Cooperative scheduling method, device, equipment, storage medium and program product

Through the three-layer model collaborative scheduling method, the poor optimization effect caused by independent optimization in virtual power plants is solved, and the coordinated optimization of power systems, virtual power plants and tram clusters is realized, which improves resource utilization efficiency and cost-effectiveness.

CN117669930BActive Publication Date: 2024-11-05TSINGHUA UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311557984.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-11-21
Publication Date
2024-11-05
Estimated Expiration
2043-11-21

AI Technical Summary

Technical Problem

The existing technology independently optimizes the profits of virtual power plants, power system production costs and tram cluster energy consumption costs in virtual power plants, resulting in poor optimization results.

Method used

The three-layer model collaborative scheduling method is adopted to obtain the initial state values ​​of each unit cluster in the power system, and the three-layer model is transformed into a single-layer problem using nonlinear planning technology and numerical optimization algorithm. The linearization processing and optimization solver are used to solve it, combined with Nested iterative algorithms obtain collaborative scheduling results to guide the operation of power systems, virtual power plants and tram clusters.

Benefits of technology

The optimal resource utilization, cost-effectiveness and efficiency between power systems, virtual power plants and tram clusters is achieved, ensuring the coordinated work of all parties' needs and goals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117669930B_ABST
    Figure CN117669930B_ABST
Patent Text Reader

Abstract

The present application relates to a collaborative scheduling method, device, equipment, storage medium and program product. The method includes: obtaining the initial state value of each unit cluster in the power system; determining the collaborative scheduling result based on the initial state value and a pre-established three-layer model of the power system, virtual power plant and tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively, and the collaborative scheduling result is used to collaboratively schedule the power system, the virtual power plant and the tram cluster. The use of this method can improve the collaborative optimization effect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of power system resource configuration, and in particular to a collaborative scheduling method, device, equipment, storage medium and program product. Background Art

[0002] A virtual power plant is a virtual energy enterprise that provides power and system services by integrating and coordinating multiple decentralized power resources. Through modern information technology and intelligent control methods, a virtual power plant integrates and optimizes various power resources to achieve more efficient, reliable and economical power system operations.

[0003] In existing technologies, virtual power plants purchase electricity from the power system at wholesale prices and then sell it to electric vehicle owners at retail prices. This process requires balancing multiple key factors: the profit of the virtual power plant needs to be maximized, the production cost of the power system needs to be minimized, and the energy cost of the electric vehicle cluster also needs to be minimized.

[0004] However, traditional techniques usually optimize these three objectives independently, resulting in poor optimization results. Summary of the invention

[0005] Based on this, it is necessary to provide a collaborative scheduling method, device, equipment, storage medium and program product that can improve the collaborative optimization effect in response to the above technical problems.

[0006] In a first aspect, the present application provides a collaborative scheduling method, comprising:

[0007] Obtain the initial state value of each unit cluster in the power system;

[0008] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0009] In one embodiment, the above-mentioned determination of the coordinated dispatch result based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the electric vehicle cluster includes:

[0010] Based on the initial state value, the three-layer problem corresponding to the three-layer model is transformed into a single-layer problem using nonlinear programming technology and numerical optimization algorithm;

[0011] The single-layer problem is solved and the collaborative scheduling result is obtained.

[0012] In one embodiment, the above single-layer problem corresponds to a nonlinear objective function and constraint condition, and the single-layer problem is solved to obtain a collaborative scheduling result, including:

[0013] The nonlinear objective function and constraint conditions are processed equivalently by using the linearization method to obtain the linear objective function and constraint conditions;

[0014] The optimization solver is used to solve the linear objective function and constraint conditions to obtain candidate solutions for the three-layer model;

[0015] A nested iterative algorithm is used to analyze and process the candidate solutions to obtain the target solution, which is used as the collaborative scheduling result.

[0016] In one embodiment, the linearization processing method includes a strong duality theorem and a linearization theorem. The linearization processing method is used to perform equivalent processing on the nonlinear objective function and constraint conditions to obtain a linear objective function and constraint conditions, including:

[0017] The nonlinear objective function is processed using the strong duality theorem to obtain a linear objective function;

[0018] The linearization theorem is used to process nonlinear constraints and obtain linear constraints.

[0019] In one embodiment, the optimization solver is used to solve the linear objective function and constraint conditions to obtain candidate solutions of the three-layer model, including:

[0020] Set the parameters of the optimization solver once and perform the solution process to obtain the initial feasible solution;

[0021] Based on the initial feasible solution, the parameters are set again and the candidate solution is obtained after solving the problem.

[0022] In one embodiment, the nested iterative algorithm is used to analyze and process the candidate solutions to obtain the target solution, and the target solution is used as the collaborative scheduling result, including:

[0023] Substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system;

[0024] Under the condition that the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative dispatch result;

[0025] When the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and returns to execute the steps of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0026] In a second aspect, the present application also provides a collaborative scheduling device, including:

[0027] An acquisition module is used to obtain the initial state value of each unit cluster in the power system;

[0028] A determination module is used to determine the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated scheduling result is used for the coordinated scheduling of the power system, the virtual power plant and the tram cluster.

[0029] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0030] Obtain the initial state value of each unit cluster in the power system;

[0031] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0033] Obtain the initial state value of each unit cluster in the power system;

[0034] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0035] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0036] Obtain the initial state value of each unit cluster in the power system;

[0037] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0038] The above-mentioned collaborative scheduling method, device, equipment, storage medium and program product first obtain the initial state value of each unit cluster in the power system; then determine the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively, and the collaborative scheduling result is used for the collaborative scheduling of the power system, virtual power plant and tram cluster; in this method, the collaborative scheduling result is calculated through the initial state value and according to the objectives and constraints of each layer, and the collaborative scheduling result will guide the operation of the power system, virtual power plant and tram cluster to achieve optimal resource utilization, cost-effectiveness and efficiency, and help ensure the collaborative work between the power system, virtual power plant and tram cluster to meet the needs and goals of all parties. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 An application environment diagram of a collaborative scheduling method in an embodiment;

[0041] Figure 2 A schematic diagram of a collaborative scheduling method in an embodiment;

[0042] Figure 3 is a flowchart of a collaborative scheduling method in another embodiment;

[0043] Figure 4 is a flowchart of a collaborative scheduling method in another embodiment;

[0044] Figure 5 is a flowchart of a collaborative scheduling method in another embodiment;

[0045] Figure 6 is a flowchart of a collaborative scheduling method in another embodiment;

[0046] Figure 7is a flowchart of a collaborative scheduling method in another embodiment;

[0047] Figure 8 is a structural block diagram of a collaborative scheduling device in one embodiment;

[0048] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] The collaborative scheduling method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in the figure, the three-layer model communicates with the server through the network. The three-layer model consists of three layers of market entities, namely the power system, the virtual power plant and the tram cluster. The virtual power plant purchases electricity from the power system at the wholesale electricity price and sells it to electric car owners at the retail electricity price. The wholesale electricity price signal and the retail electricity price signal in the three-layer model are both real-time electricity prices. In the power system dispatching process, the aggregated charging and discharging power of the virtual power plant determines the optimal dispatching result of the power system market clearing, and the wholesale electricity price signal issued by the power system according to the dispatching result will reversely affect the operation strategy of the virtual power plant; in the internal operation of the virtual power plant, the virtual power plant issues a retail electricity price signal to the tram cluster, and the tram cluster makes the best response according to this price signal, adjusts the time distribution of the charging and discharging load to reduce the charging cost, and the charging load curve of the tram cluster will also reversely affect the retail price signal design of the virtual power plant. The server can be implemented with an independent server or a server cluster consisting of multiple servers.

[0051] In an exemplary embodiment, Figure 2 As shown, a collaborative scheduling method is provided, which is applied to Figure 1 The server in the example is used for explanation, including the following S202 to S204. Among them:

[0052] S202, obtaining the initial state value of each unit cluster in the power system.

[0053] Among them, each unit cluster in the power system refers to a collection of generator sets that work together to generate electricity. These unit clusters can include different types of generator sets, such as thermal power, hydropower, nuclear power, wind power, photovoltaic power, etc., which can work together as needed to meet the power demand of the power system. The operation and management of unit clusters are crucial to ensure the stable and economic operation of the power system. The power system needs to effectively coordinate and control unit clusters to cope with fluctuations in power demand and market changes.

[0054] The initial state values ​​of each cluster of units in the power system usually refer to the initial setting values ​​of the on / off states of these clusters of units. These on / off states determine whether each unit is in operation (on) or non-operation (off) at a specific moment. The selection of initial state values ​​has an important impact on the operation and optimization of the power system because they affect the power generation capacity, fuel consumption, power supply reliability and cost.

[0055] In an embodiment of the present application, the server can obtain the initial state value of each unit cluster in the power system by connecting to the monitoring and control system of the power system.

[0056] S204, determining the coordinated dispatch result based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0057] In an embodiment of the present application, the collaborative scheduling result is determined based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster. These three-layer models involve the power system, virtual power plant and tram cluster respectively. Each layer of the model includes a set of objective functions and constraints to describe the specific optimization problem of that layer. The purpose of collaborative scheduling is to determine the best collaborative operation mode of these three levels to achieve the best performance of the entire system.

[0058] Among them, the three-layer model includes objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively, and the coordinated scheduling results are used to coordinate the scheduling of the power system, virtual power plant and tram cluster.

[0059] Continue to see Figure 1The three-layer model shown in the figure can be decomposed into a two-stage Stackelberg game problem, and each stage corresponds to a two-level optimization problem. Stackelberg game is a non-cooperative game model in game theory, in which one participant is called a leader and the other participant is called a follower. The leader first chooses a strategy, and then the follower makes his own decision based on the leader's strategy. In the Stackelberg game, the leader's strategy choice affects the follower's optimal response, so the leader can optimize his strategy by predicting the follower's response. In the power system dispatch, the leader (upper layer) is the virtual power plant and the follower (lower layer) is the power system; while in the internal operation of the virtual power plant, the leader (upper layer) is the virtual power plant and the follower (lower layer) is the tram cluster. The objective function and constraints of the optimization problem corresponding to each market entity are detailed below.

[0060] 1. Minimizing the energy cost of tram clusters

[0061] To simplify the analysis, the embodiment of the present application divides the electric vehicles in the investigation area into different electric vehicle clusters according to the parameters such as the travel characteristics, battery capacity, geographical region and power consumption of electric vehicles, and performs dynamic equivalent modeling for the spatiotemporal characteristics of different fleets. The electric vehicle cluster can minimize the energy cost by adjusting the charging and discharging power. Its objective function includes two parts: electricity expenditure and battery depreciation cost, as shown in the following formula:

[0062]

[0063] 1) In the formula, and are the aggregate charging power and aggregate discharging power of fleet F in period t respectively; is the retail electricity price for electric vehicle charging in period t issued by the virtual power plant; Δt is the unit time interval simulated by the model; λ deg It represents the battery depreciation cost per unit of electricity during the discharge of electric vehicles to the grid. This indicator will affect the willingness of electric vehicle owners to participate in the two-way interaction between the vehicle and the grid; η d The energy conversion efficiency when discharging an electric vehicle battery.

[0064] The specific constraints for fleet charging and discharging optimization include fleet energy balance constraints, upper and lower limit constraints on power state, and charging / discharging power limits, as shown below.

[0065]

[0066]

[0067]

[0068]

[0069] In the formula, is the aggregated power of fleet F during period t; and are the upper and lower limits of the aggregated electricity of fleet F in period t respectively; and are the upper and lower limits of the aggregated charging power of fleet F in period t, respectively; and are the upper and lower limits of the aggregated discharge power of the fleet F in period t; η c Energy conversion efficiency when charging electric vehicle batteries; is the total amount of electricity consumed by fleet F during road driving during period t.

[0070] The model uses the outer approximation method to integrate the operating constraints of each vehicle in different tram clusters and construct the operating constraint parameters of each fleet, as shown below.

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078] In the formula, and are the upper and lower limits of the power of electric vehicle i in period t; and are the maximum charging power and maximum discharging power of electric vehicle i in period t respectively; and are the baseline charging load and corresponding power of electric vehicle i in period t, respectively. In the model, the baseline load is taken as the disordered charging load when the electric vehicle does not participate in the virtual power plant dispatch for demand response; is the power consumption of electric vehicle i during road driving during period t; v F represents the proportion of vehicles in fleet F that participate in demand response. The model assumes that other vehicles that do not participate in demand response follow the baseline charging load pattern; It represents the charging and discharging availability of electric vehicle i in period t, that is, whether it is in a non-driving state and can interact with the power grid in a two-way manner.

[0079] 2. Minimize the cost of power production by power system dispatch

[0080] The embodiment of the present application adopts a unit combination model to simulate the power system dispatch optimization process. Since the solution of the subsequent three-layer optimization problem involves the introduction of a large number of auxiliary variables, in order to reduce the complexity of the problem, the model clusters the units according to indicators such as the power generation efficiency, fuel type, geographical location, fuel price and single-unit capacity of the thermal power units, and uses the clustered unit combination model to model the unit cluster. The basic idea of ​​this method is to use a single integer variable to replace multiple 0-1 variables that characterize the operating status of each unit in the unit cluster, effectively simulating the operating constraints of the thermal power units while reducing the complexity of the unit combination problem. The objective function of the power system dispatch is the total cost of electricity production, including the operating fuel cost and start-up and shutdown cost of the thermal power units, as shown in the following formula:

[0081]

[0082] Where P G,t is the total power generated by the cluster G during period t; and are the actual power generation of wind power and photovoltaic power in the system during period t; u G,t ,y G,t With z G,t are all integer state variables, representing the number of running units, the number of starting units, and the number of shut-down units of cluster G in time period t; b G With c G are the variable operation energy consumption coefficient and the fixed operation energy consumption coefficient of the unit cluster G; FP G Cs is the fuel price corresponding to the unit energy consumption of the unit cluster G; G with cd G They are the unit startup energy consumption and unit shutdown energy consumption of unit cluster G respectively.

[0083] The state variables in the cluster unit combination must meet the following constraints:

[0084] y G,t +z G,t ≤n g (14)

[0085] u G,t -u G,t-1 =y G,t -z G,t (15)

[0086] u G,τ ,y G,t ,zG,t ∈{0,1,2,…,n G} (16)

[0087] Where n G is the total number of thermal power units in the unit cluster G.

[0088] Other operating constraints of the thermal power plant cluster include the minimum on / off time constraint (Equations (17) and (18)), the upper and lower output limit constraints (Equation (19)), and the ramp constraint (Equations (20) and (21)).

[0089]

[0090]

[0091]

[0092]

[0093]

[0094] Where UT G With DT G They are the minimum startup time and minimum shutdown time of cluster G respectively; and are the lower and upper limits of the average output of the unit cluster G; RD G With RU G are the average maximum power decreasing limit and the average maximum power increasing limit of the unit cluster G respectively; SD G with SU G They are respectively the average power decrease limit during the shutdown phase and the average power increase limit during the startup phase of the unit cluster G.

[0095] The wind and photovoltaic output in the system is constrained by the total installed capacity and real-time resource conditions, as shown below:

[0096]

[0097]

[0098] Where, CA w With CA s They are the total installed capacity of wind power and photovoltaic power generation respectively; and They are the wind power generation capacity factor and photovoltaic power generation capacity factor in period t respectively.

[0099] At the same time, the system as a whole needs to meet the power balance constraints and spinning reserve constraints, as shown below:

[0100]

[0101]

[0102] Where, L t is the normal power load of the system during period t; r L 、r W With r S are the spinning reserve rates considering the uncertainty of power load, wind power output and photovoltaic output; r Contingency To meet the emergency backup requirements of the system; It is the total maximum power generation of the unit cluster G in period t determined by the ramp constraint and the upper and lower limits of output.

[0103] 3. Maximizing Profits of Virtual Power Plants

[0104] According to the virtual power plant operation mechanism based on electric vehicle aggregation constructed in the solution, the virtual power plant can maximize its own profits by adjusting the retail electricity price signal:

[0105]

[0106] In the formula, is the wholesale electricity price during period t issued by the power system operator, and its value is obtained by linear transformation of the system marginal generation cost:

[0107]

[0108] In the formula, is the marginal power generation cost of the system in period t, and its value is the same as the Lagrange multiplier corresponding to the unit combination model when the unit state variable takes a fixed value; fix is the fixed cost in the electricity price structure, including transmission and distribution prices, renewable energy subsidies, etc.; α is the linear conversion coefficient from the system marginal power generation cost to the wholesale electricity price.

[0109] In order to maintain the stable operation of the virtual power plant and protect the benefits of electric vehicle owners, the retail electricity price signal issued by the virtual power plant is subject to upper and lower limits and average electricity price constraints:

[0110]

[0111]

[0112] In the formula, Range ret is the maximum fixed floating range of retail electricity price relative to wholesale electricity price; m is the maximum price difference between the average retail electricity price and the average wholesale electricity price during the study period.

[0113] In the above-mentioned collaborative dispatching method, the initial state values ​​of each unit cluster in the power system are first obtained; then the collaborative dispatching results are determined based on the initial state values ​​and the pre-established three-layer model of the power system, virtual power plant and tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively, and the collaborative dispatching results are used for the collaborative dispatching of the power system, virtual power plant and tram cluster; in this method, the collaborative dispatching results are calculated through the initial state values ​​and according to the objectives and constraints of each layer, and the collaborative dispatching results will guide the operation of the power system, virtual power plant and tram cluster to achieve optimal resource utilization, cost-effectiveness and efficiency, and help ensure the collaborative work between the power system, virtual power plant and tram cluster to meet the needs and goals of all parties.

[0114] The above embodiments mentioned that the server can determine the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster. The following embodiments describe in detail the specific process of the server determining the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0115] In an exemplary embodiment, Figure 3 As shown, S204 includes S302 to S304. Among them:

[0116] S302, based on the initial state value, using nonlinear programming technology and numerical optimization algorithm to transform the three-layer problem corresponding to the three-layer model into a single-layer problem.

[0117] S304, solving the single-layer problem to obtain a collaborative scheduling result.

[0118] Among them, nonlinear programming technology is mainly used to deal with the nonlinear part of the objective function and constraints. Specifically, nonlinear programming technology can help transform the nonlinear objective function and constraints in the original three-level problem into an equivalent single-level problem, making the problem easier to solve.

[0119] Numerical optimization algorithms are used to determine whether the optimal solution to a given problem satisfies necessary conditions. Taking Karush-Kuhn-Tucker (KKT) as an example, KKT conditions are a method for dealing with constraints, especially equality and inequality constraints. When converting a multi-layer optimization problem into a single-layer problem, KKT conditions can be used to introduce Lagrange multipliers to ensure that the original upper and lower constraints are satisfied in the single-layer problem. KKT conditions are mathematically necessary conditions for determining whether a solution is a constrained optimal solution. These conditions include gradient conditions, complementary relaxation conditions, and Lagrange multiplier non-negativity conditions, which are usually used to describe and manage constraints.

[0120] Optionally, the KKT condition includes the following parts: (1) Primal feasibility: Primal feasibility requires that the decision variables satisfy the constraints of the problem. (2) Stability: The stability condition requires that the gradient of the objective function of the problem at the optimal solution is equal to zero. This means that the optimal solution should lie in the flat part of the objective function. Dual feasibility: The dual feasibility condition requires that the Lagrange multipliers (also called dual variables) are non-negative. (3) This means that the multipliers cannot take negative values. (4) Complementary slackness condition: The complementary slackness condition states that if a constraint is not strictly active (that is, it has no effect on the optimal solution), then the Lagrange multiplier corresponding to the constraint must be zero. Vice versa, if the multiplier is not zero, then the corresponding constraint must be strictly active.

[0121] In the embodiment of the present application, based on the initial state value, the three-layer problem corresponding to the three-layer model is converted into a single-layer problem using nonlinear programming technology and numerical optimization algorithm. Then the single-layer problem is solved to obtain the collaborative scheduling result. Optionally, according to the optimization objectives of the above-mentioned market entities, and Figure 1 The three-layer model in , the three-layer optimization model can be expressed as follows:

[0122]

[0123] And meet the following conditions:

[0124] (1)

[0125] (2)

[0126] (3) Formula (27) - Formula (29)

[0127] In the formula, OF VPP , OF SYS They are the optimization objective functions of the virtual power plant operator, electric vehicle fleet, and power system dispatch mentioned above, and the decision variables include the retail electricity price issued by the virtual power plant operator Electric vehicle charging and discharging power And the power system power generation unit output power P G,t , This formula shows that the virtual power plant operator must seek to maximize its own profits under the premise that the charging and discharging power of the electric vehicle fleet and the output power of the power system's power generation units are optimized to achieve their respective objective functions.

[0128] The embodiment of the present application first uses the corresponding clustered unit combination problem under the disordered charging scenario to obtain the initial value of the on / off state of each unit cluster, and uses the KKT condition to convert the three-layer optimization problem shown in Formula 30) into a single-layer optimization problem under the fixed on / off state of the unit cluster, as shown below, including the original upper-layer optimization problem constraint (Formula 32)), the original lower-layer optimization problem constraint (Formula 33)), the gradient condition (Formula 34)-Formula 38)), the complementary relaxation condition (Formula 39)) and the Lagrange multiplier non-negativity condition (Formula 40)):

[0129]

[0130]

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138] λ j ≥0 (40)

[0139] In the formula, G i is the i-th constraint in the upper optimization problem, i.e., equations (27) to (29); g j is the jth constraint in the lower optimization problem, including equations (2)-(12) and (19)-(25); I and J are the total number of upper optimization and lower optimization constraints, respectively; λ j For the constraint g j The corresponding Lagrange multiplier.

[0140] It should be noted that the purpose of obtaining the initial values ​​of the on / off status of each unit cluster is to simplify the complex three-layer optimization problem. These initial values ​​provide a starting point, making it easier for the optimization algorithm to find the optimal solution to the problem. The selection of virtual power plant and tram parameter values ​​may be affected by more factors, such as charging and discharging power, electricity prices, etc. Therefore, selecting and fixing these parameters may introduce more complexity, while obtaining the on / off status of the unit cluster is relatively simple.

[0141] In the above embodiment, since the three-layer model contains multi-level optimization problems, the processing is complicated. Merging it into a single-layer problem reduces the complexity of the problem and simplifies the mathematical modeling and solution process. In addition, by merging the optimization problems of the power system, virtual power plant and tram cluster, the interests of multiple parties can be comprehensively considered. This helps to achieve the coordinated allocation of resources to meet the needs of the power system, the profit target of the virtual power plant and the energy demand of the tram cluster.

[0142] The above embodiment mentioned that the server can solve the single-layer problem to obtain the collaborative scheduling result. In fact, the single-layer problem corresponds to nonlinear objective functions and constraints. On this basis, the following embodiment will explain in detail the specific process of solving the nonlinear objective functions and constraints to obtain the collaborative scheduling result.

[0143] In an exemplary embodiment, Figure 4 As shown, S304 includes S402 to S406. Among them:

[0144] S402, performing equivalent processing on the nonlinear objective function and the nonlinear constraint conditions by using a linear processing method to obtain a linear objective function and a linear constraint condition.

[0145] S404, using an optimization solver to solve the linear objective function and the linear constraint conditions to obtain a candidate solution of the three-layer model.

[0146] S406: Using a nested iterative algorithm to analyze and process the candidate solutions to obtain a target solution, and using the target solution as a collaborative scheduling result.

[0147] In an embodiment of the present application, first, the nonlinear objective function and nonlinear constraints in the three-layer model are linearized to obtain equivalent linear objective functions and linear constraints. For example, Taylor series expansion or other linearization methods. Then, the linearized problem is passed to the optimization solver to find the optimal solution by continuously adjusting the value of the decision variable. The optimization solver will solve the linearized problem and obtain a candidate solution, which satisfies the linear objective function and the solution of the constraint. Finally, the candidate solution is analyzed and processed by a nested iterative algorithm. This may involve repeatedly adjusting the value of the decision variable and then substituting these new values ​​into the original nonlinear problem to check whether the nonlinear objective function and the constraint are met. This iterative process is carried out at multiple levels, continuously improving the value of the decision variable. Ultimately, the nested iterative algorithm will obtain the target solution, which is a solution that satisfies the original nonlinear objective function and the constraint. This target solution is considered to be a collaborative dispatch result for collaborative dispatch of power systems, virtual power plants, and tram clusters.

[0148] In the above embodiments, the advantages of linearization processing and optimization solvers are fully utilized to find the optimal solution or near-optimal solution of the original nonlinear problem. Through continuous iteration and optimization, it is ensured that the collaborative scheduling results can meet multi-level constraints and goals to achieve optimal resource utilization, cost-effectiveness and efficiency, and help ensure the collaborative work between the power system, virtual power plant and tram cluster to meet the needs and goals of all parties.

[0149] In the above embodiment, it is mentioned that the server can use the linearization processing method to perform equivalent processing on the nonlinear objective function and the nonlinear constraint conditions to obtain the linear objective function and the linear constraint conditions. In fact, the linearization processing method includes the strong duality theorem and the linearization theorem. On this basis, the following embodiment will describe in detail the specific process of using the linearization processing method to perform equivalent processing on the nonlinear objective function and the nonlinear constraint conditions to obtain the linear objective function and the linear constraint conditions.

[0150] In an exemplary embodiment, Figure 5 As shown, S402 includes S502 to S504. Among them:

[0151] S502, using the strong duality theorem to process the nonlinear objective function to obtain a linear objective function.

[0152] S504, using the linearization theorem to process the nonlinear constraint conditions to obtain linear constraint conditions.

[0153] Among them, the strong duality theorem is an important theorem in linear programming, which involves the relationship between the original problem and the dual problem. The theorem shows that if the original linear programming problem is bounded (i.e., there is a feasible solution), then its dual problem is also bounded, and the two have the same optimal objective value. Optionally, the strong duality theorem includes the following points: 1) Original problem: The original linear programming problem is to find a linear combination that minimizes the objective function while satisfying a set of linear inequality constraints. This problem is usually expressed in a minimization form, and the goal is to minimize a linear function. 2) Dual problem: The dual problem is another linear programming problem related to the original problem. It involves finding a linear combination that maximizes linear functions while satisfying a set of dual linear inequality constraints. The dual problem is usually expressed in a maximization form. 3) Strong duality: The strong duality theorem asserts that if the original problem has a bounded solution (i.e., there is a set of feasible solutions), then the dual problem also has a bounded solution, and the optimal objective values ​​of the two are equal. This means that there is a duality relation between the primal problem and the dual problem, which guarantees the consistency of the two in terms of their optimal values. 4) Duality Gap: The Strong Duality Theorem also states that the difference between any primal and dual problem, that is, the difference between the optimal value of the primal problem and the optimal value of the dual problem, will not be greater than zero. This is called the "duality gap" and is often used to measure the optimality between the primal and dual problems. The Strong Duality Theorem plays an important role in linear programming theory and applications, as it ensures the optimality conditions for linear programming problems and that both the primal and dual problems can be considered simultaneously when solving these problems in order to better understand and optimize linear programming models.

[0154] The linearization theorem mainly uses the Fortuny-Amat linearization method, which is a mathematical technique for linearizing nonlinear problems. This method is usually used to solve nonlinear programming or optimization problems. The core idea of ​​the Fortuny-Amat linearization method is to approximate a nonlinear problem as a linear problem, so that it can be solved using common optimization methods such as linear programming or integer linear programming.

[0155] In an embodiment of the present application, the nonlinear objective function is processed using the strong duality theorem to obtain a linear objective function. The nonlinear constraints are processed using the linearization theorem to obtain linear constraints. Optionally, since the objective function and the constraints contain nonlinear terms, these nonlinear terms include the product of the retail electricity price and the charge and discharge power in the objective function and the product of the Lagrange multiplier and the decision variable in the complementary relaxation condition. These nonlinear terms make the problem difficult to solve directly. Therefore, the strong duality theorem and the Fortuny-Amat linearization method are used to linearize these nonlinear terms. This can transform the original problem into a mixed integer linear programming problem, which is easier to solve. Among them, the strong duality theorem is used to process the product of the retail electricity price and the charge and discharge power in the objective function. Through the strong duality theorem, the original problem can be linked to the dual problem and the problem can be transformed into a linear programming problem. The Fortuny-Amat linearization method is used to process the product of the Lagrange multiplier and the decision variable in the complementary relaxation condition. This solves the nonlinear terms in the complementary relaxation condition.

[0156] In the above embodiment, the nonlinear terms are linearized by the strong duality theorem and the linearization theorem, and they are approximated as linear terms, thereby linearizing the objective function of the original problem, which makes the whole problem become a mixed integer linear programming problem.

[0157] In the above embodiment, it is mentioned that the server can use the optimization solver to solve the linear objective function and the linear constraint conditions to obtain the candidate solution of the three-layer model. The following embodiment describes in detail the specific process of using the optimization solver to solve the linear objective function and the linear constraint conditions to obtain the candidate solution of the three-layer model.

[0158] In an exemplary embodiment, Figure 6 As shown, S404 includes S602 to S604. Among them:

[0159] S602, setting the parameters of the optimization solver once, and performing a solution process to obtain an initial feasible solution.

[0160] S604, performing secondary parameter setting based on the initial feasible solution, and obtaining a candidate solution after solving.

[0161] Among them, the optimization solver mainly uses the Gurobi Optimizer (GUROBI for short) to solve mathematical optimization problems such as linear programming, integer programming, mixed integer programming, quadratic programming, and nonlinear programming.

[0162] In the embodiment of the present application, the parameters of the optimization solver are first set once, and a solution process is performed to obtain multiple initial feasible solutions. Then, the parameters of the optimization solver are set a second time, and a solution process is performed to determine a candidate solution from the multiple initial feasible solutions.

[0163] Optionally, due to the introduction of a large number of auxiliary variables in the process of single-layer reconstruction and linearization, the computational complexity of the problem is high, which is specifically manifested in the difficulty in obtaining an initial feasible solution and the slow convergence of the lower bound of the optimal solution. In response to this problem, a two-stage method of differential parameter setting for the GUROBI solver is used in the embodiment of the present application for solving the problem, wherein the first stage of solution aims to obtain the initial feasible solution as soon as possible; the second stage aims to find the optimal solution, and its parameter setting is aimed at proving optimality. Optionally, when using the GUROBI solver to solve the optimization problem, due to the complexity of the problem, it is difficult to obtain the initial feasible solution and the lower bound of the optimal solution converges slowly. In order to overcome this problem, a two-stage method is adopted, in which each stage has different parameter settings: 1) First stage solution: The main goal of this stage is to obtain an initial feasible solution as soon as possible. At this stage, the parameter setting may be configured to be more relaxed to allow the GUROBI solver to explore the problem space more, that is, to allow some variables to take non-integer values. The main task of this stage is to obtain a feasible solution, even if it is not the optimal solution. 2) Second stage solution: Once an initial feasible solution is obtained in the first stage, the second stage begins. The parameter settings in this stage are adjusted to more focus on searching for the optimal solution, that is, the best solution to the problem. The parameter settings may be more stringent to ensure that the algorithm converges quickly towards the global optimal solution.

[0164] In the above embodiment, in the first stage of solving, setting the non-integer variables as loose constraints of integer variables helps to obtain an initial feasible solution. This initial feasible solution may not be the optimal solution, but it helps to start the solver and make it enter the second stage faster. This is beneficial to the speed and efficiency of problem solving, because starting from a good initial point, the solver is more likely to find the optimal solution faster in the second stage. The method of solving using two stages combines the advantages of quickly obtaining an initial solution and focusing on the optimal solution. The loose parameter setting of the first stage allows a feasible solution to be found quickly, while the strict parameter setting of the second stage helps to explore the problem space more deeply to find the optimal solution. This two-stage method combines the advantages of fast start-up and global optimal solution search, which helps to solve the problem more efficiently. In summary, setting non-integer variables as integer variables and adopting a two-stage solution method help to improve the efficiency of the solver, reduce the time of problem solving, and find better solutions in complex problems.

[0165] The above embodiments mentioned that the server can use a nested iterative algorithm to analyze and process candidate solutions to obtain a target solution, and use the target solution as a collaborative scheduling result. On this basis, the following embodiments describe in detail the specific process of using a nested iterative algorithm to analyze and process candidate solutions to obtain a target solution, and use the target solution as a collaborative scheduling result.

[0166] In an exemplary embodiment, Figure 7 As shown, S406 includes S702 to S706. Among them:

[0167] S702, substituting the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system.

[0168] S704, when the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative scheduling result.

[0169] S706, when the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and the step of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster is returned to be executed.

[0170] The cluster unit combination problem refers to grouping different generators (usually thermal power, wind power, solar power, etc.) in the power system to more effectively manage and dispatch their power generation capacity. This problem usually occurs in the optimal dispatch of the power system to ensure the stability and feasibility of power supply.

[0171] Typically, generators can be classified according to their technical characteristics, fuel type, power generation capacity, operating cost and other attributes. The goal of the cluster unit combination problem is to rationally combine these units to maximize the power demand while minimizing costs, improving efficiency or meeting other specific goals. This usually requires considering various constraints, such as the stability of the power system, the start and stop operation of the generators, emission requirements, etc.

[0172] Solving the cluster unit commitment problem usually requires the application of optimization methods, such as mathematical programming, linear programming, mixed-integer linear programming (MILP) and other techniques, to find the best unit commitment solution. This helps power system operators to effectively manage and dispatch power generation resources under different conditions to meet the needs of the power market and ensure system reliability.

[0173] In an embodiment of the present application, the candidate solution is substituted into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system. When the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative scheduling result. When the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and the step of determining the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster is returned.

[0174] Optionally, a nested iterative algorithm is used to continuously improve the on / off state of the unit cluster. Specifically, the embodiment of the present application first uses the corresponding clustered unit combination problem under the disordered charging scenario to obtain the initial value of the on / off state of each unit cluster, and after completing the solution of the three-layer optimization problem under the fixed on / off state of the unit cluster, the aggregated charging and discharging power of the virtual power plant operator at this time is substituted into the clustered unit combination problem for optimality test. If the on / off state of the unit cluster does not change under the given convergence gap, it is considered that the optimal solution of the original three-layer optimization problem has been obtained; otherwise, the new on / off state of the unit cluster is substituted into the three-layer optimization problem as an exogenous variable to solve, forming an iterative process, and solving the above-mentioned single-layer reconstruction and linearized MILP problem in each iteration. When the on / off state of the unit cluster no longer changes, or when the maximum number of iterations is reached, the iteration terminates, and the optimal solution or approximate optimal solution is obtained. Specifically, in the iterative process, the aggregated charging and discharging power of the virtual power plant operator is first obtained. This power represents the charging and discharging behavior of the virtual power plant operator and is the core operation of the virtual power plant. Next, this aggregated charge and discharge power is substituted into the "clustered unit combination problem". This problem may be an independent mathematical model that considers the operation and configuration of the unit cluster. After substitution, it is evaluated whether the on / off state of the unit cluster has changed. This is done by comparing the current unit cluster configuration with the configuration at the previous iteration. If the on / off state of the unit cluster has not changed within a given convergence interval, then it can be considered that the current unit configuration may have reached the optimal state. This convergence interval is usually a small numerical threshold used to judge small changes in the on / off state. If the on / off state changes very little within this interval, it can be considered to be close enough to the optimal solution. If the above conditions are met, it can be considered that the optimal solution to the original three-layer optimization problem has been obtained. This means that under the current behavior of the virtual power plant operator, the configuration of the unit cluster is already optimal and no further iteration is required.

[0175] In the above embodiment, the configuration of the cluster of units is improved through continuous iteration until an optimal or near-optimal state is reached. By performing an optimality test in each iteration, the iteration can be terminated early, thereby improving the computational efficiency.

[0176] A detailed embodiment is given below to illustrate the process of the collaborative scheduling method in the embodiment of the present application. Based on the above embodiment, the implementation process of the method may include the following contents:

[0177] S1, obtain the initial state value of each unit cluster in the power system.

[0178] S2, based on the initial value of the state, uses nonlinear programming technology and numerical optimization algorithm to transform the three-layer problem corresponding to the three-layer model into a single-layer problem; the three-layer model includes the objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively.

[0179] S3, use the strong duality theorem to process the nonlinear objective function and obtain a linear objective function.

[0180] S4, use the linearization theorem to process the nonlinear constraints and obtain linear constraints.

[0181] S5, setting the parameters of the optimization solver once, and performing a solution process to obtain an initial feasible solution.

[0182] S6, based on the initial feasible solution, the parameters are set again, and the candidate solution is obtained after solving.

[0183] S7, substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system.

[0184] S8, when the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the coordinated scheduling result. The coordinated scheduling result is used to coordinate the scheduling of the power system, virtual power plant and tram cluster.

[0185] S9, when the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and the step of determining the coordinated dispatch result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster is returned to be executed.

[0186] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0187] Based on the same inventive concept, the embodiment of the present application also provides a collaborative scheduling device for implementing the collaborative scheduling method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more collaborative scheduling device embodiments provided below can refer to the limitations of the collaborative scheduling method above, and will not be repeated here.

[0188] In an exemplary embodiment, Figure 8 As shown, a collaborative scheduling device is provided, including: an acquisition module 11 and a determination module 12, wherein:

[0189] An acquisition module 11 is used to acquire the initial state value of each unit cluster in the power system;

[0190] Determination module 12 is used to determine the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, virtual power plant and tram cluster respectively, and the collaborative scheduling result is used for the collaborative scheduling of the power system, virtual power plant and tram cluster.

[0191] In another embodiment, another collaborative scheduling device is provided. Based on the above embodiment, the above determination module 12 may include:

[0192] A conversion unit is used to convert the three-layer problem corresponding to the three-layer model into a single-layer problem based on the initial state value by using nonlinear programming technology and numerical optimization algorithm;

[0193] The solving unit is used to solve the single-layer problem and obtain the collaborative scheduling result.

[0194] In another embodiment, another collaborative scheduling device is provided. Based on the above embodiment, the above solving unit may include:

[0195] An equivalent processing subunit is used to perform equivalent processing on the nonlinear objective function and the nonlinear constraint conditions by using a linear processing method to obtain a linear objective function and a linear constraint condition;

[0196] A solution processing subunit is used to solve the linear objective function and the linear constraint conditions by using an optimization solver to obtain a candidate solution of the three-layer model;

[0197] The analysis and processing subunit is used to analyze and process the candidate solutions using a nested iterative algorithm to obtain a target solution, and use the target solution as a collaborative scheduling result.

[0198] In another embodiment, another collaborative scheduling device is provided. Based on the above embodiment, the above equivalent processing sub-unit is specifically used to use the strong duality theorem to process the nonlinear objective function to obtain a linear objective function; and use the linearization theorem to process the nonlinear constraints to obtain linear constraints.

[0199] In another embodiment, another collaborative scheduling device is provided. Based on the above embodiment, the above solution processing subunit is specifically used to set the parameters of the optimization solver once, and perform solution processing to obtain an initial feasible solution; based on the initial feasible solution, the parameters are set a second time, and a candidate solution is obtained after solving.

[0200] In another embodiment, another collaborative scheduling device is provided. Based on the above embodiment, the above analysis and processing subunit is specifically used to substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system; when the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative scheduling result; when the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and return to execute the step of determining the collaborative scheduling result based on the initial state value and a pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0201] Each module in the above-mentioned collaborative scheduling device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0202] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig. 9As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store collaborative scheduling data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a collaborative scheduling method is implemented.

[0203] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0204] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0205] Obtain the initial state value of each unit cluster in the power system;

[0206] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0207] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0208] Based on the initial state value, the three-layer problem corresponding to the three-layer model is transformed into a single-layer problem using nonlinear programming technology and numerical optimization algorithm;

[0209] The single-layer problem is solved and the collaborative scheduling result is obtained.

[0210] In one embodiment, the single-layer problem corresponds to a nonlinear objective function and constraint conditions, and the processor further implements the following steps when executing the computer program:

[0211] The nonlinear objective function and the nonlinear constraint conditions are processed equivalently by using the linearization processing method to obtain the linear objective function and the linear constraint conditions;

[0212] The optimization solver is used to solve the linear objective function and the linear constraints to obtain the candidate solutions of the three-layer model;

[0213] A nested iterative algorithm is used to analyze and process the candidate solutions to obtain the target solution, which is used as the collaborative scheduling result.

[0214] In one embodiment, the linearization processing method includes a strong duality theorem and a linearization theorem, and the processor further implements the following steps when executing the computer program:

[0215] The nonlinear objective function is processed using the strong duality theorem to obtain a linear objective function;

[0216] The linearization theorem is used to process nonlinear constraints and obtain linear constraints.

[0217] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0218] The parameters of the optimization solver are set once, and the solution process is performed to obtain an initial feasible solution.

[0219] Based on the initial feasible solution, the parameters are set again and the candidate solution is obtained after solving the problem.

[0220] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0221] Substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system;

[0222] Under the condition that the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative dispatch result;

[0223] When the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and returns to execute the steps of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0224] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0225] Obtain the initial state value of each unit cluster in the power system;

[0226] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0227] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0228] Based on the initial state value, the three-layer problem corresponding to the three-layer model is transformed into a single-layer problem using nonlinear programming technology and numerical optimization algorithm;

[0229] The single-layer problem is solved and the collaborative scheduling result is obtained.

[0230] In one embodiment, the single-layer problem corresponds to a nonlinear objective function and constraints, and the computer program, when executed by the processor, further implements the following steps:

[0231] The nonlinear objective function and the nonlinear constraint conditions are processed equivalently by using the linearization processing method to obtain the linear objective function and the linear constraint conditions;

[0232] The optimization solver is used to solve the linear objective function and the linear constraints to obtain the candidate solutions of the three-layer model;

[0233] A nested iterative algorithm is used to analyze and process the candidate solutions to obtain the target solution, which is used as the collaborative scheduling result.

[0234] In one embodiment, the linearization processing method includes a strong duality theorem and a linearization theorem, and the computer program further implements the following steps when executed by a processor:

[0235] The nonlinear objective function is processed using the strong duality theorem to obtain a linear objective function;

[0236] The linearization theorem is used to process nonlinear constraints and obtain linear constraints.

[0237] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0238] Set the parameters of the optimization solver once and perform the solution process to obtain the initial feasible solution;

[0239] Based on the initial feasible solution, the parameters are set again and the candidate solution is obtained after solving the problem.

[0240] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0241] Substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system;

[0242] Under the condition that the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative dispatch result;

[0243] When the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and returns to execute the steps of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0244] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0245] Obtain the initial state value of each unit cluster in the power system;

[0246] The coordinated dispatch result is determined based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for the coordinated dispatch of the power system, the virtual power plant and the tram cluster.

[0247] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0248] Based on the initial state value, the three-layer problem corresponding to the three-layer model is transformed into a single-layer problem using nonlinear programming technology and numerical optimization algorithm;

[0249] The single-layer problem is solved and the collaborative scheduling result is obtained.

[0250] In one embodiment, the single-layer problem corresponds to a nonlinear objective function and constraints, and the computer program, when executed by the processor, further implements the following steps:

[0251] The nonlinear objective function and the nonlinear constraint conditions are processed equivalently by using the linearization processing method to obtain the linear objective function and the linear constraint conditions;

[0252] The optimization solver is used to solve the linear objective function and the linear constraints to obtain the candidate solutions of the three-layer model;

[0253] A nested iterative algorithm is used to analyze and process the candidate solutions to obtain the target solution, which is used as the collaborative scheduling result.

[0254] In one embodiment, the linearization processing method includes a strong duality theorem and a linearization theorem, and the computer program further implements the following steps when executed by a processor:

[0255] The nonlinear objective function is processed using the strong duality theorem to obtain a linear objective function;

[0256] The linearization theorem is used to process nonlinear constraints and obtain linear constraints.

[0257] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0258] Set the parameters of the optimization solver once and perform the solution process to obtain the initial feasible solution;

[0259] S6, based on the initial feasible solution, the parameters are set again, and the candidate solution is obtained after solving.

[0260] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0261] Substitute the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system;

[0262] Under the condition that the on / off status of each unit cluster in the power system remains unchanged, the candidate solution is determined as the target solution, and the target solution is used as the collaborative dispatch result;

[0263] When the on / off status of each unit cluster in the power system changes, a nested iterative algorithm is used to improve the initial state value of each unit cluster in the power system, and returns to execute the steps of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

[0264] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0265] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0266] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0267] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A collaborative scheduling method, characterized in that: The method comprises: Obtain the initial state value of each unit cluster in the power system; Determine the coordinated dispatch result based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes the objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster respectively, and the coordinated dispatch result is used for coordinated dispatching of the power system, the virtual power plant and the tram cluster; The determining of the coordinated dispatch result based on the initial state value and the pre-established three-layer model of the power system, the virtual power plant and the tram cluster includes: Based on the initial state value, the three-layer problem corresponding to the three-layer model is converted into a single-layer problem by using nonlinear programming technology and numerical optimization algorithm; Solving the single-layer problem to obtain the collaborative scheduling result; The single-layer problem corresponds to a nonlinear objective function and a nonlinear constraint condition, and solving the single-layer problem to obtain the collaborative scheduling result includes: Using a linearization processing method to perform equivalent processing on the nonlinear objective function and the nonlinear constraint condition to obtain a linear objective function and a linear constraint condition; Using an optimization solver to solve the linear objective function and the linear constraint conditions to obtain a candidate solution for the three-layer model; Using a nested iterative algorithm to analyze and process the candidate solutions to obtain a target solution, and using the target solution as the collaborative scheduling result; The adopting a nested iterative algorithm to analyze and process the candidate solutions to obtain a target solution, and using the target solution as the collaborative scheduling result, includes: Substituting the candidate solution into the cluster unit combination problem to obtain the on / off status of each unit cluster in the power system; In the case where the on / off status of each cluster of units in the power system remains unchanged, determining the candidate solution as the target solution, and using the target solution as the coordinated scheduling result; When the on / off status of each cluster of units in the power system changes, the nested iterative algorithm is used to improve the initial state value of each cluster of units in the power system, and the step of determining the coordinated scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster is returned to execute.

2. The method according to claim 1, characterized in that The linearization processing method includes a strong duality theorem and a linearization theorem. The linearization processing method is used to perform equivalent processing on the nonlinear objective function and the nonlinear constraint condition to obtain a linear objective function and a linear constraint condition, including: Processing the nonlinear objective function using the strong duality theorem to obtain the linear objective function; The nonlinear constraint condition is processed using the linearization theorem to obtain the linear constraint condition.

3. The method according to claim 1, characterized in that The using of the optimization solver to solve the linear objective function and the linear constraint condition to obtain a candidate solution of the three-layer model includes: The parameters of the optimization solver are set once, and a solution process is performed to obtain an initial feasible solution; The parameters are set a second time based on the initial feasible solution, and the candidate solution is obtained after solving the problem.

4. A collaborative scheduling device, characterized in that: The device comprises: An acquisition module is used to obtain the initial state value of each unit cluster in the power system; A determination module, used to determine a coordinated dispatch result based on the initial state value and a pre-established three-layer model of the power system, the virtual power plant and the tram cluster; wherein the three-layer model includes objective functions and constraints corresponding to the power system, the virtual power plant and the tram cluster, respectively, and the coordinated dispatch result is used to coordinately dispatch the power system, the virtual power plant and the tram cluster; The determining module comprises: A conversion unit, configured to convert the three-layer problem corresponding to the three-layer model into a single-layer problem based on the initial state value by using nonlinear programming technology and numerical optimization algorithm; A solving unit, used for solving the single-layer problem to obtain the collaborative scheduling result; The single-layer problem corresponds to a nonlinear objective function and nonlinear constraints, and the solving unit includes: An equivalent processing subunit, used for performing equivalent processing on the nonlinear objective function and the nonlinear constraint condition by using a linear processing method to obtain a linear objective function and a linear constraint condition; A solution processing subunit, used for solving the linear objective function and the linear constraint conditions by using an optimization solver to obtain a candidate solution of the three-layer model; An analysis and processing subunit, configured to analyze and process the candidate solutions using a nested iterative algorithm to obtain a target solution, and use the target solution as the collaborative scheduling result; The analysis and processing subunit is specifically used to improve the initial state value of each unit cluster in the power system by using the nested iterative algorithm when the on / off state of each unit cluster in the power system changes, and return to execute the step of determining the collaborative scheduling result based on the initial state value and the pre-established three-layer model of the power system, virtual power plant and tram cluster.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.