Distributed main and distribution network scheduling method and device, electronic equipment and storage medium

Abstract

According to the distributed main and distribution network scheduling method and device, the electronic equipment and the storage medium provided by the invention, in the kth iteration, the first virtual load capacity, the first virtual power generation capacity and the penalty function multiplier of the target power distribution network are obtained and then input to the power distribution network optimization scheduling model to obtain the second virtual power generation capacity of the target power distribution network; and inputting the second virtual generator quantity, the penalty function multiplier and the first virtual load quantity into the main network optimization scheduling model to obtain a second virtual load quantity of the main network, transmitting the second virtual load quantity to the target power distribution network, and outputting a scheduling result if the second virtual generator quantity of the power distribution network and the second virtual load quantity of the main network conform to a cyclic criterion. According to the technical scheme, by calculating the virtual load capacity and the virtual generator capacity, the scheduling relation between the power distribution network and the main network is continuously optimized, the reasonability of the main and distribution network scheduling strategy is ensured, whether the convergence standard is met or not is judged by adopting the cyclic criterion, and therefore the reasonability and the calculation efficiency of the main and distribution network scheduling strategy are improved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a dispatching method, device, electronic equipment, and storage medium for a distributed main distribution network. Background Art

[0002] With the accelerated construction of new power systems, the high proportion of renewable energy access and the growth of diversified loads have posed severe challenges to the coordinated dispatching of the main and distribution networks of the power system.

[0003] Traditional coordinated dispatching methods for main and distribution networks mainly adopt a centralized dispatching architecture, rely on the complete topological parameters of the distribution network to build an accurate physical model, and achieve unified dispatching calculations through full-network state estimation. However, the existing power system optimization dispatching has low coordination efficiency between the main and distribution networks and opaque topological parameters on the distribution network side, resulting in low rationality and accuracy of the main and distribution network dispatching strategies.

[0004] Therefore, how to improve the rationality and computational efficiency of the distributed main and distribution network scheduling strategy has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The embodiments of the present application provide a scheduling method, device, electronic device and storage medium for a distributed main distribution network to solve the problems of unreasonable main distribution network scheduling strategy and low computing efficiency in related technologies.

[0006] In a first aspect, an embodiment of the present application provides a scheduling method for a distributed main distribution network, including:

[0007] Step 11: In the kth iteration, obtain the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier of the target distribution network; the target distribution network is any distribution network in the distributed main distribution network; k is a positive integer;

[0008] Step 12: Input the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity of the target distribution network;

[0009] Step 13: Input the second virtual generator quantity of the target distribution network, the penalty function multiplier and the first virtual load quantity into the main grid optimization scheduling model for calculation, obtain the second virtual load quantity of the main grid and transmit it to the target distribution network;

[0010] Step 14: determine whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion formula;

[0011] Step 15: If the condition does not meet the requirement, update the number of iterations k=k+1 and the penalty function multiplier, use the second virtual generator quantity as the new first virtual generator quantity, use the second virtual load quantity as the new first virtual load quantity, and repeat steps 12 to 14 until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion.

[0012] Step 16: If it meets the requirements, output the scheduling result.

[0013] In one possible implementation, the distribution network optimization scheduling model is trained based on the distribution network objective function, sample load, sample generator quantity and penalty function multiplier; wherein the distribution network objective function is constructed based on the distribution network loss cost, voltage deviation penalty cost, distribution network power purchase cost from the main grid and distribution network constraints.

[0014] In one possible implementation, the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier are input into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity of the target distribution network, including:

[0015] performing normalization processing on the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier respectively;

[0016] The standardized first virtual load quantity, the first virtual generator quantity and the penalty function multiplier are input into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity.

[0017] In one possible implementation, the main grid optimization scheduling model is trained based on the main grid objective function, penalty function multiplier, sample generator quantity and sample load quantity; wherein the main grid objective function is constructed based on the power generation cost of the coal-fired unit, the cost of selling electricity from the main grid to the distribution network and the main grid constraints.

[0018] In a possible implementation, the main grid constraints include: linearized power flow constraints, linearized power balance constraints, coal-fired unit output constraints, unit output constraints, and key section constraints.

[0019] In one possible implementation, the process of updating the penalty function multiplier includes:

[0020] By formula Update the penalty function multiplier;

[0021] Among them, δ is a preset constant, λ d,t,k is the linear multiplier of the Lagrangian penalty function of the dth distribution network at time t in the kth iteration, μ d,k,t is the quadratic multiplier of the Lagrange penalty function of the dth distribution network at time t in the kth iteration, The sample load of the main network transmitted to the d-th distribution network in the k-th iteration, The sample generator quantity obtained from the d-th distribution network optimization in the k-th iteration.

[0022] In a second aspect, an embodiment of the present application provides a scheduling device for a distributed main distribution network, including:

[0023] An acquisition module is configured to acquire, in a kth iteration, a first virtual load quantity, a first virtual generator quantity, and a penalty function multiplier of a target distribution network; the target distribution network is any distribution network in a distributed main distribution network; and k is a positive integer;

[0024] a calculation module, configured to input the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity of the target distribution network;

[0025] The calculation module is further used to input the second virtual generator quantity, penalty function multiplier and first virtual load quantity of the target distribution network into the main network optimization scheduling model for calculation, obtain the second virtual load quantity of the main network and transmit it to the target distribution network;

[0026] A judgment module, used for judging whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion formula;

[0027] an updating module, configured to update the number of iterations k=k+1 and the penalty function multiplier when the judgment result of the judging module is non-compliance, and use the second virtual generator quantity as the new first virtual generator quantity and the second virtual load quantity as the new first virtual load quantity;

[0028] a triggering module, configured to execute processing from the calculation module to the judgment module according to a result of the updating module, until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion formula;

[0029] The output module is used to output the scheduling result when the judgment result of the judgment module is in compliance.

[0030] In one possible implementation, the distribution network optimization scheduling model is trained based on the distribution network objective function, sample load, sample generator quantity and penalty function multiplier; wherein the distribution network objective function is constructed based on the distribution network loss cost, voltage deviation penalty cost, distribution network power purchase cost from the main grid and distribution network constraints.

[0031] In a possible implementation, the computing module is specifically configured to:

[0032] performing normalization processing on the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier respectively;

[0033] The standardized first virtual load quantity, the first virtual generator quantity and the penalty function multiplier are input into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity.

[0034] In one possible implementation, the main grid optimization scheduling model is trained based on the main grid objective function, penalty function multiplier, sample generator quantity and sample load quantity; wherein the main grid objective function is constructed based on the power generation cost of the coal-fired unit, the cost of selling electricity from the main grid to the distribution network and the main grid constraints.

[0035] In a possible implementation, the main grid constraints include: linearized power flow constraints, linearized power balance constraints, coal-fired unit output constraints, unit output constraints, and key section constraints.

[0036] In a possible implementation, the update module is specifically configured to:

[0037] By formula Update the penalty function multiplier;

[0038] Among them, δ is a preset constant, λ d,t,k is the linear multiplier of the Lagrangian penalty function of the dth distribution network at time t in the kth iteration, μ d,k,t is the quadratic multiplier of the Lagrange penalty function of the dth distribution network at time t in the kth iteration, The sample load of the main network transmitted to the d-th distribution network in the k-th iteration, The sample generator quantity obtained from the d-th distribution network optimization in the k-th iteration.

[0039] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0040] Memory stores computer-executable instructions;

[0041] The processor executes the computer-executable instructions stored in the memory to implement the method as described in the first aspect or any one of the above-mentioned methods.

[0042] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method of the first aspect or any of the above-mentioned methods.

[0043] In a fifth aspect, an embodiment of the present application provides a computer program, wherein a computer program product includes a computer program, wherein the computer program is stored in a computer-readable storage medium, and at least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, the method of the first aspect or any one of the above methods can be implemented.

[0044] The embodiments of the present application provide a scheduling method, device, electronic device and storage medium for a distributed main distribution network. The method first obtains a first virtual load amount, a first virtual generator amount and a penalty function multiplier of a target distribution network in the kth iteration, wherein the target distribution network is any distribution network in the distributed main distribution network; k is a positive integer, and then the first virtual load amount, the first virtual generator amount and the penalty function multiplier of the target distribution network are input into a distribution network optimization scheduling model for calculation to obtain a second virtual generator amount of the target distribution network, and then the second virtual generator amount, the penalty function multiplier and the first virtual load amount of the target distribution network are input into the main network optimization scheduling model for calculation to obtain a second virtual load amount of the main network and transmit it to the target distribution network, and judge the distribution network. The second virtual generator quantity of the power grid and the second virtual load quantity of the main grid meet the cyclic criterion. If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main grid do not meet the cyclic criterion, the iteration number k=k+1 and the penalty function multiplier are updated, the second virtual generator quantity is used as the new first virtual generator quantity, the second virtual load quantity is used as the new first virtual load quantity, and the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier are repeatedly input into the distribution network optimization scheduling model for calculation until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main grid meet the cyclic criterion. If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main grid meet the cyclic criterion, the scheduling result is output. This technical solution calculates the virtual load quantity and the virtual generator quantity, and combines the penalty function multiplier to perform constrained optimization, continuously iteratively optimizes the scheduling relationship between the distribution network and the main grid, ensures the rationality of the main distribution network scheduling strategy, and uses the cyclic criterion to determine whether the convergence standard has been met, thereby improving the rationality and computational efficiency of the main distribution network scheduling strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0046] Figure 1 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 1 ;

[0047] Figure 2 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 2 ;

[0048] Figure 3 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 3 ;

[0049] Figure 4A schematic diagram of the structure of a dispatching device for a distributed main distribution network provided in an embodiment of the present application;

[0050] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0051] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0052] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0053] Before introducing the embodiments of the present application, the application background of the embodiments of the present application is first explained:

[0054] With the accelerated construction of new power systems, the high proportion of renewable energy access and the growth of diversified loads have posed severe challenges to the coordinated dispatching of the main and distribution networks of the power system.

[0055] Traditional coordinated dispatch methods for main and distribution networks primarily employ a centralized dispatch architecture, relying on the complete topological parameters of the distribution network to construct a precise physical model and achieving unified dispatch calculations through full-network state estimation. However, in actual engineering applications, the distribution network topology often exhibits dynamic, time-varying characteristics due to factors such as frequent switching of distributed power sources and network reconfiguration operations. Furthermore, insufficient coverage of measurement devices leads to incomplete parameter acquisition, resulting in problems such as accumulated modeling errors and poor adaptability in traditional physical models. In particular, in scenarios with high renewable energy penetration, strong uncertainty on both the source and load sides leads to a dimensional explosion in centralized dispatch optimization. Its computational complexity increases exponentially with the number of distribution networks, making it difficult to meet the timeliness requirements of real-time dispatch.

[0056] Related technologies, such as distributed optimization based on the target cascade method, can achieve decoupled calculations for the main and distribution networks, but they still require the distribution network to provide accurate power flow equations as constraints, and cannot break through the rigid reliance on topological information. Some data-driven methods proposed in related technologies attempt to replace some physical models through machine learning, but they are limited to optimization within a single voltage level and fail to address the coordination of boundary-coupled variables in cross-level collaboration of the main and distribution networks. Furthermore, current uncertainty handling methods such as robust optimization and stochastic programming often sacrifice economic efficiency for safety, lacking a systematic solution that balances topological flexibility, computational efficiency, and robustness.

[0057] Therefore, how to improve the rationality and computational efficiency of the distributed main and distribution network scheduling strategy has become a technical problem that needs to be solved urgently.

[0058] In response to the technical problems existing in the relevant technologies, the inventors of this application have the following ideas: to solve the problems of unreasonable main and distribution network scheduling strategies and low computational efficiency in the relevant technologies, a distributed framework is adopted to realize information decoupling and collaborative optimization between the main network and the distribution network. The layer-by-layer scheduling layer serves as the upper-layer optimization subject, and constructs a main network optimization scheduling model with the goal of minimizing power generation costs and optimizing voltage levels. The distribution network, on the basis of meeting its own load demand, constructs a distribution network optimization model with the goal of minimizing network losses and pursuing the optimal solution of voltage levels. The main and distribution network scheduling layer serves as a cross-layer interaction hub, and a distributed optimization framework based on the target cascade method is designed. By constructing a correction model that considers the interconnection line power and boundary condition parameters, two-way information decoupling and collaborative convergence between the main network and the distribution network are realized, and finally a rationalized and complete optimization scheduling strategy with consistent boundaries is formed. In addition, a cyclic criterion is used to judge whether the convergence standard is met, thereby improving the computational efficiency of the main and distribution network scheduling strategy.

[0059] The technical solution of the present application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0060] Among them, the execution subject of this application is an electronic device, which can be a server, terminal device, etc.

[0061] Figure 1 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 1 ,like Figure 1 As shown, the method may include the following steps:

[0062] Step 11: In the kth iteration, obtain the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier of the target distribution network.

[0063] Among them, the target distribution network is any distribution network in the distributed main distribution network; k is a positive integer.

[0064] In this step, the first virtual load and the first virtual generator of the target distribution network are obtained for subsequent calculation and optimization. At the same time, the penalty function multiplier is also obtained to control the load adjustment process between the distribution network and the main grid to ensure the stability and efficiency of the entire power system.

[0065] Step 12: Input the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity of the target distribution network.

[0066] In this step, the aforementioned first virtual load quantity, first virtual generator quantity and penalty function multiplier are taken as input and substituted into the distribution network optimization scheduling model for calculation. By accurately calculating the load and power generation of the target distribution network, taking into account multiple factors such as load fluctuation, power generation capacity, and system stability, a new second virtual generator quantity is output.

[0067] The second virtual generator capacity represents the virtual power generation capacity required by the target distribution network under current conditions to meet power demand and ensure stable operation of the power system.

[0068] The distribution network optimization dispatching model is expressed in the following mathematical formula:

[0069]

[0070] Where, F DN is the distribution network objective function, λ d,k,t 、μ d,k,t are the linear multiplier and quadratic multiplier of the Lagrange penalty function for the d-th distribution network at time t during the k-th iteration, is the tie line power (i.e., sample load) transmitted from the main grid to the distribution grid at time t in the kth iteration, is the tie line demand power (i.e., the sample generator quantity) obtained by optimizing the distribution network at time t during the k-th iteration.

[0071] Specifically, the distribution network optimization scheduling model is trained based on the distribution network objective function, sample load quantity, sample generator quantity and penalty function multiplier.

[0072] Among them, the distribution network objective function is constructed based on the distribution network loss cost, voltage deviation penalty cost, the cost of purchasing electricity from the distribution network to the main grid, and the distribution network constraints.

[0073] In this implementation, the training process of the distribution network optimization scheduling model includes:

[0074] First, a sample data set is obtained, which includes sample load, sample wind and solar output, sample voltage amplitude, and sample transmission power received by the head-end node of each node between the main grid and the distribution network. Then, an equivalent network of the main grid and the distribution network is created based on the samples in the sample data set. The equivalent network includes multiple nodes, and the node connected to the main grid is the head-end node. The sample load, sample wind and solar output, and sample transmission power are input into the initial model to obtain the predicted voltage amplitude. Then, the loss value between the predicted voltage amplitude and the sample voltage amplitude is calculated according to the pre-constructed loss function. Finally, the weight parameters of the initial model are adjusted according to the loss value until the initial model reaches the preset convergence condition, and the distribution network optimization scheduling model is obtained.

[0075] Optionally, the pre-constructed loss function is a mean square error loss function, and the preset convergence condition is that the calculation result of the preset convergence algorithm converges (such as the Newton-Raphson method).

[0076] The process of obtaining the sample data set includes:

[0077] First, the initial sample data is obtained, and then the initial sample data is calculated based on the Newton-Raphson method to obtain the calculation results corresponding to the initial sample data. The initial sample data whose calculation results meet the preset conditions are normalized to obtain the sample data in the sample data set.

[0078] The preset condition is that the calculation result of the preset convergence algorithm converges (ie, the Newton-Raphson method).

[0079] The above distribution network objective function is expressed as follows using mathematical formula:

[0080]

[0081] Where, f d Distribution network loss cost, f v Voltage deviation penalty cost, f buy The cost of purchasing electricity from the main grid for the distribution network; c loss is the distribution network loss cost coefficient, c v is the voltage stability target coefficient, c buy I is the electricity purchase price of the tie line; d,l,t is the current of the lth line in the dth distribution network in period t, R d,l is the resistance value of the lth line in the dth distribution network; is the square of the voltage at node i at time t, U ref is the square of the reference voltage, usually set to 1. is the tie line power required by the main network for the d-th distribution network; T is the time set, N is the node set, L is the line set, and D is the distribution network set.

[0082] In the actual operation scenario of the power system, the integrity and accuracy of the distribution network topology are often limited by practical constraints such as device heterogeneity, insufficient coverage of measurement devices, and dynamic network reconstruction. The embodiment of the present application adopts a distribution network equivalent modeling technology based on data analysis, breaking through the strong dependence of traditional methods on topology, building a neural network model based on CNN, and constructing a functional relationship between the active and reactive power injected into the node and the node voltage amplitude as shown in formula (3):

[0083]

[0084] Where V d 、P d With Q d They are the voltage amplitude matrix of the distribution network nodes and the injected active and reactive power matrices respectively.

[0085] Among them, P d The active power P injected into node j j sum, Q d The reactive power Q injected into node j j The sum of the active power P injected by node j j and the reactive power Q injected by node j j The standard nonlinear expression of the distribution network power flow equation is expressed as follows:

[0086]

[0087] Where, v j is the voltage amplitude at node j, g ij is the imaginary part of the elements in the node admittance matrix.

[0088] Formula (4) can be simplified to the form shown in formula (5):

[0089] S=g(V) (5)

[0090] Where S is the power matrix injected into the node, V is the node voltage amplitude and phase angle matrix,

[0091] g(·) is a nonlinear function.

[0092] After Taylor series expansion, equation (5) can be expressed as:

[0093] S0+ΔS=g(V0+ΔV)=g(V0)+J1ΔV+o(ΔV) (6)

[0094] Where J1 is the Jacobian matrix, which is equivalent to the slope of the function g(·) at V0. To control the linearization error, J1 needs to be recalculated in each iteration of the power flow calculation. Since S0 = g(V0), the equation shown in Equation (7) can be obtained, that is, the node injection power is used as the input feature:

[0095]

[0096] Where K1 is the partial derivative matrix of ΔV with respect to ΔS. A continuous and nonzero K1 proves that there is a unique mapping from V to S. Considering that node voltage phase angles are rarely used in distribution network optimization and scheduling, to simplify the training process, the phase angle term is ignored, and only the node voltage amplitude is used as the output feature in ΔV.

[0097] Considering the incomplete distribution network topology information, the embodiment of the present application constructs the functional relationship of the tie line power based on the neural network as shown in formula (8):

[0098]

[0099] Where, P trans The active power of the main network transmitted to the distribution network tie line, Q trans The reactive power transmitted from the main grid to the distribution network tie line.

[0100] The power flow branch of the distribution network is represented as shown in formula (9):

[0101]

[0102] Formula (9) can also be derived into the form of formula (10)-formula (11):

[0103] R=h(V) (10)

[0104] R0+ΔR=h(V0+ΔV)=h(V0)+J2ΔV+o(ΔV) (11)

[0105] Where R is the branch transmission power matrix. Combining R0=h(V0) with equation (7), we can know:

[0106] ΔR=J2ΔV=J2K1ΔS=LΔS (12)

[0107] Where J2 is the first-order partial derivative matrix of branch transmission power R with respect to node voltage V, and L is the partial derivative matrix of ΔR with respect to ΔS, which proves that there is a one-to-one mapping relationship between R and S.

[0108] Step 13: Input the second virtual generator quantity, penalty function multiplier and first virtual load quantity of the target distribution network into the main grid optimization scheduling model for calculation, obtain the second virtual load quantity of the main grid and transmit it to the target distribution network.

[0109] In this step, the calculation results for the target distribution network are linked to those for the main grid. The target distribution network's second virtual generator quantity, penalty function multiplier, and first virtual load are input into the main grid's optimal scheduling model for calculation. Based on this input, the main grid's optimal scheduling model calculates the main grid's second virtual load and then transmits it to the target distribution network.

[0110] The second virtual load of the main grid represents the tie line power transmitted from the main grid to the target distribution network.

[0111] The main network optimization scheduling model can be expressed mathematically as follows:

[0112]

[0113] Where, F TG The main grid objective function includes the coal-fired unit power generation cost f in the following main grid objective function. g and the cost of electricity sold from the main grid to the distribution grid f sell ,λ d,k,t 、μ d,k,t are the linear multiplier and quadratic multiplier of the Lagrange penalty function for the d-th distribution network at time t during the k-th iteration, is the tie line power transmitted from the main grid to the distribution grid at time t in the k+1th iteration, is the required power of the tie line obtained by optimizing the distribution network at time t during the kth iteration.

[0114] Specifically, the main network optimization scheduling model is trained based on the main network objective function, penalty function multiplier, sample generator quantity and sample load quantity.

[0115] Among them, the main grid objective function is constructed based on the power generation cost of coal-fired units, the cost of electricity sales from the main grid to the distribution network, and the main grid constraints.

[0116] The main network objective function is expressed as follows using mathematical formula:

[0117]

[0118] Where, f g is the power generation cost of coal-fired units, f sell The cost of electricity sold from the main grid to the distribution grid and the cost of electricity purchased from the distribution grid to the main grid, a g 、b g 、c g are all power generation cost coefficients, c sell The electricity price of the interconnection line is is the active power output of coal-fired unit g during period t, is the tie line power transmitted from the main network to the dth distribution network, and G is the set of coal-fired units.

[0119] Among them, the main grid constraints include: linearized power flow constraints, linearized power balance constraints, coal-fired unit output constraints, unit output constraints and key section constraints.

[0120] The embodiment of the present application uses the AC power flow equation in the form of branch power as the main network power flow constraint. The main network power flow constraint is satisfied at time t as follows:

[0121]

[0122] Where, P ij,t is the active power of line ij at time t, Q ij,t is the reactive power of line ij at time t, g ij 、b ij are the real and imaginary parts of the elements in the node admittance matrix, θ ij,t is the phase angle of line ij at time t, ν i,t is the voltage amplitude of node i at time t, ν j,t is the voltage amplitude of node j at time t.

[0123] The main grid power flow model shown in formula (15) is a non-convex model and cannot be directly solved using the optimization solver provided in the embodiment of the present application. Therefore, it is linearized. First, formula (15) is rewritten as follows:

[0124]

[0125] make Then formula (16) can be rewritten as:

[0126]

[0127] Assume that the power system operates near the nominal state, that is, θ ij =θ ij0 +Δθ ij . ij and Q ij Perform a first-order Taylor expansion, ignoring Δ 2 And the higher order small quantities of the above terms, with P ij Take this as an example:

[0128]

[0129] Calculate the partial derivatives, P ij right The partial derivative of is:

[0130]

[0131] right The partial derivative of is:

[0132]

[0133] θ ij The partial derivative of is:

[0134]

[0135] Near the nominal voltage, v i ≈v j ≈1p.u, that is θ ij ≈0, then cosθ ij ≈1, sinθ ij ≈θ ij , the simplified partial derivative is:

[0136]

[0137] Similarly, we can get Q ij The linearized power flow constraint is finally obtained as follows:

[0138]

[0139] The embodiment of the present application uses interval numbers to describe the uncertainty of the renewable energy output between the main grid and the distribution network, as shown in formula (24):

[0140]

[0141] Where, represent the interval variables of centralized wind power and photovoltaic power output at time t, They represent the lower limits of wind power and photovoltaic output range at time t, They represent the upper limits of wind power and photovoltaic power output at time t respectively. The power balance constraint of the power system node expressed in the form of interval equation is:

[0142]

[0143] Where, P i,t is the injected power of node i during period t, is the active load of node i during period t, is the interconnection line power transmitted from the main grid to the distribution network during period t, which can be regarded as the virtual load of the main grid.

[0144] In interval optimization, to deal with uncertainty constraints, the interval possibility method is usually used to transform the uncertain interval constraints into deterministic constraints. That is, an appropriate possibility level is set so that the uncertainty constraints can be satisfied at this level. Through this transformation process, the original equality constraint (25) will be transformed into two inequality constraints:

[0145]

[0146] make represents the generator output and wind and solar output of node i at time t, Represents the power required by node i at time t, and transforms Equation (26) into the form shown in Equation (27):

[0147]

[0148] Where λ is the probability level of the uncertainty constraint, ranging from [0 to 1]. When λ is close to 1, the power system satisfies the constraint with a higher probability, which enhances the robustness of the power system but also increases the economic cost. Conversely, when λ is close to 0, the power system prioritizes economic efficiency but is subject to a higher risk of default. The corresponding value range is:

[0149]

[0150] In formula (28), for The lower limit of the value range of The upper limit of the value range of , formula (28) can be simplified as:

[0151]

[0152] According to the above formula, we can get After that, we need to solve and With interval number H l Taking the real number k as an example, as shown in Equation (30), Equation (27) containing the uncertain constraint can be transformed into the deterministic constraint (i.e., the linearized power balance constraint) as shown in Equation (30).

[0153]

[0154] The output constraint of coal-fired units can be expressed by the following mathematical formula:

[0155]

[0156] Where, and are the upper and lower limits of the output of generator set g respectively.

[0157] The unit output constraint can be expressed by the mathematical formula as follows:

[0158]

[0159] Where, and are the rising ramp rate and falling ramp rate of coal-fired unit g respectively.

[0160] The key section constraint can be expressed mathematically as follows:

[0161]

[0162] Where C is the critical section line set, is the upper limit of the line flow allowed at the key section C, P l is the power of line l at the critical section C.

[0163] In order to ensure the consistency of the coupling variables at the main distribution network boundary, the consistency constraint of the main distribution network boundary is established:

[0164]

[0165] Where, is the power that the distribution network needs to inject from the tie line during period t, i.e., the virtual generator.

[0166] This consistency constraint cannot be solved directly in the model, so a penalty term is introduced to relax it. In this embodiment of the application, the augmented Lagrangian method is used to construct a quadratic function penalty term:

[0167]

[0168] Where λ is the Lagrange multiplier vector, μ is the penalty coefficient vector, ξ is the constraint slack variable, i.e., the power deviation at the boundary of the main and distribution networks, and ⊙ is the Hadamard product. The penalty function is added to the optimization objective functions of the main and distribution networks, resulting in the objective function of the main network optimization dispatch model with the Lagrange penalty function added, as shown in Formula (13), and the objective function of the distribution network optimization dispatch model, as shown in Formula (1).

[0169] Step 14: Determine whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion. If not, execute step 15; if so, execute step 16.

[0170] In this step, the second virtual generator quantity of the target distribution network and the second virtual load quantity of the main grid are verified according to a preset cyclic criterion.

[0171] Among them, the cyclic criterion is used to evaluate whether the system has reached a stable state and ensure that the energy exchange between the target distribution network and the main grid is balanced. If the second virtual generator quantity of the target distribution network and the second virtual load quantity of the main grid do not meet the requirements of the cyclic criterion, it means that the power system has not yet achieved the expected stability and needs to continue iterative adjustment.

[0172] Specifically, the cyclic criterion provided in the embodiment of the present application is expressed as follows using a mathematical formula:

[0173]

[0174] Where, and are the total cost of the main network and the total cost of the distribution network during the kth iteration, ε1 and ε2 are the tie line power convergence accuracy requirements and cost iteration accuracy requirements, respectively.

[0175] Specifically, the process of updating the penalty function multiplier in step 15 includes:

[0176] By formula Update the penalty function multiplier;

[0177] Among them, δ is a preset constant, λ d,t,k is the linear multiplier of the Lagrangian penalty function of the dth distribution network at time t in the kth iteration, μ d,k,t is the quadratic multiplier of the Lagrange penalty function of the dth distribution network at time t in the kth iteration, The sample load of the main network transmitted to the d-th distribution network in the k-th iteration, The sample generator quantity obtained from the d-th distribution network optimization in the k-th iteration.

[0178] Optionally, δ is a constant, generally taking a value of 2-3, and the initial values ​​of the multipliers λ and μ are generally relatively small constants (eg, 0.0001).

[0179] Step 15: If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network do not meet the cyclic criterion, update the number of iterations k=k+1 and the penalty function multiplier, use the second virtual generator quantity as the new first virtual generator quantity, use the second virtual load quantity as the new first virtual load quantity, and repeat step 12 until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion.

[0180] In this step, if the target distribution network's second virtual generator quantity and the main grid's second virtual load quantity do not meet the cyclic criterion, an iterative optimization process is initiated, the number of iterations k is updated, and the penalty function multiplier is adjusted to further optimize the energy distribution between the distribution network and the main grid. The updated second virtual generator quantity is used as the new first virtual generator quantity, and the second virtual load quantity is used as the new first virtual load quantity. These updated parameters are then re-entered into the distribution network optimization scheduling model for new calculations until the distribution network's second virtual generator quantity and the main grid's second virtual load quantity meet the cyclic criterion, indicating that the power system has reached a stable state and optimal economic performance.

[0181] Step 16: If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion, output the scheduling result.

[0182] In this step, after multiple iterations, when the second virtual generator quantity of the target distribution network and the second virtual load quantity of the main grid finally meet the cyclic criterion, the power system has reached a stable state and the optimization scheduling process is complete. At this point, the main grid optimization scheduling model and the distribution network optimization scheduling model will output the final scheduling results, which serve as the coordinated control instructions between the distribution network and the main grid, guiding the load distribution and energy scheduling of each distribution network and the main grid.

[0183] Among them, the dispatching results include the power of the main grid, the voltage of the main grid, the current of the main grid, the power of the distribution network, the voltage of the distribution network, the current of the distribution network, the virtual generator quantity, the virtual load quantity, etc.

[0184] The embodiment of the present application provides a scheduling method for a distributed main distribution network. The method first obtains a first virtual load amount, a first virtual generator amount, and a penalty function multiplier of a target distribution network in the kth iteration, wherein the target distribution network is any distribution network in the distributed main distribution network; k is a positive integer, and then the first virtual load amount, the first virtual generator amount, and the penalty function multiplier of the target distribution network are input into a distribution network optimization scheduling model for calculation to obtain a second virtual generator amount of the target distribution network, and then the second virtual generator amount, the penalty function multiplier, and the first virtual load amount of the target distribution network are input into the main network optimization scheduling model for calculation to obtain a second virtual load amount of the main network and transmit it to the target distribution network, and judge the second virtual load amount of the distribution network. The virtual generator quantity and the second virtual load quantity of the main network meet the cyclic criterion. If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network do not meet the cyclic criterion, the iteration number k=k+1 and the penalty function multiplier are updated, the second virtual generator quantity is used as the new first virtual generator quantity, the second virtual load quantity is used as the new first virtual load quantity, and the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier are repeatedly input into the distribution network optimization scheduling model for calculation until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion. If the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion, the scheduling result is output. This technical solution calculates the virtual load quantity and the virtual generator quantity, combines the penalty function multiplier for constrained optimization, and continuously iteratively optimizes the scheduling relationship between the distribution network and the main network, ensuring the rationality of the main distribution network scheduling strategy. The cyclic criterion is used to determine whether the convergence standard has been met, thereby improving the rationality and computational efficiency of the main distribution network scheduling strategy.

[0185] Based on the above embodiments, Figure 2 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 2 ,like Figure 2 As shown, step 12 may include the following steps:

[0186] Step 21: Standardize the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier respectively.

[0187] In this step, the input first virtual load quantity, first virtual generator quantity and penalty function multiplier are standardized so as to convert these parameters into data forms with the same dimension or numerical range, eliminate the scale differences between different parameters, and thus make the model processing more efficient and accurate.

[0188] For example, the first virtual load quantity may have a different numerical range from the first virtual generator quantity and the penalty function multiplier. Standardization processing can convert these parameters into a unified numerical range (for example, 0 to 1) so that the model can better process and compare these parameters, ensuring the stability and effectiveness of the subsequent optimization scheduling process and avoiding model calculation errors or instability caused by excessive data differences.

[0189] Step 22: Input the standardized first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity.

[0190] In this step, the standardized first virtual load quantity, the first virtual generator quantity and the penalty function multiplier will be passed as input parameters to the distribution network optimization scheduling model for calculation. The distribution network optimization scheduling model will use these input values ​​to perform load forecasting and power generation scheduling, and calculate the second virtual generator quantity of the target distribution network under current conditions through the optimization algorithm.

[0191] Among them, the second virtual generator quantity is based on the current status of the power system (such as load, power generation capacity, etc.) to further optimize the power generation demand of the target distribution network and calculate a more reasonable generator quantity.

[0192] The embodiment of the present application provides a method for dispatching a distributed main distribution network, which standardizes a first virtual load quantity, a first virtual generator quantity, and a penalty function multiplier. The standardized first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier are then input into a distribution network optimization dispatching model for calculation to obtain a second virtual generator quantity. This technical solution standardizes the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier. The standardized input data enables the distribution network optimization dispatching model to more efficiently handle power generation dispatching tasks, ensures that the calculation results of the distribution network optimization dispatching model are highly reasonable, and improves the overall operating efficiency of the power system and the stability of power supply.

[0193] In one possible implementation, Figure 3 Schematic diagram of the process of the distributed main distribution network scheduling method provided in the embodiment of the present application Figure 3 ,like Figure 3 As shown, this method has the following implementation methods:

[0194] Step 1. Get started.

[0195] Step 2: Initialize parameters.

[0196] In this step, the first virtual load amount, the first virtual generator amount and the penalty function multiplier are initialized. The first virtual load amount and the first virtual generator amount are set to 0, and the penalty function multiplier is set to 1.

[0197] Step 3: On the distribution network side, the distribution network optimization scheduling model is used to solve the calculations in parallel.

[0198] In this step, distribution network 1, distribution network 2, ..., distribution network n respectively use the distribution network optimization scheduling model to parallel calculate their respective optimization parameters (including the power, voltage, current and virtual load of each distribution network), and send the virtual load calculated by each distribution network to the main network.

[0199] Step 4: The main network side performs calculations based on the main network optimization scheduling model.

[0200] Step 5: Determine whether the cycle criterion is satisfied.

[0201] In this step, if the cycle criterion is satisfied, step 7 is executed; if the cycle criterion is not satisfied, step 6 is executed.

[0202] Step 6. Update the penalty function multiplier.

[0203] Step 7. End.

[0204] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0205] Figure 4 A structural diagram of a scheduling device for a distributed main distribution network provided in an embodiment of the present application is shown as follows: Figure 4 As shown, the device includes:

[0206] The acquisition module 41 is configured to acquire, in the kth iteration, a first virtual load quantity, a first virtual generator quantity, and a penalty function multiplier of a target distribution network; the target distribution network is any distribution network in the distributed main distribution network; and k is a positive integer;

[0207] a calculation module 42 for inputting the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into the distribution network optimization scheduling model for calculation to obtain a second virtual generator quantity of the target distribution network;

[0208] The calculation module 42 is further configured to input the second virtual generator quantity, the penalty function multiplier, and the first virtual load quantity of the target distribution network into the main network optimization scheduling model for calculation, obtain the second virtual load quantity of the main network, and transmit it to the target distribution network;

[0209] A judgment module 43 is used to judge whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion formula;

[0210] An updating module 44 is configured to update the number of iterations k=k+1 and the penalty function multiplier when the judgment result of the judgment module is non-compliance, and use the second virtual generator quantity as the new first virtual generator quantity and the second virtual load quantity as the new first virtual load quantity;

[0211] a triggering module 45 for executing the processing from the calculation module 42 to the judgment module 43 according to the result of the updating module until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion;

[0212] The output module 46 is used to output the scheduling result when the judgment result of the judgment module is in compliance.

[0213] In one possible implementation, the distribution network optimization scheduling model is trained based on the distribution network objective function, sample load, sample generator quantity and penalty function multiplier; wherein the distribution network objective function is constructed based on the distribution network loss cost, voltage deviation penalty cost, distribution network power purchase cost from the main grid and distribution network constraints.

[0214] In a possible implementation, the calculation module 42 is specifically configured to:

[0215] performing normalization processing on the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier respectively;

[0216] The standardized first virtual load quantity, the first virtual generator quantity and the penalty function multiplier are input into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity.

[0217] In one possible implementation, the main grid optimization scheduling model is trained based on the main grid objective function, penalty function multiplier, sample generator quantity and sample load quantity; wherein the main grid objective function is constructed based on the power generation cost of the coal-fired unit, the cost of selling electricity from the main grid to the distribution network and the main grid constraints.

[0218] In a possible implementation, the main grid constraints include: linearized power flow constraints, linearized power balance constraints, coal-fired unit output constraints, unit output constraints, and key section constraints.

[0219] In a possible implementation, the update module 44 is specifically configured to:

[0220] By formula Update the penalty function multiplier;

[0221] Among them, δ is a preset constant, λ d,t,k is the linear multiplier of the Lagrangian penalty function of the dth distribution network at time t in the kth iteration, μ d,k,t is the quadratic multiplier of the Lagrange penalty function of the dth distribution network at time t in the kth iteration, The sample load of the main network transmitted to the d-th distribution network in the k-th iteration, The sample generator quantity obtained from the d-th distribution network optimization in the k-th iteration.

[0222] The device provided in the embodiments of the present application can be used to execute the determination method in any of the above embodiments. Its implementation principles and technical effects are similar and will not be repeated here.

[0223] It should be noted that it should be understood that the division of the various modules of the above device is merely a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. Moreover, these modules can all be implemented in the form of software called by processing elements; they can also all be implemented in the form of hardware; some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. In addition, these modules can be fully or partially integrated together or implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit in the processor element or by instructions in the form of software.

[0224] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 5 As shown, the electronic device may include: a processor 51, a memory 52, and computer program instructions stored in the memory 52 and executable on the processor 51. When the processor 51 executes the computer program instructions, the method provided in any of the aforementioned embodiments is implemented.

[0225] Optionally, the above-mentioned components of the electronic device may be connected via a system bus.

[0226] The memory 52 may be a separate storage unit or a storage unit integrated in the processor 51. The number of the processor 51 may be one or more.

[0227] It should be understood that the processor 51 can be a central processing unit (CPU), or other general-purpose processors 51, a digital signal processor 51 (DSP), an application-specific integrated circuit (ASIC), etc. The general-purpose processor 51 can be a microprocessor 51 or any conventional processor 51. The steps of the method disclosed in this application can be directly implemented by the hardware processor 51 or performed by a combination of hardware and software modules in the processor 51.

[0228] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, among others. A system bus can be divided into an address bus, a data bus, a control bus, and so on. For ease of illustration, the figure uses only one thick line, but this does not imply that there is only one bus or only one type of bus. Memory 52 may include random access memory 52 (RAM) and may also include non-volatile memory 52 (NVM), such as at least one disk storage device 52.

[0229] All or part of the steps in the above-mentioned method embodiments can be implemented by hardware associated with program instructions. The aforementioned program can be stored in a readable memory 52. ​​When executed, the program performs the steps of the above-mentioned method embodiments; and the aforementioned memory 52 (storage medium) includes: read-only memory 52 (ROM), RAM, flash memory 52, hard disk, solid-state drive, magnetic tape, floppy disk, optical disc, and any combination thereof.

[0230] The electronic device provided in the embodiments of the present application can be used to execute the method provided in any of the above method embodiments. Its implementation principles and technical effects are similar and will not be repeated here.

[0231] An embodiment of the present application provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the computer executes the above method.

[0232] The computer-readable storage medium mentioned above may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0233] Optionally, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0234] An embodiment of the present application also provides a computer program product, which includes a computer program. The computer program is stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, the above method can be implemented.

[0235] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for scheduling a distributed main distribution network, characterized in that: The method comprises: Step 11: In the kth iteration, obtain the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier of the target distribution network; the target distribution network is any distribution network in the distributed main distribution network; k is a positive integer; Step 12: Input the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into a distribution network optimization scheduling model for calculation to obtain a second virtual generator quantity of the target distribution network; Step 13: Input the second virtual generator quantity of the target distribution network, the penalty function multiplier and the first virtual load quantity into the main network optimization scheduling model for calculation, obtain the second virtual load quantity of the main network and transmit it to the target distribution network; Step 14: determining whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet the cyclic criterion formula; Step 15: If not, update the number of iterations k=k+1 and the penalty function multiplier, use the second virtual generator quantity as the new first virtual generator quantity, use the second virtual load quantity as the new first virtual load quantity, and repeat steps 12 to 14 until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main grid meet the cyclic criterion formula; Step 16: If it meets the requirements, output the scheduling result.

2. The method according to claim 1, characterized in that The distribution network optimization scheduling model is obtained by training based on the distribution network objective function, sample load, sample generator quantity and the penalty function multiplier; wherein the distribution network objective function is constructed based on the distribution network loss cost, voltage deviation penalty cost, distribution network power purchase cost from the main grid and distribution network constraint conditions.

3. The method according to claim 1, characterized in that The step of inputting the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into a distribution network optimization scheduling model for calculation to obtain a second virtual generator quantity of the target distribution network includes: performing normalization processing on the first virtual load quantity, the first virtual generator quantity and the penalty function multiplier respectively; The standardized first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier are input into the distribution network optimization scheduling model for calculation to obtain the second virtual generator quantity.

4. The method according to claim 1, wherein The main grid optimization scheduling model is trained based on the main grid objective function, the penalty function multiplier, sample generator quantity and sample load quantity; wherein, the main grid objective function is constructed based on the power generation cost of the coal-fired unit, the cost of selling electricity from the main grid to the distribution network and the main grid constraint conditions.

5. The method according to claim 4, characterized in that The main grid constraints include: linearized power flow constraints, linearized power balance constraints, coal-fired unit output constraints, unit output constraints and key section constraints.

6. The method according to claim 1, characterized in that The process of updating the penalty function multiplier includes: By formula Updating the penalty function multiplier; Among them, δ is a preset constant, λ d,t,k is the linear multiplier of the Lagrangian penalty function of the dth distribution network at time t in the kth iteration, μ d,k,t is the quadratic multiplier of the Lagrange penalty function of the dth distribution network at time t in the kth iteration, The sample load of the main network transmitted to the d-th distribution network in the k-th iteration, The sample generator quantity obtained from the d-th distribution network optimization in the k-th iteration.

7. A dispatching device for a distributed main distribution network, characterized in that: include: an acquisition module, configured to acquire, in a kth iteration, a first virtual load quantity, a first virtual generator quantity, and a penalty function multiplier of the target distribution network; The target distribution network is any distribution network in the distributed main distribution network; k is a positive integer; a calculation module, configured to input the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into a distribution network optimization scheduling model for calculation to obtain a second virtual generator quantity of the target distribution network; The calculation module is further configured to input the second virtual generator quantity of the target distribution network, the penalty function multiplier, and the first virtual load quantity into the main network optimization scheduling model for calculation, obtain the second virtual load quantity of the main network, and transmit it to the target distribution network; a judgment module, configured to judge whether the second virtual generator quantity of the distribution network and the second virtual load quantity of the main network meet a cyclic criterion; an updating module, configured to update the number of iterations k=k+1 and the penalty function multiplier when the judgment result of the judging module is non-compliance, and use the second virtual generator amount as the new first virtual generator amount and the second virtual load amount as the new first virtual load amount; a triggering module, configured to execute, upon triggering of the updating module, the processing of inputting the first virtual load quantity, the first virtual generator quantity, and the penalty function multiplier into the distribution network optimization scheduling model for calculation, until the second virtual generator quantity of the distribution network and the second virtual load quantity of the main grid meet the cyclic criterion formula; The output module is used to output the scheduling result when the judgment result of the judgment module is in compliance.

8. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program, characterized in that The computer program includes a computer program, which is stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, the method described in any one of claims 1 to 6 can be implemented.

Images