrepresents the sum of all wind farms j, represents the uncertainty cost coefficient of wind power, represents the predicted output of wind farm j, p g,j represents the actual power generation of wind farm j, Represents the cost caused by the uncertainty of wind power output.

[0021] As a preferred solution of the intraday economic dispatch method of the distribution system considering the uncertainty of power generation described in the present invention, the first stage constraint conditions include: is the active and reactive balance equation of each bus, and the formula is expressed as:

[0022]

[0023] for Using the second-order cone relaxation SOCR technique, the original quadratic equality constraint ν is relaxed using the second-order cone i l ij =(p ij ) 2 +(q ij ) 2 :

[0024] ||2p ij ;2q ij ; l ij -v i ||2≤l ij +v i

[0025] The upper and lower amplitude boundaries of each bus voltage are defined:

[0026]

[0027] The output convex constraint set of each power generation unit first limits all units The power generation capacity must meet the upper and lower bounds of the power generation capacity;

[0028] In addition, the active output of wind turbines is limited to no higher than the predicted value. And the power factor φ must satisfy the interval constraint φ∈[φ min ,φ max ];

[0029] Represents the power flow constraint of the distribution line, and the formula is expressed as:

[0030]

[0031] l ij ≥0

[0032] Among them, p g,jrepresents the active power of the generator at node j, p d,j represents the load active power of node j, represents the line active power flow from other nodes i to node j minus line losses, ∑ (j,k)∈ξ p jk represents the line active power flow from node j to other nodes k, g j ν j represents the power consumption caused by the ground admittance of node j, q g,j represents the reactive power of the generator at node j, q d,j represents the load reactive power of node j, ∑ (i,j)∈ξ (q ij -x ij l ij ) represents the line reactive power flow from other nodes i to node j minus line losses, Σ (j,k)∈ξ q jk represents the line reactive power flow from node j to other nodes k, b j ν j represents the reactive power consumption caused by the ground admittance of node j, ν j represents the square of the voltage at node j, ν i represents the square of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, l ij represents the square of the current in the line between node i and node j, p ij represents the active power of the line from node i to node j, q ij represents the line reactive power from node i to node j, ||2p ij ;2q ij ; l ij -ν i ||2 represents the quadratic cone norm, represents the minimum value of the square of the voltage at node j, represents the maximum value of the square of the voltage at node j, represents the lower limit of active power of conventional generator j, represents the upper limit of active power of conventional generator j, represents the lower limit of reactive power of conventional generator j, represents the upper limit of reactive power of conventional generator j, p g,j represents the active power of wind farm j, q g,j represents the reactive power of wind farm j, represents the predicted output of wind farm j, tanφ min Represents the reactive power coefficient corresponding to the lower limit of the wind farm power factor, tanφ maxThe reactive power coefficient corresponding to the upper limit of the wind farm power factor, q ij represents the reactive power flow of line (i, j), l ij represents the square of the current in line (i, j), p ij represents the active power flow of line (i, j), r ij represents the resistance of line (i, j), x ij represents the reactance of line (i,j), represents the maximum value of active power flow of line (i, j), represents the maximum value of reactive power flow of line (i, j);

[0033] The first stage decision is a deterministic DED optimization based on the predicted value of wind power generation. The form of the first stage constraints conforms to the global DED model in Chapter 3.

[0034] As a preferred solution of the intraday economic dispatch method of the distribution system considering the uncertainty of power generation described in the present invention, the second stage decision includes: the second stage decision is used to quantify the re-dispatch cost of CPDS to readjust the power balance based on the first stage decision solution after the actual value of the renewable energy output is determined; therefore, the second stage decision process is a correction to the first stage decision solution;

[0035] The adjustable units in the second stage of decision-making include: distributed generators, wind turbines, and flexible loads;

[0036] The first stage decision variable y s Definition, the formula is expressed as:

[0037]

[0038] in, represents the voltage change of node j, V represents the set of all nodes, represents the change in active power of line (i, j), The reactive power change of line (i,j), represents the current change of line (i, j), ξ represents the set of all lines, represents the active power change of conventional generator j, represents the reactive power change of conventional generator j, represents the active power change of conventional generator j, G represents the set of conventional generators, represents the critical wind power output of wind farm j, W represents the set of wind farms, represents the change in active power of load j, represents the reactive power change of load j, represents the absolute change in active power of load j, and L represents the set of load nodes;

[0039] The definition of the objective function of the second stage includes: The objective function of the second stage decision is to minimize the corrective power regulation cost Q of CPDS under the worst uncertainty scenario u∈U s The cost includes four parts: generator power regulation cost, wind abandonment penalty, flexible load regulation cost and uncontrollable load shedding penalty. This part models it as a linear function of the corresponding correction variable. CL represents the load bus set where the controllable flexible load is located; the coefficient are the price coefficients of the above four adjustment costs, and the formula is:

[0040]

[0041] Among them, max u min ys Q s represents the objective function, Q s Represents the state variable y s The total cost function under represents the absolute change in active power of conventional generator j, represents the power adjustment cost coefficient of conventional generator j, G\W represents the set of conventional generators in set G after removing the wind farm set W, represents the critical wind power output of wind farm j, represents the critical output cost coefficient of wind farm j, W represents the set of wind farms, represents the absolute change in active power of the uncontrollable load k, represents the power adjustment cost coefficient of controllable load j, CL represents the set of controllable loads, represents the absolute change in active power of the uncontrollable load k, represents the load reduction cost coefficient of the non-controllable load j, L\CL represents the non-controllable load set after removing the controllable load set CL from the set L;

[0042] The constraint formula of the second stage is expressed as:

[0043]

[0044] in, represents the change in active power of the generator at node j, represents the load active power change of node j, represents the change in active power flow from other nodes i to node j minus the change in line loss, represents the change in active power flow from node j to other nodes k, represents the power change caused by the voltage change caused by the ground admittance of node j, represents the square change of the voltage at node j, represents the square change of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, represents the square of the current change in the line between node i and node j, represents the change in line active power from node i to node j, represents the change in line reactive power from node i to node j, represents the quadratic cone norm, It means that the total change of active power of wind farm j is equal to the power generation of the wind farm minus the critical wind power output. Indicates that the critical wind power output must be a non-negative value, Indicates that the reactive power change of the wind farm must meet the power factor lower limit constraint, Indicates that the reactive power change of the wind farm must meet the upper limit constraint of the power factor;

[0045]

[0046] Among them, β min Indicates the minimum proportional factor of load power change, β max Indicates the maximum proportional factor of load power change, p d,j represents the original active load of node j, q d,j represents the original reactive load of node j. In the above constraints, {p g ,q g ,p ij ,q ij ,l ij ,ν} represents the “base state” operating point obtained by the first stage decision-making, which is regarded as a known constant with a fixed value in the second stage decision-making model.

[0047] As a preferred solution of the intraday economic dispatch method of the distribution system considering the uncertainty of power generation described in the present invention, wherein: according to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established, including, by means of a fully distributed optimization method, decomposing the above-mentioned centralized ARED optimization problem into decision nodes corresponding to each physical sub-area, forming an adaptive robust distributed intraday economic dispatch model, reducing the problem scale and the number of poles in a decoupled and coordinated manner, and realizing the distributed solution of the ARDED model with the help of a fully distributed consistency algorithm;

[0048] The local two-stage decision model of each sub-region α has a similar form to the centralized ARED model, and the formula is expressed as:

[0049]

[0050] There will be sub-region interconnection lines between each sub-region. In the ARDED sub-region model, the power injection from the adjacent sub-region and the power transmission to the adjacent sub-region are equivalently treated as the purchase and sale of electricity between adjacent regions, and this cost is included in the objective function of the two-stage decision of the sub-region ARDED. For the first stage decision, the formula is expressed as:

[0051]

[0052] For the second stage decision:

[0053]

[0054] in, represents the unit power exchange cost coefficient of adjacent sub-regions in the first stage of ARDED, represents the redispatching cost coefficient of power exchange between sub-regions in the second stage;

[0055] Since a new non-negative variable is introduced in the second stage objective function Add auxiliary constraints to the constraint set of the second-stage decision, and the formula is expressed as:

[0056]

[0057] Among them, G α represents the set of regular generators in subregion α, c 2,j ,c 1,j ,c 0,j represents the cost coefficient of conventional generator j, p g,j represents the power generated by conventional generator j, W α represents the set of wind farms in sub-region α, represents the uncertainty cost coefficient of wind power output, represents the predicted power generation of wind farm j, p g,j represents the actual power generation of wind farm j, represents the power exchange set between sub-region α and its adjacent sub-regions, represents the unit power exchange cost coefficient of the adjacent sub-region in the first stage, p ij represents the power exchange amount from node i to node j, {G α \W α} represents the set of conventional generators in sub-region α after removing the wind farm, represents the power adjustment cost coefficient of conventional generator j, represents the absolute active power change of conventional generator j, represents the critical wind power output cost coefficient of wind farm j, Critical wind power output of wind farm j, CL α represents the set of controllable loads in sub-region α, ρ d,j represents the power adjustment cost coefficient of controllable load j, represents the absolute active power change of the controllable load k, represents the redispatching cost coefficient of power exchange between sub-regions in the second stage, represents the absolute change in power exchange from node i to node j;

[0058]

[0059] First, each sub-area measurement node S sends local measurement information Transmitted to each sub-region decision node via information branch Initialize the decision model. After the initialization is completed, enter the DCADMM fully distributed iteration process. In the iteration process, the first step is to make a decision Φ x For local variables The decision-making process corresponds to the optimization problem:

[0060]

[0061] After the first round of information exchange among decision nodes, each decision node exchanges the decision variable solutions from adjacent decision nodes. And through the calculation of the fully distributed interactive information branch model, the next stage decision Φ is obtained z The input vector The formula is:

[0062]

[0063] In fixed After the value of Φ z and Φ λ They correspond to the following decision-making processes, and the formulas are expressed as follows:

[0064]

[0065] After Q iterations until convergence, the optimal control decision obtained by convergence is The control target instruction υ is sent to the execution node A corresponding to each controllable unit.

[0066] As a preferred solution of the intraday economic dispatch method of the distribution system considering the uncertainty of power generation described in the present invention, wherein: using the fully distributed consistent ADMM algorithm to realize the two-stage adaptive robust distributed intraday economic dispatch model solution includes two layers consisting of an outer DCADMM iteration and an inner C&CG iteration. The role of the outer DCADMM iteration is to coordinate the coupling variables x of multiple adjacent sub-regions. a , so that it finally satisfies the consistency constraint. The first step of DCADMM iteration is used as the entry of the inner C&CG algorithm iteration and is passed to the inner iteration. as well as As a fixed parameter; through the iterative calculation of the inner layer C&CG algorithm, we get Returns the outer DCADMM as the mirror variable for subsequent DCADMM updates and the multiplier variable The required fixed parameters and the remaining decision and communication interaction steps are located in the outer DCADMM iteration, which are executed in sequence as described above without calling the inner iteration. When the outer DCADMM meets the convergence criterion, ARDED is output.

[0067] The intraday economic dispatch system of the distribution system considering the uncertainty of power generation includes: the acquisition module, which collects the real-time data of the distribution system and establishes a two-stage adaptive robust centralized intraday economic dispatch model;

[0068] The partitioning module divides the global model into several sub-areas according to the physical structure of the distribution system, and establishes a two-stage adaptive robust distributed intraday economic dispatch model;

[0069] The solution module uses the fully distributed consistent ADMM algorithm to achieve a two-stage adaptive robust distributed intraday economic dispatch model solution.

[0070] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0071] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0072] Beneficial effects of the present invention: The intraday economic dispatch method for the distribution system considering the uncertainty of power generation provided by the present invention reduces the computational burden of each sub-region through regional processing, thereby improving the computational efficiency of the overall system. The distributed computing characteristics of the ADMM algorithm enable each sub-region to perform scheduling optimization in parallel, greatly improving the solution efficiency. The fast convergence characteristics enable the system to respond to scheduling requirements in a short time, thereby improving the stability and security of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0074] Figure 1 An overall flow chart of a method for intraday economic dispatch of a power distribution system taking into account power generation uncertainty provided by a first embodiment of the present invention;

[0075] Figure 2 A flowchart of DCADMM and C&CG iterative calculations in a distribution system intraday economic dispatch method considering power generation uncertainty provided by the first embodiment of the present invention;

[0076] Figure 3 An improved IEEE118 node CPDS test system diagram of a distribution system intraday economic dispatch method considering power generation uncertainty provided by a first embodiment of the present invention;

[0077] Figure 4 The node voltage dispatching results and branch SOCP relaxation error diagram of the intraday economic dispatching method for the distribution system considering power generation uncertainty provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0079] Example 1

[0080] Reference Figure 1 , which is an embodiment of the present invention, provides a method for intraday economic dispatch of a distribution system considering power generation uncertainty, comprising:

[0081] S1: Collect real-time data of the distribution system and establish a two-stage adaptive robust centralized intraday economic dispatch model. The real-time data collected from the distribution system include load data, wind power and photovoltaic power generation data, conventional power generation data, distribution system status data and uncertainty data;

[0082] Data is collected through smart meters, micro-phasor measurement units, distributed sensors and measurement equipment in wind and photovoltaic power plants.

[0083] The establishment of a two-stage adaptive robust centralized intraday economic dispatch model includes the first stage decision and the second stage decision.

[0084] The first stage decision includes the first stage decision variable definition, the first stage objective function definition and the first stage constraint conditions.

[0085] The second stage decision includes the second stage decision variable definition, the second stage objective function definition and the second stage constraint conditions.

[0086] The first stage decision variable definition y f The formula is:

[0087]

[0088] Where ν represents the set of all buses in the physical system, ξ represents the set of all distribution lines, G represents all power generation units, ν represents the square of the bus voltage amplitude of the decision variable, and p ij represents the active power on the directional distribution line i,j), q ij represents the reactive power on the directional distribution line (i, j), l ij represents the square of the current amplitude on the directional distribution line (i, j), p g Represents the active output of the generating unit, q g Indicates the reactive power output of the generating unit.

[0089] The definition of the objective function in the first stage includes: the decision-making goal of the first stage is based on the predicted value of wind power generation Minimize the power generation cost and wind curtailment cost of all power generation units. The formula is:

[0090]

[0091] Among them, c2, c1, c0 are the coefficients of the secondary power generation cost function of the generator, ρ cr represents the penalty cost coefficient for wind curtailment; represents the set of wind turbines, ∑ j∈G represents the sum of all conventional generators j, c 2,j represents the quadratic cost coefficient of conventional generator j, c1,j represents the primary cost coefficient of conventional generator j, c 0,j represents the fixed cost of conventional generator j, p g,j represents the power generated by conventional generator j, represents the power generation cost of conventional generator j, ∑ j∈W represents the sum of all wind farms j, represents the uncertainty cost coefficient of wind power, represents the predicted output of wind farm j, p g,j represents the actual power generation of wind farm j, Represents the cost caused by the uncertainty of wind power output.

[0092] The first stage constraints include: is the active and reactive balance equation of each bus, and the formula is expressed as:

[0093]

[0094] for Using the second-order cone relaxation SOCR technique, the original quadratic equality constraint ν is relaxed using the second-order cone i l ij =(p ij ) 2 +(q ij ) 2 :

[0095] ||2p ij ;2q ij ; l ij -v i ||2≤l ij +v i

[0096] The upper and lower amplitude boundaries of each bus voltage are defined:

[0097]

[0098] The output convex constraint set of each power generation unit first limits all units The power generation capacity must meet the upper and lower bounds of the power generation capacity.

[0099] In addition, the active output of wind turbines is limited to no higher than the predicted value. And the power factor φ must satisfy the interval constraint φ∈[φ min ,φ max ].

[0100] Represents the power flow constraint of the distribution line, and the formula is expressed as:

[0101]

[0102] l ij ≥0

[0103] Among them, p g,j represents the active power of the generator at node j, p d,j represents the load active power of node j, ∑ (i,j)∈ξ (p ij -r ij l ij ) represents the active power flow from other nodes i to node j minus the line loss, ∑ (j,k)∈ξ p jk represents the line active power flow from node j to other nodes k, g j ν j represents the power consumption caused by the ground admittance of node j, q g,j represents the reactive power of the generator at node j, q d,j represents the load reactive power of node j, ∑( i,j)∈ξ (q ij -x ij l ij ) represents the line reactive power flow from other nodes i to node j minus line losses, ∑ (j,k)∈ξ q jk represents the line reactive power flow from node j to other nodes k, b j ν j represents the reactive power consumption caused by the ground admittance of node j, ν j represents the square of the voltage at node j, ν i represents the square of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, l ij represents the square of the current in the line between node i and node j, p ij represents the active power of the line from node i to node j, q ij represents the line reactive power from node i to node j, ||2p ij ;2q ij ; l ij -ν i ||2 represents the quadratic cone norm, represents the minimum value of the square of the voltage at node j, represents the maximum value of the square of the voltage at node j, represents the lower limit of active power of conventional generator j, represents the upper limit of active power of conventional generator j, represents the lower limit of reactive power of conventional generator j, represents the upper limit of reactive power of conventional generator j, pg,j represents the active power of wind farm j, q g,j represents the reactive power of wind farm j, represents the predicted output of wind farm j, tanφ min Represents the reactive power coefficient corresponding to the lower limit of the wind farm power factor, tanφ max The reactive power coefficient corresponding to the upper limit of the wind farm power factor, q ij represents the reactive power flow of line (i, j), l ij represents the square of the current in line (i, j), p ij represents the active power flow of line (i, j), r ij represents the resistance of line (i, j), x ij represents the reactance of line (i,j), represents the maximum value of active power flow of line (i, j), represents the maximum value of reactive power flow of line i,j).

[0104] The first stage decision is a deterministic DED optimization based on the predicted value of wind power generation, and the form of the constraints in the first stage conforms to the global DED model in Chapter 3. The second stage decision includes: the second stage decision is used to quantify the redispatch cost of CPDS to readjust the power balance based on the first stage decision solution after the actual value of renewable energy output is determined; therefore, the second stage decision process is a correction to the first stage decision solution.

[0105] The adjustable units for the second stage decision-making include: distributed generators, wind turbines, and flexible loads.

[0106] The first stage decision variable y s Definition, the formula is expressed as:

[0107]

[0108] in, represents the voltage change of node j, V represents the set of all nodes, represents the change in active power of line i,j), The reactive power change of line (i,j), represents the current change of line (i, j), ξ represents the set of all lines, represents the active power change of conventional generator j, represents the reactive power change of conventional generator j, represents the active power change of conventional generator j, G represents the set of conventional generators, represents the critical wind power output of wind farm j, W represents the set of wind farms, represents the change in active power of load j, represents the reactive power change of load j, represents the absolute change in active power of load j, and L represents the set of load nodes.

[0109] The definition of the objective function of the second stage includes: The objective function of the second stage decision is to minimize the corrective power regulation cost Q of CPDS under the worst uncertainty scenario u∈U s The cost includes four parts: generator power regulation cost, wind abandonment penalty, flexible load regulation cost and uncontrollable load shedding penalty. This part models it as a linear function of the corresponding correction variable. CL represents the load bus set where the controllable flexible load is located; the coefficient are the price coefficients of the above four adjustment costs, and the formula is:

[0110]

[0111] in, represents the objective function, Q s Represents the state variable y s The total cost function under represents the absolute change in active power of conventional generator j, represents the power adjustment cost coefficient of conventional generator j, G\W represents the set of conventional generators in set G after removing the wind farm set W, represents the critical wind power output of wind farm j, represents the critical output cost coefficient of wind farm j, W represents the set of wind farms, represents the absolute change in active power of the uncontrollable load k, represents the power adjustment cost coefficient of controllable load j, CL represents the set of controllable loads, represents the absolute change in active power of the uncontrollable load k, represents the load reduction cost coefficient of the uncontrollable load j, and L\CL represents the uncontrollable load set after removing the controllable load set CL from the set L.

[0112] The constraint formula of the second stage is expressed as:

[0113]

[0114]

[0115] in, represents the change in active power of the generator at node j, represents the load active power change of node j, represents the change in active power flow from other nodes i to node j minus the change in line loss, represents the change in active power flow from node j to other nodes k, represents the power change caused by the voltage change caused by the ground admittance of node j, represents the square change of the voltage at node j, represents the square change of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, represents the square of the current change in the line between node i and node j, represents the change in line active power from node i to node j, represents the change in line reactive power from node i to node j, represents the quadratic cone norm, It means that the total change of active power of wind farm j is equal to the power generation of the wind farm minus the critical wind power output. Indicates that the critical wind power output must be a non-negative value, Indicates that the reactive power change of the wind farm must meet the power factor lower limit constraint, It means that the reactive power change of the wind farm must meet the upper limit constraint of the power factor.

[0116]

[0117] Among them, β min Indicates the minimum proportional factor of load power change, β max Indicates the maximum proportional factor of load power change, p d,j represents the original active load of node j, q d,j represents the original reactive load of node j. In the above constraints, {p g ,q g ,p ij ,q ij ,l ij ,ν} represents the “base state” operating point obtained by the first stage decision, which is regarded as a known constant with a fixed value in the second stage decision model. S2: According to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established.

[0118] According to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established. This includes decomposing the above-mentioned centralized ARED optimization problem into decision nodes corresponding to each physical sub-area through a fully distributed optimization method to form an adaptive robust distributed intraday economic dispatch model. The problem scale is reduced and the number of poles is reduced in a decoupled and coordinated manner. With the help of a fully distributed consistency algorithm, the distributed solution of the ARDED model is realized.

[0119] The local two-stage decision model of each sub-region α has a similar form to the centralized ARED model, and the formula is expressed as:

[0120]

[0121] There will be sub-region interconnection lines between each sub-region. In the ARDED sub-region model, the power injection from the adjacent sub-region and the power transmission to the adjacent sub-region are equivalently treated as the purchase and sale of electricity between adjacent regions, and this cost is included in the objective function of the two-stage decision of the sub-region ARDED. For the first stage decision, the formula is expressed as:

[0122]

[0123] For the second stage decision:

[0124]

[0125] in, represents the unit power exchange cost coefficient of adjacent sub-regions in the first stage of ARDED, Represents the redispatching cost coefficient for exchanging power between sub-regions in the second stage.

[0126] Since a new non-negative variable is introduced in the second stage objective function Add auxiliary constraints to the constraint set of the second-stage decision, and the formula is expressed as:

[0127]

[0128] Among them, G α represents the set of regular generators in subregion α, c 2,j ,c 1,j ,c 0,j represents the cost coefficient of conventional generator j, p g,j represents the power generated by conventional generator j, W α represents the set of wind farms in sub-region α, represents the uncertainty cost coefficient of wind power output, represents the predicted power generation of wind farm j, p g,jrepresents the actual power generation of wind farm j, represents the power exchange set between sub-region α and its adjacent sub-regions, represents the unit power exchange cost coefficient of the adjacent sub-region in the first stage, p ij represents the power exchange amount from node i to node j, {G α \W α} represents the set of conventional generators in sub-region α after removing the wind farm, represents the power adjustment cost coefficient of conventional generator j, represents the absolute active power change of conventional generator j, represents the critical wind power output cost coefficient of wind farm j, Critical wind power output of wind farm j, CL α represents the set of controllable loads in sub-region α, represents the power adjustment cost coefficient of controllable load j, represents the absolute active power change of the controllable load k, represents the redispatching cost coefficient of power exchange between sub-regions in the second stage, Represents the absolute change in power exchange from node i to node j.

[0129]

[0130] First, each sub-area measurement node S sends local measurement information Transmitted to each sub-region decision node via information branch Initialize the decision model. After the initialization is completed, enter the DCADMM fully distributed iteration process. In the iteration process, the first step is to make a decision Φ x For local variables The decision-making process corresponds to the optimization problem:

[0131]

[0132] After the first round of information exchange among decision nodes, each decision node exchanges the decision variable solutions from adjacent decision nodes. And through the calculation of the fully distributed interactive information branch model, the next stage decision Φ is obtained z The input vector The formula is:

[0133]

[0134] In fixed After the value of Φ z and Φ λ They correspond to the following decision-making processes, and the formulas are expressed as follows:

[0135]

[0136] After Q iterations until convergence, the optimal control decision obtained by convergence is The control target instruction υ is sent to the execution node A corresponding to each controllable unit.

[0137] S3: Use the fully distributed consistent ADMM algorithm to achieve a two-stage adaptive robust distributed intraday economic dispatch model solution.

[0138] The ARDED fully distributed decision and control process based on DCADMM is expressed as:

[0139]

[0140] First, each sub-area measurement node S sends local measurement information Transmitted to each sub-region decision node via information branch Initialize the decision model. After the initialization is completed, enter the DCADMM fully distributed iteration process. In the iteration process, the first step is to decide Φ x For local variables The decision-making process corresponds to the following optimization problem:

[0141]

[0142] The first-stage decision variables obtained after convergence To comprehensively consider the economic and cost-effectiveness of the second-stage redispatch, the coupling variable component is extracted from it. This is the final output of this step. After the first round of information exchange among decision nodes, each decision node exchanges the decision variable solutions from adjacent decision nodes. And through the calculation of the fully distributed interactive information branch model (3-40) established in Chapter 3, the next stage decision Φ is obtained z The input vector The models of the subsequent communication interaction processes are similar and will not be described in detail in this section.

[0143]

[0144] In fixed After the value of , all related terms are considered constants. z and Φ λ They correspond to the following decision-making processes:

[0145]

[0146] After Q iterations until convergence, the optimal control decision obtained by convergence is The control target instruction υ is sent to the execution nodes corresponding to each controllable unit (generator, wind power plant, flexible load)

[0147] It can be seen that Φ x It is the core step of local decision making in each sub-region, which corresponds to solving its local ARDED model. Analysis shows that each sub-region ARDED model has the following characteristics: (1) Its first-stage decision (4-26) is a quadratic programming (QCQP) problem with convex quadratic constraints; (2) Its second-stage decision (4-27) is a non-convex two-level max-min problem. However, for a given first-stage variable y f and uncertain variables u, the question is about y s Linear. Taking into account the above characteristics of the ARDED model, the Benders Decompostion (BD) algorithm and the Column & Constraints Generation (C&CG) algorithm can be used to design its solution strategy. With the help of the C&C&CG algorithm to design the solution strategy of the ARDED model, the basic step is that the optimal solution can always be obtained at a certain pole of the convex polyhedron. Therefore, the aforementioned two-stage ARDED problem can be divided into a relaxed master problem (Master Problem, MP) and a sub-problem (Slaver Problem, SP). SP is used to continuously add poles to MP to converge to the global optimal point after a finite number of alternating iterations. The relaxed main problem MP does not consider or only considers some of the scene uncertainty constraints. After solving, a lower bound LB of the optimal value is obtained, and a ground state solution y is passed to the sub-problem SP. f ; Subproblem SP finds the worst uncertain scenario u corresponding to a given base state solution l (l is the round index of this iteration) and its corresponding variable correction value, thus obtaining an upper bound UB of the original ARDED problem. Afterwards, SP will use u l The generated cutting plane is added back to the relaxed main problem MP, thereby tightening the feasible region of MP and affecting the MP ground state solution y in the next iteration f . In the above classical C&CGG algorithm, the coupling constraints of the first-stage decision variables and the second-stage decision variables are required to be linear. However, in the ARDED model, the coupling constraints of the first-stage and second-stage decision variables include not only linear constraints but also nonlinear SOCP constraints. Therefore, it is necessary to make appropriate improvements to the classical C&CG algorithm. The following will derive the mathematical expressions of each step of the improved C&CG algorithm applicable to the ARDED model. To simplify the expression, the formulas are explained and derived in compact matrix and vector forms, and the sub-region subscripts are omitted.

[0148]

[0149] Using the fully distributed consistent ADMM algorithm, a two-stage adaptive robust distributed intraday economic dispatch model is solved, which consists of two layers: the outer DCADMM iteration and the inner C&CG iteration. The role of the outer DCADMM iteration is to coordinate the coupling variables x of multiple adjacent sub-regions. a , so that it finally satisfies the consistency constraint. The first step of DCADMM iteration is used as the entry of the inner C&CG algorithm iteration and is passed to the inner iteration. as well as As a fixed parameter; through the iterative calculation of the inner layer C&CG algorithm, we get Returns the outer DCADMM as the mirror variable for subsequent DCADMM updates and the multiplier variable The required fixed parameters and the remaining decision and communication interaction steps are located in the outer DCADMM iteration, which are executed in sequence as described above without calling the inner iteration. When the outer DCADMM meets the convergence criterion, ARDED is output.

[0150] On the other hand, this embodiment also provides a distribution system intraday economic dispatch system considering power generation uncertainty, which includes:

[0151] The acquisition module collects real-time data of the distribution system and establishes a two-stage adaptive robust centralized intraday economic dispatch model.

[0152] The partitioning module divides the global model into several sub-areas according to the physical structure of the distribution system, and establishes a two-stage adaptive robust distributed intraday economic dispatch model.

[0153] The solution module uses the fully distributed consistent ADMM algorithm to achieve a two-stage adaptive robust distributed intraday economic dispatch model solution.

[0154] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0155] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0156] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0157] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0158] Example 2

[0159] For the IEEE-33 node CPDS test system, this section shows its maximum deviation parameter Δw g=0.2MW. The results of the first-stage pre-dispatch and the second-stage re-dispatch are shown in Table 1. Analyzing the results of the first-stage decision, since the wind abandonment penalty cost in the objective function is higher than the marginal cost of the distributed generator output, the optimal pre-dispatch scheme is forced to reduce the wind abandonment as much as possible. Therefore, it is observed that the wind power plant output pre-dispatch plan value at bus #9 and #19 is the same as the predicted value, thereby achieving the maximum utilization of wind power generation. The second-stage decision reveals that in Δw g =0.2MW, the worst wind power output uncertainty scenario of CPDS: the wind power plant output at bus #9 and #19 deviates from the predicted value by -0.2MW at the same time (i.e., it drops to 0.3MW and 0.2MW respectively, resulting in an active power shortage of 0.4MW). In order to compensate for this power shortage, distributed generators #24 and #27 increase their active output by 0.274MW and 0.055MW respectively, and the flexible loads at bus 7, 8, 25 and 30 reduce the total power demand by 0.066MW, thus achieving the best power shortage.

[0160] Balance of power shortage in severe uncertain scenarios. In addition, in the second-stage decision, the active power regulation at substation bus #1 (i.e., the common connection bus between CPDS and the external transmission network) is 0, which means that the power shortage caused by the uncertainty of wind power output has been completely balanced within CPDS, so that the exchange power between CPDS and the external transmission network is stable at the planned value, reducing the power balance burden of the transmission network. In addition, Figure 4 The results of pre-scheduling and re-scheduling of node voltages, as well as the SOCP relaxation errors of the voltage and current constraints and complex power constraints of each branch are shown. It can be seen that the voltages of all nodes are maintained in a reasonable voltage range during the pre-scheduling and re-scheduling stages. In order to prevent the system voltage from exceeding the limit in the worst scenario, the voltages of some nodes are moderately reduced during the pre-scheduling stage, so that the system voltage can still be maintained at a reasonable level after the voltage adjustment during the re-scheduling stage. For the 32 branches, it can be seen that the SOCP relaxation errors introduced by constraints (4-16) are all 10 -7 This means that although this constraint is an inequality constraint in the form of less than or equal to, the final dispatch result strictly satisfies the equality constraint, that is, the quadratic equality constraint ν of voltage, current and complex power i l ij =(p ij ) 2 +(q ij 0 2 , which confirms the accuracy of the scheduling results. The experimental results are shown in Table 1.

[0161] Table 1 Cost and operating parameters of each generator set (118 node CPDs)

[0162]

[0163] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for intraday economic dispatch of distribution systems considering power generation uncertainty, characterized in that: include: Collect real-time data of the distribution system and establish a two-stage adaptive robust centralized intraday economic dispatch model; According to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established. The fully distributed consistent ADMM algorithm is used to solve the two-stage adaptive robust distributed intraday economic dispatch model.

2. The method for intraday economic dispatch of a power distribution system considering power generation uncertainty according to claim 1, characterized in that: The real-time data collected from the power distribution system includes load data, wind power and photovoltaic power generation data, conventional power generation data, power distribution system status data and uncertainty data; Data is collected through smart meters, micro-phasor measurement units, distributed sensors and measurement equipment in wind and photovoltaic power plants.

3. The intraday economic dispatch method of a power distribution system considering power generation uncertainty as claimed in claim 2, characterized in that: The establishment of a two-stage adaptive robust centralized intraday economic dispatch model includes a first-stage decision and a second-stage decision; The first stage decision includes the first stage decision variable definition, the first stage objective function definition and the first stage constraint conditions; The second stage decision includes the definition of the second stage decision variables, the definition of the second stage objective function and the second stage constraints; The first stage decision variable definition y f The formula is: Where ν represents the set of all buses in the physical system, ξ represents the set of all distribution lines, G represents all power generation units, ν represents the square of the bus voltage amplitude of the decision variable, and p ij represents the active power on the directional distribution line i,j), q ij represents the reactive power on the directional distribution line (i, j), l ij represents the square of the current amplitude on the directional distribution line (i, j), p g Represents the active output of the generating unit, q g Indicates the reactive power output of the generating unit; The definition of the objective function in the first stage includes: the decision-making goal of the first stage is based on the predicted value of wind power generation Minimize the power generation cost and wind curtailment cost of all power generation units. The formula is: Among them, c2, c1, c0 are the coefficients of the secondary power generation cost function of the generator, ρ cr represents the penalty cost coefficient for wind curtailment; represents the set of wind turbines, ∑ j∈G represents the sum of all conventional generators j, c 2,j represents the quadratic cost coefficient of conventional generator j, c 1,j represents the primary cost coefficient of conventional generator j, c 0,j represents the fixed cost of conventional generator j, p g,j represents the power generated by conventional generator j, represents the power generation cost of conventional generator j, Σ j∈W represents the sum of all wind farms j, represents the uncertainty cost coefficient of wind power, represents the predicted output of wind farm j, p g,j represents the actual power generation of wind farm j, Represents the cost caused by the uncertainty of wind power output.

4. The intraday economic dispatch method of a power distribution system considering power generation uncertainty as claimed in claim 3, characterized in that: The first stage constraints include: is the active and reactive balance equation of each bus, and the formula is expressed as: for Using the second-order cone relaxation SOCR technique, the original quadratic equality constraint ν is relaxed using the second-order cone i l ij =(p ij ) 2 +(q ij ) 2 : ||2p ij ;2q ij ;l ij -in i ||2≤l ij +v i The upper and lower amplitude boundaries of each bus voltage are defined: The output convex constraint set of each power generation unit first limits all units The power generation capacity must meet the upper and lower bounds of the power generation capacity; In addition, the active output of wind turbines is limited to no higher than the predicted value. And the power factor φ must satisfy the interval constraint φ∈[φ min ,φ max ]; Represents the power flow constraint of the distribution line, and the formula is expressed as: l ij ≥0 Among them, p g,j represents the active power of the generator at node j, p d,j represents the load active power of node j, Σ( i,j)∈ξ (p ij -r ij l ij ) represents the active power flow from other nodes i to node j minus the line loss, ∑( j,k)∈ξ p jk represents the line active power flow from node j to other nodes k, g j ν j represents the power consumption caused by the ground admittance of node j, q g,j represents the reactive power of the generator at node j, q d,j represents the load reactive power of node j, ∑( i,j)∈ξ (q ij -x ij l ij ) represents the line reactive power flow from other nodes i to node j minus line losses, Σ j,k)∈ξ q jk represents the line reactive power flow from node j to other nodes k, b j ν j represents the reactive power consumption caused by the ground admittance of node j, ν j represents the square of the voltage at node j, ν i represents the square of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, l ij represents the square of the current in the line between node i and node j, p ij represents the active power of the line from node i to node j, q ij represents the line reactive power from node i to node j, ||2p ij ;2q ij ; l ij -ν i ||2 represents the quadratic cone norm, represents the minimum value of the square of the voltage at node j, represents the maximum value of the square of the voltage at node j, represents the lower limit of active power of conventional generator j, represents the upper limit of active power of conventional generator j, represents the lower limit of reactive power of conventional generator j, represents the upper limit of reactive power of conventional generator j, p g,j represents the active power of wind farm j, q g,j represents the reactive power of wind farm j, represents the predicted output of wind farm j, tanφ min Represents the reactive power coefficient corresponding to the lower limit of the wind farm power factor, tanφ max The reactive power coefficient corresponding to the upper limit of the wind farm power factor, q ij represents the reactive power flow of line (i, j), l ij represents the square of the current of line (i, j, p ij represents the active power flow of line (i, j), r ij represents the resistance of line (i, j), x ij represents the reactance of line (i,j), represents the maximum value of active power flow of line (i, j), represents the maximum value of reactive power flow of line i,j); The first stage decision is a deterministic DED optimization based on the predicted value of wind power generation. The form of the first stage constraints conforms to the global DED model in Chapter 3.

5. The method for intraday economic dispatch of a power distribution system considering power generation uncertainty as claimed in claim 4, characterized in that: The second stage decision includes: the second stage decision is used to quantify the redispatch cost of CPDS to readjust the power balance based on the first stage decision solution when the actual value of renewable energy output is determined; Therefore, the second stage decision process is a correction of the first stage decision solution; The adjustable units in the second stage of decision-making include: distributed generators, wind turbines, and flexible loads; The first stage decision variable y s Definition, the formula is expressed as: in, represents the voltage change of node j, V represents the set of all nodes, represents the active power change of line i,j), The reactive power change of line (i,j), represents the current change of line (i, j), ξ represents the set of all lines, represents the active power change of conventional generator j, represents the reactive power change of conventional generator j, represents the active power change of conventional generator j, G represents the set of conventional generators, represents the critical wind power output of wind farm j, W represents the set of wind farms, represents the change in active power of load j, represents the reactive power change of load j, represents the absolute change in active power of load j, and L represents the set of load nodes; The definition of the objective function of the second stage includes: The objective function of the second stage decision is to minimize the corrective power regulation cost Q of CPDS under the worst uncertainty scenario u∈U s The cost includes four parts: generator power regulation cost, wind abandonment penalty, flexible load regulation cost and uncontrollable load shedding penalty. This part models it as a linear function of the corresponding correction variable. CL represents the load bus set where the controllable flexible load is located; the coefficient are the price coefficients of the above four adjustment costs, and the formula is: in, represents the objective function, Q s Represents the state variable y s The total cost function under represents the absolute change in active power of conventional generator j, represents the power adjustment cost coefficient of conventional generator j, G\W represents the set of conventional generators in set G after removing the wind farm set W, represents the critical wind power output of wind farm j, represents the critical output cost coefficient of wind farm j, W represents the set of wind farms, represents the absolute change in active power of the uncontrollable load k, represents the power adjustment cost coefficient of controllable load j, CL represents the set of controllable loads, represents the absolute change in active power of the uncontrollable load k, represents the load reduction cost coefficient of the non-controllable load j, L\CL represents the non-controllable load set after removing the controllable load set CL from the set L; The constraint formula of the second stage is expressed as: in, represents the change in active power of the generator at node j, represents the load active power change of node j, represents the change in active power flow from other nodes i to node j minus the change in line loss, represents the change in active power flow from node j to other nodes k, represents the power change caused by the voltage change caused by the ground admittance of node j, represents the square change of the voltage at node j, represents the square change of the voltage at node i, r ij represents the resistance of the line between node i and node j, x ij represents the reactance of the line between node i and node j, represents the square of the current change in the line between node i and node j, represents the change in line active power from node i to node j, represents the change in line reactive power from node i to node j, represents the quadratic cone norm, It means that the total change of active power of wind farm j is equal to the power generation of the wind farm minus the critical wind power output. Indicates that the critical wind power output must be a non-negative value, Indicates that the reactive power change of the wind farm must meet the power factor lower limit constraint, Indicates that the reactive power change of the wind farm must meet the upper limit constraint of the power factor; Among them, β min Indicates the minimum proportional factor of load power change, β max Indicates the maximum proportional factor of load power change, p d,j represents the original active load of node j, q d,j represents the original reactive load of node j. In the above constraints, {p g ,q g ,p ij ,q ij ,l ij ,ν} represents the "base state" operating point obtained by the first stage decision-making, which is regarded as a known constant with a fixed value in the second stage decision-making model.

6. The method for intraday economic dispatch of a power distribution system considering power generation uncertainty as claimed in claim 5, characterized in that: According to the physical structure of the distribution system, the global model is divided into several sub-areas, and a two-stage adaptive robust distributed intraday economic dispatch model is established. This includes decomposing the above centralized ARED optimization problem into decision nodes corresponding to each physical sub-area through a fully distributed optimization method to form an adaptive robust distributed intraday economic dispatch model. The problem scale is reduced and the number of poles is reduced in a decoupled and coordinated manner. With the help of a fully distributed consistency algorithm, the distributed solution of the ARDED model is realized; The local two-stage decision model of each sub-region α has a similar form to the centralized ARED model, and the formula is expressed as: There will be sub-region interconnection lines between each sub-region. In the ARDED sub-region model, the power injection from the adjacent sub-region and the power transmission to the adjacent sub-region are equivalently treated as the purchase and sale of electricity between adjacent regions, and this cost is included in the objective function of the two-stage decision of the sub-region ARDED. For the first stage decision, the formula is expressed as: For the second stage decision: in, represents the unit power exchange cost coefficient of adjacent sub-regions in the first stage of ARDED, represents the redispatching cost coefficient of power exchange between sub-regions in the second stage; Since a new non-negative variable is introduced in the second stage objective function Add auxiliary constraints to the constraint set of the second-stage decision, and the formula is expressed as: Among them, G α represents the set of regular generators in subregion α, c 2,j ,c 1,j ,c 0,j represents the cost coefficient of conventional generator j, p g,j represents the power generated by conventional generator j, W α represents the set of wind farms in sub-region α, represents the uncertainty cost coefficient of wind power output, represents the predicted power generation of wind farm j, p g,j represents the actual power generation of wind farm j, represents the power exchange set between sub-region α and its adjacent sub-regions, represents the unit power exchange cost coefficient of the adjacent sub-region in the first stage, p ij represents the power exchange amount from node i to node j, {G α \W α } represents the set of conventional generators in sub-region α after removing the wind farm, represents the power adjustment cost coefficient of conventional generator j, represents the absolute active power change of conventional generator j, represents the critical wind power output cost coefficient of wind farm j, Critical wind power output of wind farm j, CL α represents the set of controllable loads in sub-region α, represents the power adjustment cost coefficient of controllable load j, represents the absolute active power change of the controllable load k, represents the redispatching cost coefficient of power exchange between sub-regions in the second stage, represents the absolute change in power exchange from node i to node j; First, each sub-area measurement node S sends local measurement information Transmitted to each sub-region decision node via information branch Initialize the decision model. After the initialization is completed, enter the DCADMM fully distributed iteration process. In the iteration process, the first step is to make a decision Φ x For local variables The decision-making process corresponds to the optimization problem: After the first round of information exchange among decision nodes, each decision node exchanges the decision variable solutions from adjacent decision nodes. And through the calculation of the fully distributed interactive information branch model, the next stage decision Φ is obtained z The input vector The formula is: In fixed After the value of Φ z and Φ λ They correspond to the following decision-making processes, and the formulas are expressed as follows: After Q iterations until convergence, the optimal control decision obtained by convergence is The control target instruction υ is sent to the execution node A corresponding to each controllable unit.

7. The method for intraday economic dispatch of a power distribution system considering power generation uncertainty as claimed in claim 6, characterized in that: Using the fully distributed consistent ADMM algorithm, a two-stage adaptive robust distributed intraday economic dispatch model is solved, which consists of two layers: the outer DCADMM iteration and the inner C&CG iteration. The role of the outer DCADMM iteration is to coordinate the coupling variables x of multiple adjacent sub-regions. a , so that it finally satisfies the consistency constraint. The first step of DCADMM iteration is used as the entry of the inner C&CG algorithm iteration and is passed to the inner iteration. as well as As a fixed parameter; through the iterative calculation of the inner layer C&CG algorithm, we get Returns the outer DCADMM as the mirror variable for subsequent DCADMM updates and the multiplier variable The required fixed parameters and the remaining decision and communication interaction steps are located in the outer DCADMM iteration, which are executed in sequence as described above without calling the inner iteration. When the outer DCADMM meets the convergence criterion, ARDED is output.

8. An intraday economic dispatch system for a power distribution system taking into account power generation uncertainty using the method as claimed in any one of claims 1 to 7, characterized in that: The acquisition module collects real-time data of the distribution system and establishes a two-stage adaptive robust centralized intraday economic dispatch model; The partitioning module divides the global model into several sub-areas according to the physical structure of the distribution system, and establishes a two-stage adaptive robust distributed intraday economic dispatch model; The solution module uses the fully distributed consistent ADMM algorithm to achieve a two-stage adaptive robust distributed intraday economic dispatch model solution.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the method for intraday economic dispatch of a distribution system taking into account power generation uncertainty are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for intraday economic dispatch of a power distribution system taking into account power generation uncertainty are implemented.

Images