Virtual power plant global optimization scheduling method, system and device based on cloud edge collaboration and storage medium

Abstract

The invention discloses a virtual power plant global optimization scheduling method, system and device based on cloud edge collaboration, and a storage medium, and belongs to the technical field of power system scheduling. The method comprises the following steps: based on a cloud edge coordinated regulation and control framework comprising a cloud layer, an edge layer and an end side layer, taking minimization of the total operation cost of a system as a target, comprehensively considering a power balance constraint, a main network interaction constraint, a distribution network transmission constraint, a distributed resource operation constraint, an energy storage equipment constraint and a renewable energy consumption constraint; establishing a global optimization scheduling model; edge collaborative optimization is realized by adopting an alternating direction multiplier method, a global coupling problem is decomposed into local optimization sub-problems and a cloud coordination problem of each region, and aggregation power information is sent to the cloud after the local optimization sub-problems are solved in parallel in each region; and the cloud performs global coordination optimization to generate an optimal scheduling strategy, and issues a scheduling instruction to the edge layer to control the actual operation of the distributed power supply, the energy storage equipment and the controllable load, thereby realizing the collaborative optimization scheduling of the virtual power plant. The problems that in virtual power plant large-scale distributed resource coordination optimization, calculation complexity is high, communication burden is heavy, and real-time performance and global optimality are difficult to consider at the same time are effectively solved.

Description

Technical Field

[0001] This invention belongs to the field of power system dispatching technology, specifically relating to a global optimization dispatching method, system, equipment, and storage medium for virtual power plants based on cloud-edge collaboration. Background Technology

[0002] With the advancement of new power system construction, a large number of distributed energy resources are being connected to the power grid, posing a severe challenge to the traditional centralized dispatching model. Virtual power plants, based on power Internet of Things (IoT) technology, aggregate large-scale distributed resources to achieve intelligent sensing and refined management of distributed energy, and are an important technical means to improve the flexible operation performance and renewable energy absorption capacity of new power systems.

[0003] While current multi-level planning models in virtual power plant optimization scheduling can better simulate the complexity of real-world problems, they still face challenges such as high complexity of global optimization problems under multi-level coupling constraints, heavy computational and communication burdens in large-scale distributed systems, and difficulty in balancing real-time performance and global optimality. Existing research mainly focuses on local optimization and lacks distributed collaborative optimization mechanisms oriented towards global optimality. To address this, we propose a cloud-edge collaborative global optimization scheduling method and system for virtual power plants. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a cloud-edge collaborative global optimization scheduling method, system, device, and storage medium for virtual power plants, thereby solving the problems in existing technologies.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] In a first aspect, embodiments of the present invention provide a global optimization scheduling method for virtual power plants based on cloud-edge collaboration, comprising the following steps:

[0007] Based on a cloud-edge collaborative control framework that includes cloud layer, edge layer and terminal layer, with the goal of minimizing the total system operating cost, a global optimization scheduling model is established by comprehensively considering power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints and renewable energy consumption constraints.

[0008] The alternating direction multiplier method is used to achieve edge collaborative optimization. The global optimization scheduling model is solved, and the global coupling problem is decomposed into local optimization sub-problems in each region and cloud coordination problem. After solving the local optimization sub-problems in parallel in each region, the aggregated power information is sent to the cloud.

[0009] The cloud performs global coordination and optimization to generate the optimal scheduling strategy, and sends scheduling instructions to the edge layer to control the actual operation of distributed power sources, energy storage devices and controllable loads, thereby realizing the collaborative optimization scheduling of virtual power plants.

[0010] Furthermore, the cloud layer includes a virtual power plant control center, which is responsible for aggregating global information, performing optimization calculations, formulating global scheduling strategies, and issuing control commands to the edge layer;

[0011] The edge layer includes edge computing nodes deployed by VPP, which are responsible for the perception, control and preliminary optimization of local resources. They can execute instructions issued by the cloud layer and make autonomous decisions in the event of communication interruption or emergency.

[0012] The terminal layer includes various terminal devices, which are responsible for executing specific control commands and collecting real-time data.

[0013] Furthermore, the objective function of the global optimization scheduling model is:

[0014]

[0015] C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t

[0016] C buy,t =λ buy,t P buy,t Δt

[0017] R sell,t =λ sell,t P sell,t Δt

[0018] In the formula: F represents the total operating cost of the virtual power plant; C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t The cost of purchasing electricity from the main grid during time period t; R sell,t t represents the revenue obtained from selling electricity to the main grid during time period t; N represents the number of distributed resource types; T represents the total number of time periods in the scheduling cycle; a i b i c i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of resource type i in time period t; u i,t λ represents the start / stop status of resource type i during time period t; buy,t P represents the main grid electricity price for period t; buy,t λ represents the power purchased from the main grid during time period t; Δt is the length of the time period; λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

[0019] Furthermore, the constraints of the global optimization scheduling model include:

[0020]

[0021] P grid,t =P buy,t -P sell,t

[0022] P grid,min ≤P grid,t ≤P grid,max

[0023]

[0024] |P r,i,t -P r,i,t-1 |≤R i

[0025]

[0026] In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of load type j during time period t; M be the number of load types; P be the load type. grid,t P represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum capacity of a virtual power plant to sell electricity to the main grid. grid,max This represents the maximum capacity of a virtual power plant to purchase electricity from the main grid. P represents the maximum transmission capacity of distribution network line k; m,j,t Ω represents the net output of the j-th microgrid during time period t. k For the set of microgrids associated with distribution network line k; R represents the minimum and maximum output constraints for resource type i during time period t; i For the ramp rate constraint of resource type i; Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; Maximum charging and discharging power limits for energy storage device i; Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

[0027] Furthermore, the local optimization subproblem is:

[0028]

[0029] In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region. Let be the dual variable of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

[0030] Furthermore, the cloud coordination problem is as follows:

[0031]

[0032] In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. The power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let P be the dual variable of region j in time period t and the k-th iteration; R is the number of marginal regions; load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

[0033] Secondly, embodiments of the present invention provide a global optimization scheduling system for virtual power plants based on cloud-edge collaboration, comprising:

[0034] Framework building module: Constructs a cloud-edge collaborative control framework that includes cloud layer, edge layer, and terminal layer;

[0035] Model building module: Based on the cloud-edge collaborative control framework, with the goal of minimizing the total system operating cost, a global optimization scheduling model is established by comprehensively considering power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints, and renewable energy consumption constraints.

[0036] Collaborative optimization module: The alternating direction multiplier method is used to achieve edge collaborative optimization. The global optimization scheduling model is solved, and the global coupling problem is decomposed into local optimization sub-problems in each region and cloud coordination problem. After solving the local optimization sub-problems in parallel in each region, the aggregated power information is sent to the cloud.

[0037] Furthermore, the cloud layer includes a virtual power plant control center, which is responsible for aggregating global information, performing optimization calculations, formulating global scheduling strategies, and issuing control commands to the edge layer;

[0038] The edge layer includes edge computing nodes deployed by VPP, which are responsible for the perception, control and preliminary optimization of local resources. They can execute instructions issued by the cloud layer and make autonomous decisions in the event of communication interruption or emergency.

[0039] The terminal layer includes various terminal devices, which are responsible for executing specific control commands and collecting real-time data.

[0040] Furthermore, the objective function of the global optimization scheduling model is:

[0041]

[0042] C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t

[0043] C buy,t =λ buy,t P buy,t Δt

[0044] R sell,t =λ sell,t P sell,t Δt

[0045] In the formula: F represents the total operating cost of the virtual power plant; C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t The cost of purchasing electricity from the main grid during time period t; R sell,t t represents the revenue obtained from selling electricity to the main grid during time period t; N represents the number of distributed resource types; T represents the total number of time periods in the scheduling cycle; a i b i c i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of resource type i in time period t; u i,t λ represents the start / stop status of resource type i during time period t; buy,t P represents the main grid electricity price for period t;buy,t λ represents the power purchased from the main grid during time period t; Δt is the length of the time period; λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

[0046] Furthermore, the constraints of the global optimization scheduling model include:

[0047]

[0048] P grid,t =P buy,t -P sell,t

[0049] P grid,min ≤P grid,t ≤P grid,max

[0050]

[0051] |P r,i,t -P r,i,t-1 |≤R i

[0052]

[0053] In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of load type j during time period t; M be the number of load types; P be the load type. grid,t P represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum capacity of a virtual power plant to sell electricity to the main grid. grid,max This represents the maximum capacity of a virtual power plant to purchase electricity from the main grid. P represents the maximum transmission capacity of distribution network line k; m,j,t Ω represents the net output of the j-th microgrid during time period t. k For the set of microgrids associated with distribution network line k; R represents the minimum and maximum output constraints for resource type i during time period t; i For the ramp rate constraint of resource type i; Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; Maximum charging and discharging power limits for energy storage device i; Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

[0054] Furthermore, the local optimization subproblem is:

[0055]

[0056] In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region. Let be the dual variable of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

[0057] Furthermore, the cloud coordination problem is as follows:

[0058]

[0059] In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. The power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let P be the dual variable of region j in time period t and the k-th iteration; R is the number of marginal regions; load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

[0060] Thirdly, embodiments of the present invention provide a computer storage medium storing a readable program. When the program is run, the program can instruct a computing device to execute a cloud-edge collaborative virtual power plant global optimization scheduling method as described above.

[0061] Fourthly, embodiments of the present invention provide an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0062] The memory is used to store at least one executable instruction, which causes the processor to perform an operation corresponding to the cloud-edge collaborative virtual power plant global optimization scheduling method described above.

[0063] Fifthly, embodiments of the present invention provide a computer program product, including computer instructions, which instruct a computing device to perform operations corresponding to the cloud-edge collaborative virtual power plant global optimization scheduling method described above.

[0064] The beneficial effects of this invention are:

[0065] 1. This invention utilizes a three-layer cloud-edge collaborative control framework based on the cloud layer, edge layer, and terminal layer. The cloud layer is responsible for aggregating global information and performing complex optimization calculations, the edge layer is responsible for local resource perception, control, and preliminary optimization, and the terminal layer is responsible for executing specific control commands. This achieves reasonable hierarchical allocation and localized processing of computing tasks. Compared with the traditional centralized scheduling mode, it effectively reduces communication bandwidth requirements, improves the system's real-time response capability, and solves the technical problem of heavy communication burden in centralized scheduling of large-scale distributed resources.

[0066] 2. This invention uses ADMM to achieve edge collaborative optimization, decomposing the global coupling problem into local optimization subproblems in each region and cloud coordination problems. After solving the local optimization subproblems in parallel in each region, the aggregated power information is sent to the cloud. Global optimization is achieved through cloud coordination and dual variable updates, which solves the problems of high computational complexity and difficulty in balancing real-time performance and global optimality. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 This is a flowchart of the optimized scheduling method of the present invention;

[0069] Figure 2 This is a schematic diagram of the cloud-edge collaborative regulation framework structure of the present invention;

[0070] Figure 3 This is a schematic diagram of the ADMM edge collaboration optimization process of the present invention. Detailed Implementation

[0071] 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.

[0072] Example 1

[0073] like Figure 1 As shown, a cloud-edge collaborative global optimization scheduling method for virtual power plants includes the following steps:

[0074] S1, based on a cloud-edge collaborative control framework that includes cloud layer, edge layer and terminal layer, realizes computing task allocation and information flow control;

[0075] The virtual power plant cloud-edge collaborative control framework is a key technological support for achieving global optimized scheduling. Through the collaboration between the "cloud" and the "edge," it enables efficient optimization and rapid control of resources; for example... Figure 2 As shown, the cloud-edge collaborative control framework comprises three levels:

[0076] The cloud layer includes a virtual power plant control center, which is responsible for aggregating global information, performing complex optimization calculations, formulating global scheduling strategies, and issuing control commands to the edge. The cloud layer has powerful computing capabilities and a global perspective, enabling it to achieve overall system optimization.

[0077] Edge layer: This includes edge computing nodes deployed by VPP, which are responsible for the perception, control and initial optimization of local resources. They can execute instructions issued by the cloud and make autonomous decisions in the event of communication interruption or emergency.

[0078] End-side layer: This includes various terminal devices (such as distributed power sources, energy storage, controllable loads, etc.), which are responsible for executing specific control commands and collecting real-time data.

[0079] Furthermore, to ensure efficient collaboration between different levels of the virtual power plant, a clear information flow and control flow mechanism needs to be established:

[0080] 1) Upstream information flow:

[0081] Terminal device → Edge controller: Basic data such as device status, adjustable resource quantity, and operational constraints;

[0082] Edge controller → Cloud platform: Aggregated resource information, regional load forecasts, local network status and other comprehensive data.

[0083] 2) Downlink control flow:

[0084] Cloud platform → Edge controller: Global optimization results, scheduling targets for each region, and market trading strategies;

[0085] Edge controller → Terminal device: Specific control commands, operating settings, and adjustment parameters.

[0086] 3) Horizontal collaboration between edges: Neighboring edge controllers can directly exchange information and coordinate the optimal utilization of boundary resources; distributed optimization across regions is achieved through distributed optimization algorithms.

[0087] The design of information flow and control flow follows the principle of "layered partitioning and hierarchical decision-making". A large number of routine computing and control tasks are completed at the edge layer, while complex optimization and global coordination are performed in the cloud, thereby achieving efficient utilization of computing resources and optimization of system response performance.

[0088] S2, a cloud-edge collaborative control framework built on S1, aims to minimize the total system operating cost. It comprehensively considers power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints, and renewable energy consumption constraints, and establishes a global optimization scheduling model across three levels: main grid, distribution network, and microgrid.

[0089] The virtual power plant global optimization scheduling model takes economic efficiency as its core objective, aiming to minimize the total operating cost of the system; the objective function of the model is:

[0090]

[0091] In the formula: F represents the economic objective, indicating the total operating cost of the virtual power plant (yuan); C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t The cost of purchasing electricity from the main grid during time period t; R sell,t t represents the revenue obtained from selling electricity to the main grid during time period t; N represents the number of distributed resource types; and T represents the total number of time periods in the scheduling cycle.

[0092] The cost calculation process for each item is as follows:

[0093] 1) The operating cost C of the i-th type of distributed resource in time period t. oper,i,t The formula for calculation is:

[0094] C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t

[0095] In the formula: a i b ic i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of resource type i in time period t; u i,t Let represent the start / stop status of resource type i during time period t (0-1 variable).

[0096] 2) Cost of purchasing electricity from the main grid during period t, C buy,t The formula for calculation is:

[0097] C buy,t =λ buy,t P buy,t Δt

[0098] In the formula: λ buy,t P represents the main grid electricity price for period t; buy,t Δt represents the power purchased from the main grid during time period t; Δt is the length of the time period.

[0099] 3) Revenue R obtained from selling electricity to the main grid during period t sell,t The formula for calculation is:

[0100] R sell,t =λ sell,t P sell,t Δt

[0101] In the formula: λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

[0102] The global optimization scheduling model includes the following constraints:

[0103] 1) Power balance constraints:

[0104]

[0105] In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of the j-th type of load during time period t; M is the number of load types.

[0106] 2) Constraints on interaction with the mainnet:

[0107]

[0108] P grid,t =P buy,t -P sell,t

[0109] P grid,min ≤P grid,t ≤P grid,max

[0110] In the formula: P grid,tP represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum (negative) capacity of the virtual power plant to sell electricity to the main grid. grid,max This represents the maximum capacity (positive value) for the virtual power plant to purchase electricity from the main grid.

[0111] 3) Distribution network transmission constraints:

[0112]

[0113] In the formula: P represents the maximum transmission capacity (positive value) of distribution network line k; m,j,t Let Ω be the net output of the j-th microgrid during time period t. k Let be the set of microgrids associated with distribution network line k.

[0114] 4) Constraints on the operation of distributed resources:

[0115]

[0116] |P r,i,t -P r,i,t-1 |≤R i

[0117] In the formula: R represents the minimum and maximum output constraints for resource type i during time period t; i The ramp-up rate limit is set for the i-th type of resource.

[0118] 5) Constraints on energy storage devices:

[0119]

[0120] Where: Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; The maximum charging and discharging power limits for energy storage device i.

[0121] 6) Renewable energy consumption constraints:

[0122]

[0123] Where: Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

[0124] S3 employs the Alternating Direction Multiplier Method (ADMM) to achieve edge collaborative optimization, solving the global optimization scheduling model constructed in S2. It decomposes the global coupling problem into local optimization sub-problems in each region and cloud coordination problems. After solving the local optimization sub-problems in parallel in each region, it sends aggregated power information to the cloud to achieve distributed collaborative scheduling.

[0125] The ADMM algorithm implements edge collaborative optimization by decomposing the edge collaborative optimization problem into local optimization subproblems for each region. For region i, the local optimization objective is to minimize the running cost.

[0126]

[0127] In the formula: f i (x i P represents the operating cost of region i; i,t C represents the total output of region i during time period t; i,t This is the corresponding cost function.

[0128] The regions are coupled through power balance constraints:

[0129]

[0130] In the formula: R is the number of edge regions; P load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

[0131] Introducing auxiliary variable z i,t The global coupling problem is decomposed into:

[0132] Local optimization subproblems (solved in parallel across regions):

[0133]

[0134] In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region (constraint set). Let be the dual variable (Lagrange multiplier) of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

[0135] Coordination Problem (Cloud-based Solution):

[0136]

[0137] In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. Let be the power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let R be the dual variable (Lagrange multiplier) of region j in time period t and the k-th iteration; R is the number of marginal regions; P is the number of marginal regions. load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

[0138] Dual variable update:

[0139]

[0140] like Figure 3 As shown, the steps for edge collaborative optimization using the alternating direction multiplier method include:

[0141] Step 1, Initialization: Set the initial output for each region. and dual variables Setting initial coordination variables in the cloud and penalty parameter ρ;

[0142] Step 2, Parallel Optimization in Regions: Each edge controller solves the local optimization subproblem in parallel to obtain the optimal output. And send the aggregated regional power information to the cloud;

[0143] Step 3, Cloud-based Coordination and Optimization: The cloud solves the coordination problem based on the power information of each region;

[0144] Step 4, Dual Variable Update: Each region updates the dual variables according to the data distributed from the cloud. Update the dual variable;

[0145] The algorithm converges when both the original residual and the dual residual simultaneously meet the following accuracy requirements:

[0146] Original residual Furthermore, dual residuals Where, ε pri To meet the convergence accuracy requirements of the original residuals, ε pri This is the convergence accuracy requirement for the dual residual;

[0147] The iteration may terminate when the maximum number of iterations is reached: k ≥ k max Where k is the number of iterations, k max This represents the maximum number of iterations.

[0148] Based on a similar inventive concept, embodiments of the present invention also provide a computer storage medium storing a readable program, which, when run by a processor, can execute the aforementioned cloud-edge collaborative virtual power plant global optimization scheduling method.

[0149] Based on a similar inventive concept, this invention provides an electronic device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus;

[0150] The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the above-described cloud-edge collaborative virtual power plant global optimization scheduling method.

[0151] Based on a similar inventive concept, this embodiment of the invention also provides a computer program product, including computer instructions, which instruct a computing device to perform the operation corresponding to the above-described cloud-edge collaborative virtual power plant global optimization scheduling method.

[0152] Example 2

[0153] In this embodiment, a global optimization scheduling system for virtual power plants based on cloud-edge collaboration is proposed, specifically including:

[0154] Framework building module: Constructs a cloud-edge collaborative control framework that includes cloud layer, edge layer, and terminal layer;

[0155] Model building module: Based on the cloud-edge collaborative control framework, with the goal of minimizing the total system operating cost, a global optimization scheduling model is established by comprehensively considering power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints, and renewable energy consumption constraints.

[0156] Collaborative optimization module: The alternating direction multiplier method is used to achieve edge collaborative optimization. The global optimization scheduling model is solved, and the global coupling problem is decomposed into local optimization sub-problems in each region and cloud coordination problem. After solving the local optimization sub-problems in parallel in each region, the aggregated power information is sent to the cloud.

[0157] Furthermore, the cloud layer includes a virtual power plant control center, which is responsible for aggregating global information, performing optimization calculations, formulating global scheduling strategies, and issuing control commands to the edge layer;

[0158] The edge layer includes edge computing nodes deployed by VPP, which are responsible for the perception, control and preliminary optimization of local resources. They can execute instructions issued by the cloud layer and make autonomous decisions in the event of communication interruption or emergency.

[0159] The terminal layer includes various terminal devices, which are responsible for executing specific control commands and collecting real-time data.

[0160] Furthermore, the objective function of the global optimization scheduling model is:

[0161]

[0162] C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t

[0163] C buy,t =λ buy,t P buy,t Δt

[0164] R sell,t =λ sell,t P sell,t Δt

[0165] In the formula: F represents the total operating cost of the virtual power plant; C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t The cost of purchasing electricity from the main grid during time period t; R sell,t t represents the revenue obtained from selling electricity to the main grid during time period t; N represents the number of distributed resource types; T represents the total number of time periods in the scheduling cycle; a i b i c i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of resource type i in time period t; u i,t λ represents the start / stop status of resource type i during time period t; buy,t P represents the main grid electricity price for period t; buy,t λ represents the power purchased from the main grid during time period t; Δt is the length of the time period; λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

[0166] Furthermore, the constraints of the global optimization scheduling model include:

[0167]

[0168] Pgrid,t =P buy,t -P sell,t

[0169] P grid,min ≤P grid,t ≤P grid,max

[0170]

[0171] |P r,i,t -P r,i,t-1 |≤R i

[0172]

[0173] In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of load type j during time period t; M be the number of load types; P be the load type. grid,t P represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum capacity of a virtual power plant to sell electricity to the main grid. grid,max This represents the maximum capacity of a virtual power plant to purchase electricity from the main grid. P represents the maximum transmission capacity of distribution network line k; m,j,t Ω represents the net output of the j-th microgrid during time period t. k For the set of microgrids associated with distribution network line k; R represents the minimum and maximum output constraints for resource type i during time period t; i For the ramp rate constraint of resource type i; Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; Maximum charging and discharging power limits for energy storage device i; Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

[0174] Furthermore, the local optimization subproblem is:

[0175]

[0176] In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region. Let be the dual variable of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

[0177] Furthermore, the cloud coordination problem is as follows:

[0178]

[0179] In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. The power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let P be the dual variable of region j in time period t and the k-th iteration; R is the number of marginal regions; load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

[0180] The methods of the present invention can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code originally stored on a remote recording medium or a non-transitory machine-readable medium and subsequently stored on a local recording medium, downloaded via a network. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the methods described herein. Furthermore, when a general-purpose computer accesses the code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for performing the methods shown herein.

[0181] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A global optimization scheduling method for virtual power plants based on cloud-edge collaboration, characterized in that, Includes the following steps: Based on a cloud-edge collaborative control framework that includes cloud layer, edge layer and terminal layer, with the goal of minimizing the total system operating cost, a global optimization scheduling model is established by comprehensively considering power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints and renewable energy consumption constraints. The alternating direction multiplier method is used to achieve edge collaborative optimization. The global optimization scheduling model is solved, and the global coupling problem is decomposed into local optimization sub-problems in each region and cloud coordination problem. After solving the local optimization sub-problems in parallel in each region, the aggregated power information is sent to the cloud. The cloud performs global coordination and optimization to generate the optimal scheduling strategy, and sends scheduling instructions to the edge layer to control the actual operation of distributed power sources, energy storage devices and controllable loads, thereby realizing the collaborative optimization scheduling of virtual power plants.

2. The global optimization scheduling method for virtual power plants based on cloud-edge collaboration according to claim 1, characterized in that, The cloud layer includes a virtual power plant control center, which is used to aggregate global information, perform optimization calculations, formulate global scheduling strategies, and send control instructions to the edge layer. The edge layer includes edge computing nodes deployed by VPP, which are used for local resource perception, control and preliminary optimization, execution of instructions issued by the cloud layer, and autonomous decision-making in the event of communication interruption or emergency. The terminal layer includes various terminal devices used to execute control commands and collect real-time data.

3. The global optimization scheduling method for virtual power plants based on cloud-edge collaboration according to claim 1, characterized in that, The objective function of the global optimization scheduling model is: C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t C buy,t =λ buy,t P buy,t Δt R sell,t =λ sell,t P sell,t Δt In the formula: F represents the total operating cost of the virtual power plant; C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t R represents the cost of purchasing electricity from the main grid during time period t; sell,t The revenue obtained from selling electricity to the main grid during time period t; N is the number of distributed resource types; T is the total number of scheduling periods; a i b i c i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of the i-th type of distributed resource in time period t; u i,t λ represents the start / stop status of resource type i during time period t; buy,t P represents the main grid electricity price for period t; buy,t Δt represents the power purchased from the main grid during time period t; Δt is the length of the time period. λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

4. The global optimization scheduling method for virtual power plants based on cloud-edge collaboration according to claim 3, characterized in that, The constraints of the global optimization scheduling model include: Power balance constraints: In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of load type j during time period t; M is the number of load types; Constraints on interaction with the mainnet: P grid,t =P buy,t -P sell,t P grid,min ≤P grid,t ≤P grid,max In the formula: P grid,t P represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum capacity of a virtual power plant to sell electricity to the main grid. grid,max The maximum capacity of the virtual power plant to purchase electricity from the main grid; distribution network transmission constraints: In the formula: P represents the maximum transmission capacity of distribution network line k; m,j,t Ω represents the net output of the j-th microgrid during time period t. k For the set of microgrids associated with distribution network line k; Distributed resource operation constraints: |P r,i,t -P r,i,t-1 |≤R i In the formula: Maximum charging and discharging power limits for energy storage device i; R i For the ramp-up rate constraint of resource type i; Constraints of energy storage devices: Where: Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; Maximum charging and discharging power limits for energy storage device i; Renewable energy consumption constraints: Where: Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

5. The global optimization scheduling method for virtual power plants based on cloud-edge collaboration according to claim 4, characterized in that, The local optimization subproblem is: In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region. Let be the dual variable of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

6. The global optimization scheduling method for virtual power plants based on cloud-edge collaboration according to claim 5, characterized in that, The cloud coordination issue is as follows: In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. The power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let P be the dual variable of region j in time period t and the k-th iteration; R is the number of marginal regions; load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

7. A global optimization scheduling system for virtual power plants based on cloud-edge collaboration, characterized in that, include: Model building module: Based on the cloud-edge collaborative control framework that includes cloud layer, edge layer and terminal layer, with the goal of minimizing the total system operating cost, a global optimization scheduling model is established by comprehensively considering power balance constraints, interaction constraints with the main grid, distribution network transmission constraints, distributed resource operation constraints, energy storage device constraints and renewable energy consumption constraints. Model Solving Module: The alternating direction multiplier method is used to achieve edge collaborative optimization and solve the global optimization scheduling model. The global coupling problem is decomposed into local optimization sub-problems in each region and cloud coordination problem. After solving the local optimization sub-problems in parallel in each region, the aggregated power information is sent to the cloud. Optimized scheduling module: The cloud performs global coordination and optimization to generate the optimal scheduling strategy, and sends scheduling instructions to the edge layer to control the actual operation of distributed power sources, energy storage devices and controllable loads, so as to realize the collaborative optimization scheduling of virtual power plants.

8. A cloud-edge collaborative virtual power plant global optimization scheduling system according to claim 7, characterized in that, The cloud layer includes a virtual power plant control center, which is used to aggregate global information, perform optimization calculations, formulate global scheduling strategies, and send control instructions to the edge layer. The edge layer includes edge computing nodes deployed by VPP, which are used for local resource perception, control and preliminary optimization, execution of instructions issued by the cloud layer, and autonomous decision-making in the event of communication interruption or emergency. The terminal layer includes various terminal devices used to execute control commands and collect real-time data.

9. A global optimization scheduling system for virtual power plants based on cloud-edge collaboration according to claim 7, characterized in that, The objective function of the global optimization scheduling model is: C oper,i,t =a i (P r,i,t ) 2 +b i P r,i,t +c i u i,t C buy,t =λ buy,t P buy,t Δt R sell,t =λ sell,t P sell,t Δt In the formula: F represents the total operating cost of the virtual power plant; C oper,i,t Let C be the operating cost of the i-th type of distributed resource during time period t; buy,t R represents the cost of purchasing electricity from the main grid during time period t; sell,t The revenue obtained from selling electricity to the main grid during time period t; N is the number of distributed resource types; T is the total number of scheduling periods; a i b i c i P represents the cost coefficient for the i-th type of resource; r,i,t For the output of the i-th type of distributed resource in time period t; u i,t λ represents the start / stop status of resource type i during time period t; buy,t P represents the main grid electricity price for period t; buy,t Δt represents the power purchased from the main grid during time period t; Δt is the length of the time period. λ sell,t P is the main grid electricity price for period t; sell,t The power sold to the main grid during time period t.

10. A cloud-edge collaborative virtual power plant global optimization scheduling system according to claim 9, characterized in that, The constraints of the global optimization scheduling model include: Power balance constraints: In the formula: P r,i,t P represents the output of the i-th type of distributed resource during time period t; load,j,t Let M be the demand of load type j during time period t; M is the number of load types; Constraints on interaction with the mainnet: P grid,t =P buy,t -P sell,t P grid,min ≤P grid,t ≤P grid,max In the formula: P grid,t P represents the net exchange power between the virtual power plant and the main grid during time period t; grid,min P represents the maximum capacity of a virtual power plant to sell electricity to the main grid. grid,max The maximum capacity of the virtual power plant to purchase electricity from the main grid; distribution network transmission constraints: In the formula: P represents the maximum transmission capacity of distribution network line k; m,j,t Ω represents the net output of the j-th microgrid during time period t. k For the set of microgrids associated with distribution network line k; Distributed resource operation constraints: |P r,i,t -P r,i,t-1 |≤R i In the formula: Maximum charging and discharging power limits for energy storage device i; R i For the ramp-up rate constraint of resource type i; Constraints of energy storage devices: Where: Ω ess A collection of energy storage devices; E i,t P represents the energy state of energy storage device i during time period t; c,i,t P d,i,t The charging and discharging power of energy storage device i during time period t; η c,i η d,i The charging and discharging efficiency of energy storage device i; The minimum and maximum energy limits for energy storage device i; Maximum charging and discharging power limits for energy storage device i; Renewable energy consumption constraints: Where: Ω ren A collection of renewable energy sources; P j,t The actual output of renewable energy j during time period t; Let α be the theoretical maximum output of renewable energy j during time period t; α is the renewable energy consumption rate requirement.

11. A global optimization scheduling system for virtual power plants based on cloud-edge collaboration as described in claim 10, characterized in that, The local optimization subproblem is: In the formula: Let x be the decision variable vector for region i in the (k+1)th iteration. i Let X be the decision variable vector (in general form) for region i. i Let i be the feasible region. Let be the dual variable of region i in time period t and the k-th iteration. Let f be the auxiliary variable for region i in time period t and the k-th iteration, and ρ be the penalty parameter of the ADMM algorithm; i (x i P represents the operating cost of region i; i,t This represents the total output of region i during time period t.

12. A cloud-edge collaborative virtual power plant global optimization scheduling system according to claim 11, characterized in that, The cloud coordination issue is as follows: In the formula: Let z be the vector of coordination variables for time period t in the (k+1)th iteration. t z is an auxiliary variable for region i. i,t Let be an auxiliary variable for region i during time period t. The power output of region i in time period t and the (k+1)th iteration. Let be the power output of region j in time period t and the (k+1)th iteration. Let P be the dual variable of region j in time period t and the k-th iteration; R is the number of marginal regions; load,t P represents the total load demand during period t; grid,t The power exchanged with the main network during time period t.

13. A computer storage medium storing a readable program, characterized in that, When the program runs, it can instruct the computing device to execute a cloud-edge collaborative virtual power plant global optimization scheduling method as described in any one of claims 1-6.

14. An electronic device, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the cloud-edge collaborative virtual power plant global optimization scheduling method as described in any one of claims 1-6.

15. A computer program product comprising computer instructions, characterized in that, The computer instructions instruct the computing device to perform the operation corresponding to the cloud-edge collaborative virtual power plant global optimization scheduling method as described in any one of claims 1-6.

Images