Micro-grid scheduling method, device and equipment based on edge computing, and storage medium

Abstract

The application discloses a micro-grid scheduling method and device based on edge computing, equipment and storage medium, the method comprises the following steps: constructing an information-energy double-layer optimization structure model of a micro-grid; the information-energy double-layer optimization structure model comprises an upper information network and a lower energy network; obtaining the power generation cost and operation cost of each distributed power supply, constructing a micro-grid grid-connected cost model of the distributed power supply; according to the preset constraint condition, the micro-grid grid-connected cost model is optimized; according to the optimized micro-grid grid-connected cost model and the information-energy double-layer optimization structure model, a parallel constraint optimization iteration algorithm is constructed, and according to the preset minimum operation cost of the micro-grid, the parallel constraint optimization iteration algorithm is iteratively calculated, so that the corresponding power generation power of each distributed power supply is obtained as the scheduling operation scheme of the micro-grid; the application solves the technical problems of low efficiency and accuracy of micro-grid scheduling in the prior art.

Description

Technical Field

[0001] This invention relates to the field of power grid technology, and in particular to a microgrid scheduling method, apparatus, equipment and storage medium based on edge computing. Background Technology

[0002] New energy sources offer excellent sustainability and environmental friendliness. Using wind and solar power is an effective way to reduce carbon emissions, making new energy power generation technology a focus of attention for many countries. To increase the penetration rate of renewable energy, large-scale combinations of renewable energy generation to form microgrids (MGs) are employed, and these microgrids are controlled and dispatched by a control center. This significantly reduces the impact of renewable energy volatility on the main power grid and improves power supply reliability. The combination of a large power grid and appropriate control methods is considered by many experts and scholars both domestically and internationally to be a primary way to reduce energy consumption and improve the reliability and flexibility of power systems.

[0003] Existing MG control schemes mainly fall into three categories: hierarchical control, centralized control, and distributed or decentralized control. Hierarchical control includes primary control, secondary control, and sometimes tertiary control. Distributed generation (DG) is regulated by primary control, while deviations introduced by primary control are eliminated by secondary control. However, hierarchical control cannot be applied to complex power grid systems with multiple energy sources operating in parallel, and it cannot ensure the efficiency of power grid dispatch information transmission. Distributed or decentralized control, on the other hand, uses only local information. Even if multiple edge nodes fail, the system can still operate. To reduce communication and computational complexity, distributed or decentralized control is more suitable for complex systems. However, it cannot control deviations caused by regulation, resulting in low dispatch control accuracy.

[0004] Therefore, there is an urgent need for a method that can improve the efficiency and accuracy of microgrid dispatching. Summary of the Invention

[0005] This invention provides a microgrid scheduling method, apparatus, device, and storage medium based on edge computing to solve the technical problems of low efficiency and accuracy in microgrid scheduling in the prior art.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a microgrid scheduling method based on edge computing, comprising:

[0007] A dual-layer optimization structure model for information and energy in a microgrid is constructed. The dual-layer optimization structure model includes an upper-layer information network and a lower-layer energy network. The upper-layer information network includes edge-sensing computing nodes corresponding to each distributed power source, and the lower-layer energy network includes each distributed power source and load.

[0008] Obtain the generation cost and operating cost of each distributed power source, and construct a microgrid grid connection cost model for distributed power sources;

[0009] The microgrid grid connection cost model is optimized based on preset constraints.

[0010] Based on the optimized microgrid grid connection cost model and the information and energy dual-layer optimization structure model, a parallel constraint optimization iterative algorithm is constructed. The algorithm is iteratively calculated according to the preset minimum operating cost of the microgrid to obtain the power generation of each distributed power source, which serves as the scheduling and operation scheme for the microgrid.

[0011] As a preferred embodiment, the step of obtaining the generation cost and operating cost of each distributed power source and constructing a microgrid grid connection cost model for the distributed power sources specifically involves:

[0012] The grid-connected generation cost, environmental governance cost, and equipment depreciation and loss cost of each distributed power source in the microgrid are obtained as the generation cost and operating cost.

[0013] The grid-connected generation cost of each distributed power source's microgrid is: C i (P cg,i (t))=α i P cg,i (t) 2 +β i P cg,i (t)+γ i ;P cg,i (t) represents the power generation of the i-th controllable distributed power source at time t, α i β i and γ i These are the primary, secondary, and constant term coefficients of the generation cost of the i-th controllable distributed power source, respectively, C. i (P cg,i (t) is about P cg,i A function of (t);

[0014] The cost of environmental remediation is: e CO e HC e PM , and The treatment cost coefficients for carbon monoxide, hydrocarbons, soot, carbon dioxide, and nitrogen oxides are respectively given by e. gas E represents the total cost coefficient for the treatment of various pollutants. i (P cg,i (t) is about P cg,i A function of (t);

[0015] The equipment depreciation and loss cost is: k iD is the depreciation and wear cost factor for controllable distributed power sources. i (P cg,i (t) is about P cg,i A function of (t);

[0016] Based on the grid-connected power generation cost, environmental governance cost, and equipment depreciation cost of the microgrid, as well as the power generation capacity of each distributed power source, a microgrid grid-connected cost model for distributed power sources is constructed.

[0017] Among them, the microgrid grid connection cost model for distributed generation is: n cg P represents the number of distributed controllable power sources. cg,i (t) represents the power output of the i-th controllable distributed power source at time t, and C i (P cg,i (t) represents the generation cost of the i-th controllable distributed power source, E i (P cg,i (t) represents the environmental remediation cost of gas pollution caused by the i-th controllable distributed power source, D. i (P cg,i (t) is the depreciation and wear cost of the i-th controllable distributed power source.

[0018] As a preferred embodiment, the optimization of the microgrid grid connection cost model based on preset constraints specifically involves:

[0019] Based on the first preset constraint on the grid-connected power generation cost of the microgrid, the second constraint on the environmental governance cost, and the third constraint on the equipment depreciation and loss cost, the microgrid grid-connected cost model is constrained, thereby completing the optimization of the microgrid grid-connected cost model;

[0020] The constraint conditions are as follows: P PCC (t) represents the switching power of the common coupling node in the microgrid at time t. cg,i,min and P cg,i,max dP represents the lower and upper limits of the power generation of the i-th controllable distributed power source, respectively. cg,i,down and dP cg,i,up These are the maximum downhill rate and maximum uphill rate of the i-th controllable distributed power source, respectively.

[0021] As a preferred embodiment, the parallel constrained optimization iterative algorithm is constructed based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, specifically as follows:

[0022] Based on the edge-sensing computing nodes of the information-energy dual-layer optimization structure model and the power generation of each distributed power source in the optimized microgrid grid-connected cost model, a parallel constrained optimization iterative algorithm is constructed; wherein, the parallel constrained optimization iterative algorithm is:

[0023]

[0024]

[0025]

[0026] Where argmin represents the value of the variable that minimizes the objective function argmin(.), C i (P cg,i (t) represents the grid-connected generation cost of the microgrid, E i (P cg,i (t) represents the environmental governance cost, D i (P cg,i (t) represents the equipment depreciation and loss cost, P cg,i (t) represents the power output of the i-th controllable distributed power source at time t. This represents the intermediate solution for the controllable distributed power output of the i-th controllable edge node after the k-th iteration. Let ρ be the local Lagrange multiplier of the i-th controllable edge node before the k-th iteration, where ρ is the penalty factor and ρ > 0. 1×n Represents a 1-dimensional n-column vector consisting entirely of 1s. Let n be the intermediate solution of the controllable distributed power output received by the i-th controllable edge node before the (k+1)-th iteration from its neighboring edge nodes. cg P represents the number of distributed controllable power sources. BESS (t) represents the output power of the energy storage system in the microgrid at time t. These are the maximum likelihood estimates of the local Lagrange multipliers for the (i-1), ith, and (i+1)th controllable edge nodes after the kth iteration, respectively. This represents the maximum likelihood estimate of the intermediate solution after the i-th controllable edge node updates the output of the locally controllable distributed power source in the k-th iteration. Let be the local Lagrange multiplier of the i-th controllable edge node after the k-th iteration.

[0027] As a preferred embodiment, the parallel constraint optimization iterative algorithm is iteratively calculated based on the preset minimum operating cost of the microgrid to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme for the microgrid. Specifically:

[0028] Based on the parallel constraint optimization iterative algorithm, the intermediate solutions and node information received by each edge sensing computing node from its two adjacent edge sensing computing nodes, as well as the communication distance between each edge sensing computing node, are calculated to obtain the iteration results of each edge sensing computing node in two adjacent iterations; where each edge sensing computing node corresponds to a distributed power source.

[0029] With the goal of minimizing the operating cost of the microgrid, the power generation of each distributed power source is calculated based on the results of two adjacent iterations. When the deviation between the results of two adjacent iterations is less than a preset value, the calculated power generation of each distributed power source is used as the microgrid's scheduling and operation scheme.

[0030] As a preferred solution, when the number of edge nodes in the upper-layer information network is n, before the k-th iteration, the intermediate solutions received by the i-th edge node from the two adjacent edge-aware computing nodes are: s i (j) represents the communication distance from the j-th edge node to the i-th edge node in a ring information network composed of controllable distributed power sources, s i (j) = min[|ij|, (n+ij)];

[0031] Specifically, before the k-th iteration, the node information received by the i-th edge node from its two adjacent edge-aware computing nodes is as follows:

[0032] The goal is to achieve the minimum operating cost of the microgrid. Based on the results of two consecutive iterations, the power generation of each distributed power source is calculated, specifically as follows:

[0033]

[0034]

[0035] Among them, P i (t) represents the current generating power of the i-th distributed power source, P. i,min Let P be the minimum power output of the i-th distributed power source. i,max Let P be the maximum power output of the i-th distributed power source. i,maxdown P represents the maximum value of the sudden decrease in the power generation of the i-th distributed power source. i,maxup This represents the maximum value of the sudden increase in the power generation of the current i-th distributed power source;

[0036] When all and If the deviation is less than 0.1%, the iterative calculation ends, and the power generation of each distributed power source is used as the scheduling and operation scheme of the microgrid.

[0037] As a preferred embodiment, the distributed power source includes photovoltaics, wind turbines, gas turbines, diesel generators, and energy storage networks.

[0038] As a preferred embodiment, there is communication between the upper-layer information network and the lower-layer energy network regarding the output power of the distributed power source and the demand data between the loads.

[0039] Accordingly, the present invention also provides a microgrid scheduling device based on edge computing, comprising: a structural model construction module, a cost model construction module, a model optimization module, and an iterative calculation module;

[0040] The structural model construction module is used to construct a dual-layer optimized structure model of information and energy for microgrids. The dual-layer optimized structure model of information and energy includes an upper-layer information network and a lower-layer energy network. The upper-layer information network includes edge-sensing computing nodes corresponding to each distributed power source, and the lower-layer energy network includes each distributed power source and load.

[0041] The cost model construction module is used to obtain the generation cost and operating cost of each distributed power source and construct a microgrid grid connection cost model for the distributed power sources.

[0042] The model optimization module is used to optimize the microgrid grid connection cost model according to preset constraints.

[0043] The iterative calculation module is used to construct a parallel constraint optimization iterative algorithm based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, and to perform iterative calculations on the parallel constraint optimization iterative algorithm according to the preset minimum operating cost of the microgrid, so as to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme of the microgrid.

[0044] As a preferred embodiment, the step of obtaining the generation cost and operating cost of each distributed power source and constructing a microgrid grid connection cost model for the distributed power sources specifically involves:

[0045] The grid-connected generation cost, environmental governance cost, and equipment depreciation and loss cost of each distributed power source in the microgrid are obtained as the generation cost and operating cost.

[0046] The grid-connected generation cost of each distributed power source's microgrid is: C i (P cg,i (t))=α i P cg,i (t) 2 +β i P cg,i(t)+γ i ;P cg,i (t) represents the power generation of the i-th controllable distributed power source at time t, α i β i and γ i These are the primary, secondary, and constant term coefficients of the generation cost of the i-th controllable distributed power source, respectively, C. i (P cg,i (t) is about P cg,i A function of (t);

[0047] The cost of environmental remediation is: e CO e HC e PM , and The treatment cost coefficients for carbon monoxide, hydrocarbons, soot, carbon dioxide, and nitrogen oxides are respectively given by e. gas E represents the total cost coefficient for the treatment of various pollutants. i (P cg,i (t) is about P cg,i A function of (t);

[0048] The equipment depreciation and loss cost is: k i D is the depreciation and wear cost factor for controllable distributed power sources. i (P cg,i (t) is about P cg,i A function of (t);

[0049] Based on the grid-connected power generation cost, environmental governance cost, and equipment depreciation cost of the microgrid, as well as the power generation capacity of each distributed power source, a microgrid grid-connected cost model for distributed power sources is constructed.

[0050] Among them, the microgrid grid connection cost model for distributed generation is: n cg P represents the number of distributed controllable power sources. cg,i (t) represents the power output of the i-th controllable distributed power source at time t, and C i (P cg,i (t) represents the generation cost of the i-th controllable distributed power source, E i (P cg,i (t) represents the environmental remediation cost of gas pollution caused by the i-th controllable distributed power source, D. i (P cg,i (t) is the depreciation and wear cost of the i-th controllable distributed power source.

[0051] As a preferred embodiment, the optimization of the microgrid grid connection cost model based on preset constraints specifically involves:

[0052] Based on the first preset constraint on the grid-connected power generation cost of the microgrid, the second constraint on the environmental governance cost, and the third constraint on the equipment depreciation and loss cost, the microgrid grid-connected cost model is constrained, thereby completing the optimization of the microgrid grid-connected cost model;

[0053] The constraint conditions are as follows: P PCC (t) represents the switching power of the common coupling node in the microgrid at time t. cg,i,min and P cg,i,max dP represents the lower and upper limits of the power generation of the i-th controllable distributed power source, respectively. cg,i,down and dP cg,i,up These are the maximum downhill rate and maximum uphill rate of the i-th controllable distributed power source, respectively.

[0054] As a preferred embodiment, the parallel constrained optimization iterative algorithm is constructed based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, specifically as follows:

[0055] Based on the edge-sensing computing nodes of the information-energy dual-layer optimization structure model and the power generation of each distributed power source in the optimized microgrid grid-connected cost model, a parallel constrained optimization iterative algorithm is constructed; wherein, the parallel constrained optimization iterative algorithm is:

[0056]

[0057]

[0058]

[0059] Where argmin represents the value of the variable that minimizes the objective function argmin(.), C i (P cg,i (t) represents the grid-connected generation cost of the microgrid, E i (P cg,i (t) represents the environmental governance cost, D i (P cg,i (t) represents the equipment depreciation and loss cost, P cg,i (t) represents the power output of the i-th controllable distributed power source at time t. This represents the intermediate solution for the controllable distributed power output of the i-th controllable edge node after the k-th iteration. Let ρ be the local Lagrange multiplier of the i-th controllable edge node before the k-th iteration, where ρ is the penalty factor and ρ > 0. 1×nRepresents a 1-dimensional n-column vector consisting entirely of 1s. Let n be the intermediate solution of the controllable distributed power output received by the i-th controllable edge node before the (k+1)-th iteration from its neighboring edge nodes. cg P represents the number of distributed controllable power sources. BESS (t) represents the output power of the energy storage system in the microgrid at time t. These are the maximum likelihood estimates of the local Lagrange multipliers for the (i-1), ith, and (i+1)th controllable edge nodes after the kth iteration, respectively. This represents the maximum likelihood estimate of the intermediate solution after the i-th controllable edge node updates the output of the locally controllable distributed power source in the k-th iteration. Let be the local Lagrange multiplier of the i-th controllable edge node after the k-th iteration.

[0060] As a preferred embodiment, the parallel constraint optimization iterative algorithm is iteratively calculated based on the preset minimum operating cost of the microgrid to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme for the microgrid. Specifically:

[0061] Based on the parallel constraint optimization iterative algorithm, the intermediate solutions and node information received by each edge sensing computing node from its two adjacent edge sensing computing nodes, as well as the communication distance between each edge sensing computing node, are calculated to obtain the iteration results of each edge sensing computing node in two adjacent iterations; where each edge sensing computing node corresponds to a distributed power source.

[0062] With the goal of minimizing the operating cost of the microgrid, the power generation of each distributed power source is calculated based on the results of two adjacent iterations. When the deviation between the results of two adjacent iterations is less than a preset value, the calculated power generation of each distributed power source is used as the microgrid's scheduling and operation scheme.

[0063] As a preferred solution, when the number of edge nodes in the upper-layer information network is n, before the k-th iteration, the intermediate solutions received by the i-th edge node from the two adjacent edge-aware computing nodes are: s i (j) represents the communication distance from the j-th edge node to the i-th edge node in a ring information network composed of controllable distributed power sources, s i (j) = min[|ij|, (n+ij)];

[0064] Specifically, before the k-th iteration, the node information received by the i-th edge node from its two adjacent edge-aware computing nodes is as follows:

[0065] The goal is to achieve the minimum operating cost of the microgrid. Based on the results of two consecutive iterations, the power generation of each distributed power source is calculated, specifically as follows:

[0066]

[0067]

[0068] Among them, P i (t) represents the current generating power of the i-th distributed power source, P. i,min Let P be the minimum power output of the i-th distributed power source. i,max Let P be the maximum power output of the i-th distributed power source. i,maxdown P represents the maximum value of the sudden decrease in the power generation of the i-th distributed power source. i,maxup This represents the maximum value of the sudden increase in the power generation of the current i-th distributed power source;

[0069] When all and If the deviation is less than 0.1%, the iterative calculation ends, and the power generation of each distributed power source is used as the scheduling and operation scheme of the microgrid.

[0070] As a preferred embodiment, the distributed power source includes photovoltaics, wind turbines, gas turbines, diesel generators, and energy storage networks.

[0071] As a preferred embodiment, there is communication between the upper-layer information network and the lower-layer energy network regarding the output power of the distributed power source and the demand data between the loads.

[0072] Accordingly, the present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the microgrid scheduling method based on edge computing as described in any of the preceding claims.

[0073] Accordingly, the present invention also provides a computer-readable storage medium comprising a stored computer program; wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the microgrid scheduling method based on edge computing as described in any of the preceding claims.

[0074] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0075] The technical solution of this invention constructs a two-layer optimization structure model of information and energy for microgrids and a microgrid grid-connected cost model. It optimizes the microgrid grid-connected cost model using preset constraints. After constructing a parallel constraint optimization iterative algorithm, the optimization problem for all edge-aware computing nodes is iteratively solved, thereby obtaining the power generation corresponding to each distributed power source. This serves as the microgrid's scheduling and operation scheme, gradually leading to the global optimal solution for the microgrid cost-economic optimization problem. This invention achieves faster convergence of the constraint optimization algorithm than the standard constraint optimization algorithm without reducing optimization accuracy, ensuring high efficiency and high accuracy in microgrid scheduling. Attached Figure Description

[0076] Figure 1 : A flowchart illustrating the steps of a microgrid scheduling method based on edge computing provided in an embodiment of the present invention;

[0077] Figure 2 : A schematic diagram of the information energy dual-layer optimization structure model provided in the embodiment of the present invention;

[0078] Figure 3 : A schematic diagram of the microgrid simulation model provided in the embodiments of the present invention;

[0079] Figure 4 : This is a schematic diagram of the microgrid simulation model environment settings provided in the embodiments of the present invention;

[0080] Figure 5 : This is a controllable distributed power supply output curve provided in an embodiment of the present invention;

[0081] Figure 6 : A voltage and frequency curve diagram of a microgrid provided in an embodiment of the present invention;

[0082] Figure 7 : This is a power output curve diagram of the energy storage system provided in an embodiment of the present invention;

[0083] Figure 8 : This is a graph showing the incremental rate of power consumption of a controllable distributed power source provided in an embodiment of the present invention;

[0084] Figure 9 : This is a schematic diagram of a microgrid scheduling device based on edge computing provided in an embodiment of the present invention. Detailed Implementation

[0085] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0086] Example 1

[0087] Please refer to Figure 1 The microgrid scheduling method based on edge computing provided in this embodiment of the invention includes the following steps S101-S104:

[0088] Step S101: Construct a dual-layer optimization structure model for information and energy in a microgrid; the dual-layer optimization structure model includes an upper-layer information network and a lower-layer energy network; the upper-layer information network includes edge-sensing computing nodes corresponding to each distributed power source, and the lower-layer energy network includes each distributed power source and load.

[0089] As a preferred embodiment, the distributed power source includes photovoltaics, wind turbines, gas turbines, diesel generators, and energy storage networks.

[0090] In a preferred embodiment, there is communication between the upper-layer information network and the lower-layer energy network regarding the output power of the distributed power source and the demand data between the loads.

[0091] In this embodiment, a two-layer optimized structure model for information and energy of the microgrid is constructed. Please refer to [link / reference]. Figure 2 Edge-sensing computing nodes form the upper-layer information network, while the distributed power sources and load connections of the microgrid constitute the lower-layer energy network. The upper-layer information network contains edge-sensing computing nodes corresponding to various types of distributed power sources, mainly for sensing, calculating, analyzing, and controlling the output of local distributed power sources. Distributed power sources include controllable distributed power sources and semi-controllable distributed power sources. Uncontrollable distributed power sources (such as photovoltaics and wind turbines) operate in maximum power point tracking mode, while the edge nodes corresponding to controllable distributed power sources (micro gas turbines and small diesel generators) are connected in a ring in sequence.

[0092] Furthermore, the upper-layer information network senses the power output of the common coupling nodes in the microgrid in real time, obtains the power fluctuation of the microgrid at the current moment, and transmits this information to the edge sensing computing nodes corresponding to the adjacent controllable distributed power sources.

[0093] In this embodiment, leveraging multi-edge node system theory, a two-layer optimization structure model for microgrid information and energy based on edge nodes (edge-sensing computing nodes) is constructed. This model includes an upper-layer communication network composed of intelligent agents and a lower-layer microgrid connected by electrical equipment. The lower-layer microgrid consists of distributed power sources and loads. Distributed power sources include photovoltaics, wind turbines, micro gas turbines, small diesel generators, and energy storage systems. These distributed power sources can be further classified into controllable distributed power sources, semi-controllable distributed power sources, and uncontrollable distributed power sources based on their control methods. The upper-layer communication network contains controllable edge nodes and semi-controllable edge nodes corresponding to the controllable and semi-controllable distributed power sources. Figure 1 As shown in the diagram, controllable and semi-controllable edge nodes in the upper-layer communication network are represented by diamonds and ellipses, respectively. Directed arrows indicate the communication links between edge nodes and their directions. Edge nodes read the output and load requirements of the corresponding underlying distributed power source through the dashed bidirectional arrows between the two network layers, perform calculations, and then control the output of the corresponding distributed power source.

[0094] To maximize the absorption of renewable energy and reduce the carbon emissions of microgrids, distributed power sources driven by renewable energy, such as photovoltaic and wind turbines, operate in maximum power point tracking (MPPT) mode. In this mode, their power output cannot be manually adjusted, hence they are called uncontrollable distributed power sources. Micro gas turbines and small diesel generators, on the other hand, can have their power output manually adjusted as needed, and are called controllable distributed power sources. Their corresponding controllable edge nodes in the upper-level network are represented by diamond-shaped boxes. When a microgrid operates in islanded mode, it requires energy storage systems to provide voltage and frequency support, allowing the energy storage systems to operate in V / F control mode. Since their output power can only change according to the microgrid's operating state, they are called semi-controllable distributed power sources, and their corresponding semi-controllable edge nodes in the upper-level network are represented by elliptical boxes.

[0095] In a communication network, a semi-controllable edge node transmits unprocessed information read from the underlying microgrid to adjacent controllable edge nodes, but does not receive information from other controllable edge nodes; that is, a semi-controllable edge node only has outgoing edges and no incoming edges. A controllable edge node, after processing the information, exchanges information with other controllable edge nodes through bidirectional communication links; therefore, it has both outgoing and incoming edges. Furthermore, controllable edge nodes can also collect their own information through self-loops. Figure 1 The upper-layer communication network shown has controllable edge nodes numbered sequentially and connected in a ring to ensure that the results converge to the global optimum when solving the optimization problem in a distributed manner. The semi-controllable edge node connected to the energy storage system in the lower-layer electrical network by dashed lines is connected to the controllable edge node numbered 1 in the ring communication network composed of controllable edge nodes by a unidirectional solid arrow.

[0096] Step S102: Obtain the generation cost and operating cost of each distributed power source, and construct a microgrid grid connection cost model for distributed power sources.

[0097] As a preferred embodiment, the step of obtaining the generation cost and operating cost of each distributed power source and constructing a microgrid grid connection cost model for the distributed power sources specifically involves:

[0098] Obtain the grid-connected generation cost, environmental governance cost, and equipment depreciation cost of each distributed power source in the microgrid, as the generation cost and operating cost; based on the grid-connected generation cost, environmental governance cost, and equipment depreciation cost of the microgrid, as well as the power generation capacity of each distributed power source, construct a microgrid grid-connected cost model for the distributed power sources.

[0099] In this embodiment, the microgrid grid connection cost model for distributed power sources includes the controllable distributed power generation cost, environmental governance cost, and equipment depreciation and loss cost. The controllable distributed power generation cost is a quadratic function, as shown in the following equation:

[0100] C i (P cg,i (t))=α i P cg,i (t) 2 +β i P cg,i (t)+γ i (1)

[0101] In equation (1) above, P cg,i (t) represents the power generation of the i-th controllable distributed power source at time t, α i β i and γ i These are the primary, secondary, and constant term coefficients of the generation cost of the i-th controllable distributed power source, respectively, C. i (P cg,i (t) is about P cg,i A function of (t); α is determined by market energy prices. i The value is generally taken as 0.4 to 0.7, β i The value is generally taken as 0.4–0.7, and γ i Generally, a value of 40 to 70 is used.

[0102] For example, the main driving energy source for controllable distributed power sources is fossil fuels such as natural gas and diesel, which produce a series of polluting gases when burned, as shown in the following formula:

[0103]

[0104] In equation (2) above, e CO e HCe PM , and The treatment cost coefficients for carbon monoxide, hydrocarbons, soot, carbon dioxide, and nitrogen oxides are respectively given by e. gas E represents the total cost coefficient for the treatment of various pollutants. i (P cg,i (t) is about P cg,i A function of (t).

[0105] For example, the depreciation losses of distributed power sources during use:

[0106]

[0107] In equation (3) above, k i D is the depreciation and wear cost factor for controllable distributed power sources. i (P cg,i (t) is about P cg,i A function of (t).

[0108] Step S103: Optimize the microgrid grid connection cost model according to preset constraints.

[0109] As a preferred embodiment, the optimization of the microgrid grid connection cost model based on preset constraints specifically involves:

[0110] Based on the first preset constraint on the grid-connected power generation cost of the microgrid, the second constraint on the environmental governance cost, and the third constraint on the equipment depreciation and loss cost, the microgrid grid-connected cost model is constrained, thereby optimizing the microgrid grid-connected cost model.

[0111] In this embodiment, the optimized microgrid grid connection cost model is as follows:

[0112]

[0113] Constraints

[0114] In equation (4) above, n cg P represents the number of distributed controllable power sources. cg,i (t) represents the power output of the i-th controllable distributed power source at time t, and C i (P cg,i (t) represents the generation cost of the i-th controllable distributed power source, E i (P cg,i (t) represents the environmental remediation cost of gas pollution caused by the i-th controllable distributed power source, D. i (P cg,i(t) represents the depreciation and wear cost of the i-th controllable distributed power source; P PCC (t) represents the switching power of the common coupling node in the microgrid at time t. cg,i,min and P cg,i,max dP represents the lower and upper limits of the power generation of the i-th controllable distributed power source, respectively. cg,i,down and dP cg,i,up These are the maximum downhill rate and maximum uphill rate of the i-th controllable distributed power source, respectively.

[0115] Step S104: Based on the optimized microgrid grid connection cost model and the information and energy dual-layer optimization structure model, a parallel constraint optimization iterative algorithm is constructed. The parallel constraint optimization iterative algorithm is iteratively calculated according to the preset minimum operating cost of the microgrid to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme of the microgrid.

[0116] In this embodiment, the constructed parallel constraint optimization iterative algorithm is as follows:

[0117]

[0118] In equation (5) above, argmin represents the value of the variable that minimizes the objective function argmin(.), and C i (P cg,i (t) represents the grid-connected generation cost of the microgrid, E i (P cg,i (t) represents the environmental governance cost, D i (P cg,i (t) represents the equipment depreciation and loss cost, P cg,i (t) represents the power output of the i-th controllable distributed power source at time t. This represents the intermediate solution for the controllable distributed power output of the i-th controllable edge node after the k-th iteration. Let ρ be the local Lagrange multiplier of the i-th controllable edge node before the k-th iteration, where ρ is the penalty factor and ρ > 0. 1×n Represents a 1-dimensional n-column vector consisting entirely of 1s. Let n be the intermediate solution of the controllable distributed power output received by the i-th controllable edge node before the (k+1)-th iteration from its neighboring edge nodes. cg P represents the number of distributed controllable power sources. BESS (t) represents the output power of the energy storage system in the microgrid at time t. These are the maximum likelihood estimates of the local Lagrange multipliers for the (i-1), ith, and (i+1)th controllable edge nodes after the kth iteration, respectively. This represents the maximum likelihood estimate of the intermediate solution after the i-th controllable edge node updates the output of the locally controllable distributed power source in the k-th iteration. Let be the local Lagrange multiplier of the i-th controllable edge node after the k-th iteration.

[0119] As a preferred embodiment, the step of constructing a parallel constrained optimization iterative algorithm based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model specifically involves:

[0120] Based on the edge-sensing computing nodes of the information-energy dual-layer optimization structure model and the power generation of each distributed power source in the optimized microgrid grid-connected cost model, a parallel constrained optimization iterative algorithm is constructed.

[0121] In a preferred embodiment, the parallel constraint optimization iterative algorithm is iteratively calculated based on the preset minimum operating cost of the microgrid to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme for the microgrid. Specifically:

[0122] Based on the parallel constrained optimization iterative algorithm, the intermediate solutions and node information received by each edge sensing computing node from its two adjacent edge sensing computing nodes, as well as the communication distance between each edge sensing computing node, are calculated to obtain the iteration results of each edge sensing computing node in two adjacent iterations. Each edge sensing computing node corresponds to a distributed power source. With the preset minimum operating cost of the microgrid as the objective, the power generation of each distributed power source is calculated based on the results of two adjacent iterations. When the deviation between the results of two adjacent iterations is less than a preset value, the calculated power generation of each distributed power source is used as the scheduling and operation scheme of the microgrid.

[0123] In this embodiment, when the number of edge nodes in the upper-layer information network is n, before the k-th iteration calculation, the intermediate solutions received by the i-th edge node from its left and right adjacent edge nodes are summarized by the following formula:

[0124]

[0125] In equation (6) above: s i (j) is the communication distance from the j-th edge node to the i-th edge node in a ring information network composed of controllable distributed power sources, calculated according to the following formula (7):

[0126] s i (j)=min[|ij|,(n+ij)] (7)

[0127] Equation (7) above represents the minimum value between the absolute value of (ij) and the value of (n+ij).

[0128] Furthermore, before the k-th iteration, the information received by the i-th edge node from its left and right adjacent edge nodes is as follows:

[0129]

[0130] Before the next iteration begins, each edge node (edge-aware computing node) uses the received information to complete each iteration according to the above equation (5). After each iteration, each controllable edge node further determines the final value of the intermediate solution of this iteration according to the upper and lower limit constraints shown in equation (4).

[0131] Furthermore, with the goal of minimizing operating costs, the operation scheme of the distributed power supply is determined. The i-th controllable edge node is determined based on the results of the current k and k-1 iterations. The power generation P of the current i-th controllable distributed power source is calculated using the following formula. i (t):

[0132]

[0133] In equation (9), P i (t) represents the current generating power of the i-th distributed power source, P. i,min Let P be the minimum power output of the i-th distributed power source. i,max Let P be the maximum power output of the i-th distributed power source. i,max down P represents the maximum value of the sudden decrease in the power generation of the i-th distributed power source. i,max up This represents the maximum value of the sudden increase in the power generation of the current i-th distributed power source. When all... and When the deviation is less than 0.1%, the iteration ends, thus obtaining the microgrid's scheduling and operation scheme.

[0134] In this embodiment, to verify the effectiveness of the invention, a radial microgrid simulation platform including 11 distributed generation (DG) sources, a common coupling node (PCC), and its loads was built in the Matlab / Simulink environment, as follows: Figure 3 As shown in Table 1, the generation cost coefficients of the six controllable distributed generation (DG) units are as follows.

[0135] Table 1. Controllable Distributed Generator (DG) Generation Cost Coefficients

[0136]

[0137] The capacity and operating mode of each distributed generation (DG) in the microgrid, as well as their corresponding maximum load demand, are shown in Table 2.

[0138] Table 2 Distributed Generation (DG) and Load Parameters

[0139]

[0140]

[0141] To test the performance of the proposed parallel constraint optimization algorithm, this example compares and tests the solution accuracy and convergence speed of standard serial constraint optimization and improved parallel constraint optimization. Four typical convex functions, as shown in Table 3, are selected as objective functions to compare and test the two algorithms.

[0142] Table 3 Test Function Table

[0143]

[0144] To eliminate the impact of randomness on the algorithm's performance, 100 sets of equality constraints with different coefficients were set for each test objective function, and the solutions were obtained using standard serial constraint optimization and parallel constraint optimization respectively. The relevant performance indicators were then averaged for comparison.

[0145] Using the interior-point approach (IPA) as the reference optimal solution, the mean absolute deviation (MAE) between the optimal solutions obtained by the standard constraint optimization and improved constraint optimization algorithms and the reference optimal solution is calculated to compare their solution accuracy. The formula for calculating MAE is shown below.

[0146]

[0147] In the above formula (10), Let i be the i-th component of the reference optimal solution obtained by the interior-point method. It is the i-th component of the optimal solution obtained by standard constraint optimization or improved constraint optimization.

[0148] The penalty factor for both the existing standard constraint optimization algorithm and the constraint optimization algorithm of this invention is set to 0.05, and the convergence accuracy is set to 0.01. After eliminating randomness under the 100 sets of constraints, the average absolute deviation (MAE) between the optimal solution obtained by solving the typical convex optimization problem shown in the table above using the standard constraint optimization algorithm and the improved constraint optimization algorithm, and the reference optimal solution of the interior point method, as well as the required number of iterations (NoI), are shown below:

[0149] Table 4 Comparison of Results between Serial Constraint Optimization and Parallel Constraint Optimization

[0150]

[0151] Table 4 shows that when solving the tested optimization problem using both the existing standard constraint optimization algorithm and the improved constraint optimization algorithm of this invention, the MAE value is very small, and the optimal solution obtained is basically consistent with the reference optimal solution obtained using the interior point method. Specifically, comparing the same test function longitudinally, as the value of n increases (i.e., the problem complexity increases), the number of iterations for both algorithms increases. Furthermore, comparing horizontally, when solving the same optimization problem using the standard constraint optimization algorithm and the improved constraint optimization algorithm respectively, the improved constraint optimization algorithm obtains a more accurate optimal solution with fewer iterations and a faster convergence speed. For the first type of function, the number of iterations for the improved constraint optimization algorithm is only one-third that of the standard constraint optimization algorithm; for the second type of function, the number of iterations is only one-half that of the standard constraint optimization algorithm; for the third type of function, the number of iterations for the improved constraint optimization algorithm is at least one-quarter that of the standard constraint optimization algorithm; and for the fourth type of function, the average number of iterations for the improved constraint optimization algorithm is at least one-tenth that of the standard constraint optimization algorithm.

[0152] During the simulation, the output curves of wind turbines and photovoltaics within the microgrid, as well as the total active and reactive power demands of the load within the microgrid, are shown as follows: Figure 4 As shown.

[0153] On the established microgrid simulation platform, the parallel constraint optimization solution formula shown in equation (5) is used to solve the minimum power generation cost optimization scheduling of the islanded microgrid. The output curves of each controllable distributed generation source (DG) during the scheduling period are as follows: Figure 5 As shown.

[0154] like Figure 5 As shown, there were no significant power fluctuations in the microgrid before t=2h, and the output of each controllable distributed generation (DG) did not change significantly. Furthermore, as shown in the figure, the consumption increment rate of each controllable distributed generation (DG) remained equal throughout, indicating that the output of the controllable distributed generation (DG) in the microgrid achieved economic optimization.

[0155] At t=2h, the load demand within the microgrid suddenly increases dramatically. At this time, if... Figure 6 As shown, the voltage and frequency of the microgrid are affected, resulting in significant fluctuations. For example... Figure 7 As shown, the energy storage system located at the balance node discharges promptly upon detecting voltage and frequency anomalies, thus ensuring the voltage and frequency stability of the microgrid and preventing it from exceeding safety limits. Subsequently, each controllable distributed generation (DG) increases its power output to share the discharge load of the energy storage system, such as... Figure 8 As shown, after increasing the power output, the consumption rate of each controllable distributed generation (DG) remains consistent, and the power generation cost within the microgrid achieves economic optimization.

[0156] At t=4h, the load demand within the microgrid suddenly decreases significantly. At this time, if... Figure 6 The voltage and frequency of the microgrid shown are affected and fluctuate significantly. For example... Figure 7 As shown, the energy storage system operating in V / F mode charges in a timely manner to absorb excess electricity within the microgrid, further stabilizing the microgrid's voltage and frequency. Subsequently, each controllable distributed generation (DG) reduces its power output to offset the charging input of the energy storage system, such as... Figure 8 As shown, after reducing the power output, the consumption increase rate of each controllable distributed generation (DG) remains consistent, and the power generation cost within the microgrid achieves economic optimization.

[0157] At other times, when the load demand within the microgrid suddenly decreases or increases, the energy storage system operating in V / F mode at the balance node can charge or discharge in a timely manner, thereby maintaining the stability of the microgrid voltage and frequency. Subsequently, each controllable distributed generation (DG) uses an improved constraint optimization algorithm to solve for the magnitude of its own output in a distributed manner, achieving economic optimization while ensuring that the microgrid has a large safety margin.

[0158] Implementing the above embodiments has the following effects:

[0159] The technical solution of this invention constructs a two-layer optimization structure model of information and energy for microgrids and a microgrid grid-connected cost model. It optimizes the microgrid grid-connected cost model using preset constraints. After constructing a parallel constraint optimization iterative algorithm, the optimization problem for all edge-aware computing nodes is iteratively solved, thereby obtaining the power generation corresponding to each distributed power source. This serves as the microgrid's scheduling and operation scheme, gradually leading to the global optimal solution for the microgrid cost-economic optimization problem. This invention achieves faster convergence of the constraint optimization algorithm than the standard constraint optimization algorithm without reducing optimization accuracy, ensuring high efficiency and high accuracy in microgrid scheduling.

[0160] Furthermore, in this invention, all edge nodes solve the optimization problem in parallel and send the obtained local intermediate solutions to adjacent edge nodes. Through repeated iterations, the global optimal solution to the optimization problem is gradually obtained. Without reducing the optimization accuracy, the improved constraint optimization algorithm converges faster than the standard constraint optimization algorithm. In addition, a microgrid optimal scheduling model is established, and this method is applied to solve the microgrid economic scheduling problem. Simulation results show that, for solving the same optimization problem, the number of iterations of the parallel constraint optimization algorithm is only half or even less than that of the standard constraint optimization algorithm. When solving the optimal scheduling problem to minimize the generation cost of the microgrid, the incremental costs of each controllable distributed generation (DG) tend to be consistent, indicating that this method can minimize the operating cost of the microgrid.

[0161] Example 2

[0162] Please see Figure 9 The present invention provides a microgrid scheduling device based on edge computing, comprising: a structural model construction module 201, a cost model construction module 202, a model optimization module 203, and an iterative calculation module 204.

[0163] The structural model construction module 201 is used to construct a two-layer optimized structure model of information and energy for a microgrid. The two-layer optimized structure model includes an upper-layer information network and a lower-layer energy network. The upper-layer information network includes edge-aware computing nodes corresponding to each distributed power source, and the lower-layer energy network includes each distributed power source and load.

[0164] The cost model construction module 202 is used to obtain the generation cost and operating cost of each distributed power source and construct a microgrid grid connection cost model for the distributed power sources.

[0165] The model optimization module 203 is used to optimize the microgrid grid connection cost model according to preset constraints.

[0166] The iterative calculation module 204 is used to construct a parallel constraint optimization iterative algorithm based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, and to perform iterative calculations on the parallel constraint optimization iterative algorithm based on the preset minimum operating cost of the microgrid, so as to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme of the microgrid.

[0167] As a preferred embodiment, the step of obtaining the generation cost and operating cost of each distributed power source and constructing a microgrid grid connection cost model for the distributed power sources specifically involves:

[0168] Obtain the grid-connected generation cost, environmental remediation cost, and equipment depreciation cost of each distributed power source in the microgrid, as the generation cost and operating cost; among which, the grid-connected generation cost of each distributed power source in the microgrid is: C i (P cg,i (t))=α i P cg,i (t) 2 +β i P cg,i (t)+γ i ;P cg,i (t) represents the power generation of the i-th controllable distributed power source at time t, α i β i and γ i These are the primary, secondary, and constant term coefficients of the generation cost of the i-th controllable distributed power source, respectively, C. i (P cg,i (t) is about P cg,i A function of (t); the cost of environmental remediation is: e CO e HC e PM , and The treatment cost coefficients for carbon monoxide, hydrocarbons, soot, carbon dioxide, and nitrogen oxides are respectively given by e. gas E represents the total cost coefficient for the treatment of various pollutants. i (P cg,i (t) is about P cg,i The function of (t); the equipment depreciation cost is: k i D is the depreciation and wear cost factor for controllable distributed power sources. i (P cg,i (t) is about P cg,i The function of (t); based on the microgrid grid-connected power generation cost, environmental governance cost, and equipment depreciation cost, as well as the power generation capacity of each distributed power source, a microgrid grid-connected cost model for distributed power sources is constructed; wherein, the microgrid grid-connected cost model for distributed power sources is... n cg P represents the number of distributed controllable power sources. cg,i (t) represents the power output of the i-th controllable distributed power source at time t, and C i (P cg,i (t) represents the generation cost of the i-th controllable distributed power source, E i (P cg,i (t) represents the environmental remediation cost of gas pollution caused by the i-th controllable distributed power source, D. i (p cg,i (t) is the depreciation and wear cost of the i-th controllable distributed power source.

[0169] As a preferred embodiment, the optimization of the microgrid grid connection cost model based on preset constraints specifically involves:

[0170] Based on the first preset constraint on the microgrid grid-connected power generation cost, the second constraint on the environmental governance cost, and the third constraint on the equipment depreciation and loss cost, the microgrid grid-connected cost model is constrained, thereby optimizing the microgrid grid-connected cost model; wherein, the constraints are as follows: P PCC (t) represents the switching power of the common coupling node in the microgrid at time t. cg,i,min and P cg,i,max dP represents the lower and upper limits of the power generation of the i-th controllable distributed power source, respectively. cg,i,down and dP cg,i,up These are the maximum downhill rate and maximum uphill rate of the i-th controllable distributed power source, respectively.

[0171] As a preferred embodiment, the parallel constrained optimization iterative algorithm is constructed based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, specifically as follows:

[0172] Based on the edge-sensing computing nodes of the information-energy dual-layer optimization structure model and the power generation of each distributed power source in the optimized microgrid grid-connected cost model, a parallel constrained optimization iterative algorithm is constructed; wherein, the parallel constrained optimization iterative algorithm is:

[0173]

[0174]

[0175]

[0176] Where argmin represents the value of the variable that minimizes the objective function argmin(.), C i (P cg,i (t) represents the grid-connected generation cost of the microgrid, E i (P cg,i (t) represents the environmental governance cost, D i (P cg,i (t) represents the equipment depreciation and loss cost, P cg,i (t) represents the power output of the i-th controllable distributed power source at time t. This represents the intermediate solution for the controllable distributed power output of the i-th controllable edge node after the k-th iteration. Let ρ be the local Lagrange multiplier of the i-th controllable edge node before the k-th iteration, where ρ is the penalty factor and ρ > 0. 1×n Represents a 1-dimensional n-column vector consisting entirely of 1s. Let n be the intermediate solution of the controllable distributed power output received by the i-th controllable edge node before the (k+1)-th iteration from its neighboring edge nodes. cg P represents the number of distributed controllable power sources. BESS (t) represents the output power of the energy storage system in the microgrid at time t. These are the maximum likelihood estimates of the local Lagrange multipliers for the (i-1), ith, and (i+1)th controllable edge nodes after the kth iteration, respectively. This represents the maximum likelihood estimate of the intermediate solution after the i-th controllable edge node updates the output of the locally controllable distributed power source in the k-th iteration. Let be the local Lagrange multiplier of the i-th controllable edge node after the k-th iteration.

[0177] As a preferred embodiment, the parallel constraint optimization iterative algorithm is iteratively calculated based on the preset minimum operating cost of the microgrid to obtain the power generation corresponding to each distributed power source, which serves as the scheduling and operation scheme for the microgrid. Specifically:

[0178] Based on the parallel constrained optimization iterative algorithm, the intermediate solutions and node information received by each edge sensing computing node from its two adjacent edge sensing computing nodes, as well as the communication distance between each edge sensing computing node, are calculated to obtain the iteration results of each edge sensing computing node in two adjacent iterations. Each edge sensing computing node corresponds to a distributed power source. With the preset minimum operating cost of the microgrid as the objective, the power generation of each distributed power source is calculated based on the results of two adjacent iterations. When the deviation between the results of two adjacent iterations is less than a preset value, the calculated power generation of each distributed power source is used as the scheduling and operation scheme of the microgrid.

[0179] As a preferred solution, when the number of edge nodes in the upper-layer information network is n, before the k-th iteration, the intermediate solutions received by the i-th edge node from the two adjacent edge-aware computing nodes are: s i (j) represents the communication distance from the j-th edge node to the i-th edge node in a ring information network composed of controllable distributed power sources, s i (j) = min[|ij|, (n+ij)]; where, before the k-th iteration, the node information received by the i-th edge node from its two adjacent edge-aware computing nodes is: The goal is to achieve the minimum operating cost of the microgrid. Based on the results of two consecutive iterations, the power generation of each distributed power source is calculated, specifically as follows:

[0180]

[0181]

[0182] Among them, P i (t) represents the current generating power of the i-th distributed power source, P. i,min Let P be the minimum power output of the i-th distributed power source. i,max Let P be the maximum power output of the i-th distributed power source. i,maxdown P represents the maximum value of the sudden decrease in the power generation of the i-th distributed power source. i,maxup This represents the maximum value of the sudden increase in the power generation of the i-th distributed power source; when all and If the deviation is less than 0.1%, the iterative calculation ends, and the power generation of each distributed power source is used as the scheduling and operation scheme of the microgrid.

[0183] As a preferred embodiment, the distributed power source includes photovoltaics, wind turbines, gas turbines, diesel generators, and energy storage networks.

[0184] As a preferred embodiment, there is communication between the upper-layer information network and the lower-layer energy network regarding the output power of the distributed power source and the demand data between the loads.

[0185] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0186] Implementing the above embodiments has the following effects:

[0187] The technical solution of this invention constructs a two-layer optimization structure model of information and energy for microgrids and a microgrid grid-connected cost model. It optimizes the microgrid grid-connected cost model using preset constraints. After constructing a parallel constraint optimization iterative algorithm, the optimization problem for all edge-aware computing nodes is iteratively solved, thereby obtaining the power generation corresponding to each distributed power source. This serves as the microgrid's scheduling and operation scheme, gradually leading to the global optimal solution for the microgrid cost-economic optimization problem. This invention achieves faster convergence of the constraint optimization algorithm than the standard constraint optimization algorithm without reducing optimization accuracy, ensuring high efficiency and high accuracy in microgrid scheduling.

[0188] Example 3

[0189] Accordingly, the present invention also provides a terminal device, comprising: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the microgrid scheduling method based on edge computing as described in any of the above embodiments.

[0190] The terminal device in this embodiment includes a processor, a memory, and a computer program and computer instructions stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps described in Embodiment 1 above, for example... Figure 1 The steps S101 to S104 are shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above device embodiment, such as the structural model construction module 201.

[0191] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device. For example, the iterative calculation module 204 is used to construct a parallel constraint optimization iterative algorithm based on the optimized microgrid grid connection cost model and the information-energy dual-layer optimization structure model, and to iteratively calculate the parallel constraint optimization iterative algorithm according to the preset minimum operating cost of the microgrid, thereby obtaining the power generation corresponding to each distributed power source, as a scheduling and operation scheme for the microgrid.

[0192] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0193] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0194] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the use of the mobile terminal, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0195] Wherein, if the modules / units integrated in the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the steps of the various method embodiments described above. Wherein, the computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0196] Example 4

[0197] Accordingly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the microgrid scheduling method based on edge computing as described in any of the above embodiments.

[0198] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A microgrid scheduling method based on edge computing, characterized in that, The application relates to a micro-grid information-energy double-layer optimization structure model, which comprises an upper-layer information network and a lower-layer energy network; the upper-layer information network comprises edge perception calculation nodes corresponding to distributed power supplies; and the lower-layer energy network comprises the distributed power supplies and loads. The power generation cost and operation cost of each distributed power supply are obtained, and a micro-grid grid-connection cost model of the distributed power supply is constructed. The micro-grid grid-connection cost model is optimized according to preset constraint conditions. A parallel constraint optimization iterative algorithm is constructed according to the optimized micro-grid grid-connection cost model and the information-energy double-layer optimization structure model, and the parallel constraint optimization iterative algorithm is iteratively calculated according to the preset minimum operation cost of the micro-grid, so that the power generation power of each distributed power supply is obtained as a scheduling operation scheme of the micro-grid. The parallel constraint optimization iterative algorithm is constructed according to the edge perception calculation nodes of the information-energy double-layer optimization structure model and the power generation power of each distributed power supply in the optimized micro-grid grid-connection cost model. The power generation cost and operation cost of each distributed power supply are obtained, and a micro-grid grid-connection cost model of the distributed power supply is constructed. The micro-grid grid-connection cost model is optimized according to preset constraint conditions. where argmin denotes the variable value that makes the objective function argmin(.) take the minimum value, C i (P cg,i (t)) is the micro-grid grid-connected power generation cost, E i (P cg,i (t)) is the environmental governance cost, D i (P cg,i (t)) is the equipment depreciation loss cost, P cg,i (t) is the power generation of the ith controllable distributed power source at time t, is the intermediate solution of the controllable distributed power output of the ith controllable edge node after the kth iteration, is the local Lagrange multiplier of the ith controllable edge node before the kth iteration, ρ is a penalty factor, and ρ>0, 1 1×n denotes a 1-dimensional n-column vector with all 1s, is the intermediate solution of the controllable distributed power output received by the ith controllable edge node before the k+1th iteration from the adjacent edge nodes, n cg denotes the number of distributed controllable power sources, P BESS (t) is the output power of the energy storage system in the micro-grid at time t, are the maximum likelihood estimation values of the local Lagrange multipliers of the i-1th, ith, and i+1th controllable edge nodes after the kth iteration, respectively, is the maximum likelihood estimation value of the intermediate solution of the ith controllable edge node after updating the local controllable distributed power output in the kth iteration, is the local Lagrange multiplier of the ith controllable edge node after the kth iteration. 2.The microgrid scheduling method based on edge computing of claim 1, wherein, A parallel constraint optimization iterative algorithm is constructed according to the optimized micro-grid grid-connection cost model and the information-energy double-layer optimization structure model, and the parallel constraint optimization iterative algorithm is iteratively calculated according to the preset minimum operation cost of the micro-grid, so that the power generation power of each distributed power supply is obtained as a scheduling operation scheme of the micro-grid. The parallel constraint optimization iterative algorithm is constructed according to the edge perception calculation nodes of the information-energy double-layer optimization structure model and the power generation power of each distributed power supply in the optimized micro-grid grid-connection cost model. The grid-connected generation cost of each distributed power source's microgrid is: C i (P cg,i (t))=α i P cg,i (t) 2 +β i P cg,i (t)+γ i ;P cg,i (t) represents the power output of the i-th controllable distributed power source at time t, α i β i and γ i These are the primary, secondary, and constant term coefficients of the generation cost of the i-th controllable distributed power source, respectively, C. i (P cg,i (t) is about P cg,i A function of (t); The environmental governance cost is: e CO , e HC , e PM , and are the governance cost coefficients of carbon monoxide, hydrocarbon, soot, carbon dioxide and nitrogen oxide pollution gases respectively, e gas is the total governance cost coefficient of various pollution gases, E i (P cg,i (t)) is a function of P cg,i (t); The equipment depreciation wear and tear cost is: k i D is the depreciation wear and tear cost coefficient of the controllable distributed power supply, i (P cg,i (t)) is a function of P cg,i (t). The power generation cost and operation cost of each distributed power supply are obtained, and a micro-grid grid-connection cost model of the distributed power supply is constructed. Wherein, the micro-grid grid-connected cost model of the distributed power supply is n cg P cg,i (t) is the power generation of the i-th controllable distributed power supply at time t, C i (P cg,i (t)) is the power generation cost of the i-th controllable distributed power supply, E i (P cg,i (t)) is the environmental governance cost of the i-th controllable distributed power supply caused by power generation, D i (P cg,i (t)) is the depreciation wear cost of the i-th controllable distributed power supply for power generation. 3.The microgrid scheduling method based on edge computing of claim 2, wherein, The micro-grid grid-connection cost model is optimized according to preset constraint conditions. A parallel constraint optimization iterative algorithm is constructed according to the optimized micro-grid grid-connection cost model and the information-energy double-layer optimization structure model, and the parallel constraint optimization iterative algorithm is iteratively calculated according to the preset minimum operation cost of the micro-grid, so that the power generation power of each distributed power supply is obtained as a scheduling operation scheme of the micro-grid. wherein the constraint condition is P PCC (t) is the exchange power of the public coupling node in the micro-grid at time t, P cg,i,min and P cg,i,max are the lower and upper limits of the power generation of the ith controllable distributed power supply, dP cg,i,down and dP cg,i,up are the maximum downhill rate and the maximum uphill rate of the ith controllable distributed power supply. 4.The microgrid scheduling method based on edge computing of claim 1, wherein, The parallel constraint optimization iterative algorithm is constructed according to the edge perception calculation nodes of the information-energy double-layer optimization structure model and the power generation power of each distributed power supply in the optimized micro-grid grid-connection cost model. The power generation cost and operation cost of each distributed power supply are obtained, and a micro-grid grid-connection cost model of the distributed power supply is constructed. The micro-grid grid-connection cost model is optimized according to preset constraint conditions. A parallel constraint optimization iterative algorithm is constructed according to the optimized micro-grid grid-connection cost model and the information-energy double-layer optimization structure model, and the parallel constraint optimization iterative algorithm is iteratively calculated according to the preset minimum operation cost of the micro-grid, so that the power generation power of each distributed power supply is obtained as a scheduling operation scheme of the micro-grid.

5. The edge computing based microgrid dispatching method according to any one of claims 1-4, characterized in that, When the number of edge nodes in the upper layer information network is n, before the kth iteration calculation, the intermediate solution sent by the adjacent two edge perception calculation nodes received by the ith edge node is: s i (j) is the communication distance from the jth edge node to the ith edge node in the ring information network composed of controllable distributed power sources, s i (j) = min[|i-j|, (n+i-j)]; wherein, before the kth iteration, the ith edge node receives node information sent by its two adjacent edge-aware computing nodes as: The preset minimum operation cost of the micro-grid is taken as a target, and the power generation of each distributed power supply is calculated according to the results of adjacent two iterations, specifically as follows: P i (t) is the current power of the i-th distributed power source, P i,min Pmin(i) is the minimum power of the current i-th distributed power source, P i,max Pmax(i) is the maximum power of the current i-th distributed power source, P i,maxdown Pdown(i) is the maximum sudden drop of the power of the current i-th distributed power source, P i,maxup Pup(i) is the maximum sudden rise of the power of the current i-th distributed power source. When the deviation of all and is less than 0.1%, the iterative calculation is ended, and the obtained power generation of each distributed power supply is taken as the dispatching operation scheme of the micro-grid.

6. The edge computing based microgrid dispatching method of any one of claims 1-4, wherein, The distributed power supply includes a photovoltaic, a fan, a gas turbine, a diesel generator and an energy storage network.

7. An edge computing based microgrid scheduling apparatus, characterized by, It comprises: a structure model construction module, a cost model construction module, a model optimization module and an iterative calculation module; The structure model construction module is configured to construct an information-energy double-layer optimization structure model of the micro-grid, wherein the information-energy double-layer optimization structure model comprises an upper-layer information network and a lower-layer energy network, the upper-layer information network comprises edge perception computing nodes corresponding to each distributed power supply, and the lower-layer energy network comprises each distributed power supply and a load. The cost model construction module is configured to obtain the power generation cost and the operation cost of each distributed power supply, and construct a micro-grid grid-connected cost model of the distributed power supply; The model optimization module is configured to optimize the micro-grid grid-connected cost model according to a preset constraint condition; The iterative calculation module is configured to construct a parallel constraint optimization iterative algorithm according to the optimized micro-grid grid-connected cost model and the information-energy double-layer optimization structure model, and perform iterative calculation on the parallel constraint optimization iterative algorithm according to the preset minimum operation cost of the micro-grid, so as to obtain the power generation of each distributed power supply as a dispatching operation scheme of the micro-grid. The parallel constraint optimization iterative algorithm is constructed according to the edge perception computing nodes of the information-energy double-layer optimization structure model and the power generation of each distributed power supply in the optimized micro-grid grid-connected cost model, and the parallel constraint optimization iterative algorithm is as follows: The computer readable storage medium comprises a stored computer program; wherein the computer program controls the device where the computer readable storage medium is located to execute the micro-grid dispatching method based on edge computing according to any one of claims 1-6 when running. where argmin denotes the variable value that makes the objective function argmin(.) take the minimum value, C i (P cg,i (t)) is the micro-grid grid-connected power generation cost, E i (P cg,i (t)) is the environmental governance cost, D i (P cg,i (t)) is the equipment depreciation loss cost, P cg,i (t) is the power generation of the ith controllable distributed power source at time t, is the intermediate solution of the controllable distributed power output of the ith controllable edge node after the kth iteration, is the local Lagrange multiplier of the ith controllable edge node before the kth iteration, ρ is a penalty factor, and ρ > 0, 1 1×n denotes a 1-dimensional n-column vector with all 1s, is the intermediate solution of the controllable distributed power output received by the ith controllable edge node from the adjacent edge nodes before the k+1th iteration, n cg denotes the number of distributed controllable power sources, P BESS (y) is the output power of the energy storage system in the micro-grid at time t, are the maximum likelihood estimation values of the local Lagrange multipliers of the i-1th, ith, and i+1th controllable edge nodes after the kth iteration, is the maximum likelihood estimation value of the intermediate solution of the ith controllable edge node after updating the local controllable distributed power output in the kth iteration, is the local Lagrange multiplier of the ith controllable edge node after the kth iteration.

8. A terminal device, comprising: The computer readable storage medium comprises a stored computer program; wherein the computer program controls the device where the computer readable storage medium is located to execute the micro-grid dispatching method based on edge computing according to any one of claims 1-6 when running.

9. A computer-readable storage medium, characterized in that, ​

Images