Method and apparatus for online scheduling of a virtual power plant utilizing wind energy storage
The rolling periodic MDP model with a greedy algorithm addresses dimensionality issues in virtual power plant scheduling, enhancing computational efficiency and optimization performance for wind energy storage systems.
Patent Information
- Application Number
- JP2023578863
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2022-09-26
- Filing Date
- 2023-09-25
- Publication Date
- 2024-11-28
- Estimated Expiration
- 2043-09-25
AI Technical Summary
Existing virtual power plants utilizing wind energy storage face challenges in scheduling due to large state and decision spaces, leading to dimensionality issues and high computational overhead in traditional methods like Markov decision processes and reinforcement learning.
A rolling periodic MDP model is formulated to address these challenges, incorporating a greedy algorithm to convert multi-stage scheduling into a single-stage optimization problem, reducing computational overhead and improving optimization performance.
The proposed method effectively handles large decision spaces, providing real-time scheduling plans with reduced computational burden and improved optimization performance, ensuring efficient operation of virtual power plants.
Smart Images

Figure 0007591669000519 
Figure 0007591669000520 
Figure 0007591669000521
Abstract
Description
[Technical field]
[0001] The present invention relates to the field of new energy power generation technology, and in particular to a method and apparatus for online scheduling of a virtual power plant utilizing wind energy storage. [Background technology]
[0002] A virtual power plant is a power supply adjustment management system that uses advanced information and communication technology and software systems to aggregate, coordinate, and optimize renewable energy and flexible resources, and participates in the power market and power network operation as a special power plant. A virtual power plant using wind energy storage is a typical virtual power plant, and its wind power output capacity and load demand are random, so it is somewhat difficult to realize the scheduling of a virtual power plant using wind energy storage according to the real-time predicted wind power output capacity and load demand.
[0003] Traditionally, virtual power plants using wind energy storage often build Markov decision process (MDP) models to describe the dynamics of the system by state transition probabilities to analyze the multi-stage decision problems in the real-time scheduling process. However, because it is difficult to estimate the transition probabilities and the dimensions of the state space and decision space are too large, multi-stage decisions often encounter dimensional disasters, making it difficult to obtain MDP models using traditional methods. Virtual power plants using wind energy storage also develop scheduling plans based on methods such as neural dynamic planning and reinforcement learning, but the computational overhead is still large.
[0004] Therefore, there is an urgent need to propose a highly efficient scheduling plan for a virtual power plant utilizing wind energy storage. Summary of the Invention [Problem to be solved by the invention]
[0005] In order to solve the above technical problems, the present invention proposes an online scheduling method and apparatus for a virtual power plant utilizing wind energy storage, which describes the joint scheduling problem of the virtual power plant utilizing wind energy storage by formulating a rolling period MDP model, and uses a simulation algorithm to obtain the rolling period MDP model to obtain a real-time scheduling plan for the virtual power plant utilizing wind energy storage, thereby avoiding the problem of the increased difficulty of obtaining the scheduling plan due to the dimensionality problem, and at the same time reducing the calculation overhead. [Means for solving the problem]
[0006] In a first aspect, the present invention provides a method for online scheduling of a virtual power plant utilizing wind energy storage, said method comprising: Configuring a target rolling period by the target time period and a plurality of subsequent time periods; Inputting the state of a virtual power plant utilizing wind energy storage in a target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling cycle into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model, and obtaining a basic feasible plan consisting of a scheduling plan of the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling cycle; determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model; The state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with the power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with the power network, and the operating power of an energy storage battery.
[0007] According to the method for online scheduling of a virtual power plant utilizing wind energy storage provided by the present invention, obtaining the rolling optimization sub-model and obtaining a basic feasible plan, which is a scheduling plan of a virtual power plant utilizing wind energy storage in a non-target time period of the target rolling period, includes: The method includes employing a greedy algorithm to solve for the rolling optimization sub-model to obtain a baseline feasible plan comprising a scheduling plan for a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period.
[0008] According to the method for online scheduling of a virtual power plant utilizing wind energy storage provided by the present invention, the process of constructing the rolling optimization sub-model includes: Developing a first objective function targeting a minimum operating cost of a virtual power plant utilizing wind energy storage; Constructing a power balance constraint, a first energy storage battery charge / discharge power upper and lower limit constraint, a first power network exchange power upper and lower limit constraint, and a first cogeneration equipment output power upper and lower limit constraint for the first objective function; generating the rolling optimization sub-model based on the first objective function and its corresponding constraints.
[0009] According to the online scheduling method for a virtual power plant utilizing wind energy storage provided by the present invention, the expression of the first objective function is shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0010] According to the online scheduling method for a virtual power plant utilizing wind energy storage provided by the present invention, obtaining a scheduling plan for the virtual power plant utilizing wind energy storage during the target time period based on the basic feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model includes the following steps: determining an operating cost of a virtual power plant utilizing wind energy storage corresponding to a scheduling plan of the virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; The operating cost of the virtual power plant utilizing wind energy storage is input into a lower-level optimization sub-model, and the simulation algorithm is employed to obtain the lower-level optimization sub-model to obtain a scheduling plan for the virtual power plant utilizing wind energy storage during the target time period.
[0011] According to the method for online scheduling of a virtual power plant with wind energy storage provided by the present invention, the process of constructing the lower optimization sub-model comprises: constructing a second objective function with a goal of minimizing the sum of an operating cost of a virtual power plant utilizing wind energy storage during the target time period and an expected value of an operating cost of a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; For the second objective function, establish a power balance constraint, a charge state constraint of an energy storage battery, a second energy storage battery charge / discharge power upper and lower limit constraints, a second power network exchange power upper and lower limit constraints, and a second cogeneration equipment output power upper and lower limit constraints; generating the lower-level optimization sub-model based on the second objective function and its corresponding constraints.
[0012] According to the online scheduling method for a virtual power plant utilizing wind energy storage provided by the present invention, the expression of the second objective function is shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0013] In a second aspect, the present invention provides an apparatus for online scheduling of a virtual power plant utilizing wind energy storage, said apparatus comprising: a target rolling period determination module for configuring a target rolling period with a target time period and a plurality of subsequent time periods; a first optimization obtaining module for inputting the state of a virtual power plant utilizing wind energy storage in a target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling cycle into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model and obtain a basic feasible plan consisting of a scheduling plan of the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling cycle; a second optimization module for determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model; Wherein, the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with a power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with the power network, and the operating power of an energy storage battery.
[0014] In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable by the processor, wherein, when the processor executes the program, the method for online scheduling of a virtual power plant with wind energy storage as described in the first aspect is performed.
[0015] In a fourth aspect, the present invention provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, performs the method for online scheduling of a virtual power plant with wind energy storage as described in the first aspect. Effect of the Invention
[0016] The method and apparatus for online scheduling of a virtual power station utilizing wind energy storage provided by the present invention considers the randomness of wind power generation and the requirements for real-time scheduling, formulates a rolling period MDP model including a rolling optimization submodel and a sub-optimization submodel to describe the joint scheduling problem of a virtual power station utilizing wind energy storage, and obtains a rolling optimization submodel during application to obtain a basic feasible plan corresponding to the rolling period to which the target time slot belongs, and then employs a simulation algorithm to obtain a sub-optimization submodel based on the basic feasible plan to obtain a scheduling plan for the target time slot. The scheduling plan for the target time slot obtained based on the simulation algorithm is necessarily better than the scheduling plan inferred by the rolling optimization submodel for the target time slot, and the optimization performance is improved. In addition, the simulation algorithm can effectively solve the multi-energy real-time scheduling problem, and can avoid the dimensionality problem caused by the state and decision space being too large, so that the difficulty of optimization and the computation overhead are significantly reduced. [Brief description of the drawings]
[0017] In order to more clearly describe the present invention or the solutions of the prior art, the drawings that need to be used in the description of the embodiments or the prior art are briefly described below. Of course, the drawings described below are some embodiments of the present invention, and those skilled in the art can think of other drawings based on these drawings without creative efforts. [Figure 1]FIG. 1 is a structural schematic diagram of a virtual power station utilizing wind energy storage provided by the present invention. [Diagram 2] 1 is a flow chart of a method for online scheduling of a virtual power plant utilizing wind energy storage provided by the present invention. [Diagram 3] FIG. 1 is a structural schematic diagram of an online scheduling device for a virtual power plant utilizing wind energy storage provided by the present invention; [Figure 4] FIG. 2 is a structural schematic diagram of an electronic device implementing the online scheduling method for a virtual power plant utilizing wind energy storage provided by the present invention; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions are described below clearly and completely with reference to the drawings according to the present invention, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments thereof, and all other embodiments obtained by those skilled in the art without creative labor based on the embodiments in the present invention all belong to the protection scope of the present invention.
[0019] The specific structural schematic diagram of the virtual power plant utilizing wind energy storage considered in the present invention is as shown in Figure 1, which includes a wind power generator, an energy storage battery, a cogeneration (CHP) facility, an information control center, and a smart meter, where the virtual power plant utilizing wind energy storage communicates with an upstream power network, and relies on the flexibility of energy storage control (the energy storage battery can be charged by the wind power generator unit, and can discharge electric energy to the power transmission network through the power line) to improve the schedulability of the existing wind power generation and smooth the operation state of the virtual power plant utilizing wind energy storage. The output power of the cogeneration (CHP) facility in the virtual power plant utilizing wind energy storage, the power exchanged with the power network (the power exchanged between the virtual power plant utilizing wind energy storage and the upstream power network), and the charging and discharging power of the energy storage are scheduled and controlled in real time.
[0020] Hereinafter, a method and device for online scheduling of a virtual power plant utilizing wind energy storage according to the present invention will be described with reference to FIGS. 1 to 4. FIG.
[0021] In a first aspect, the present invention provides an online scheduling method for a virtual power plant utilizing wind energy storage, and as shown in FIG. 2, the method includes S11 to S13.
[0022] In S11, a target rolling period is configured by the target time period and a plurality of subsequent time periods. The target time period is
number
number
number
number
[0023] In S12, the state of a virtual power plant utilizing wind energy storage in a target time zone and information predicted by the virtual power plant utilizing wind energy storage in a non-target time zone of the target rolling cycle are input into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model, and a basic feasible plan consisting of a scheduling plan for the virtual power plant utilizing wind energy storage in the non-target time zone of the target rolling cycle is obtained. The present invention considers the application of a simulation algorithm to the multi-stage real-time scheduling of a virtual power plant utilizing wind energy storage. The core idea of the simulation algorithm is to first find a group of feasible solutions that satisfy the system safety constraints for the target rolling period, i.e., the target time period (
number
number
number
number
number
number
number
number
[0024] Therefore, the present invention first constructs a rolling optimization sub-model to obtain a basic feasible plan required for the simulation algorithm.
number
number
number
number
number
number
number
[0025] In S13, a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period is obtained based on the basic feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model. Using a simulation algorithm, the target time period (
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0026] Wherein, the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with a power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with the power network, and the operating power of an energy storage battery.
[0027] The online scheduling method of the virtual power station utilizing wind energy storage provided by the present invention considers the randomness of wind power generation and the requirements of real-time scheduling, formulates a rolling period MDP model including a rolling optimization submodel and a sub-optimization submodel to describe the joint scheduling problem of the virtual power station utilizing wind energy storage, and obtains the rolling optimization submodel during application to obtain a basic feasible plan corresponding to the rolling period to which the target time slot belongs, and then employs a simulation algorithm to obtain a sub-optimization submodel to obtain a scheduling plan for the target time slot based on the basic feasible plan. The scheduling plan for the target time slot obtained based on the simulation algorithm is necessarily better than the scheduling plan inferred by the rolling optimization submodel for the target time slot, and the optimization performance is improved. In addition, the simulation algorithm can effectively solve the multi-energy real-time scheduling problem, and can avoid the dimensionality problem caused by the state and decision space being too large, so that the difficulty of the optimization and the computation overhead are significantly reduced.
[0028] Based on the above embodiments, as an optional embodiment, obtaining the rolling optimization sub-model and obtaining a basic feasible plan, which is a scheduling plan for a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period, includes: The method includes employing a greedy algorithm to solve for the rolling optimization sub-model to obtain a baseline feasible plan comprising a scheduling plan for a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period.
[0029] The present invention uses a greedy algorithm to make a local optimal selection in each time slot, thereby transforming the multi-stage real-time scheduling problem into a single-stage optimization problem, and determining the target time slot (
number
number
number
number
number
number
[0030] Based on the above embodiments, as an optional embodiment, the rolling optimization sub-model construction process includes: Developing a first objective function targeting a minimum operating cost of a virtual power plant utilizing wind energy storage; Constructing a power balance constraint, a first energy storage battery charge / discharge power upper and lower limit constraint, a first power network exchange power upper and lower limit constraint, and a first cogeneration equipment output power upper and lower limit constraint for the first objective function; generating the rolling optimization sub-model based on the first objective function and its corresponding constraints.
[0031] Considering the uncertainties present in wind power generation and the requirement of real-time scheduling, a Markov decision process (MDP), which is a rolling periodic multi-stage decision, is usually adopted to describe the scheduling problem of a virtual power plant with wind energy storage.
[0032] Here, the rolling optimization sub-model in the rolling periodic MDP model constructs a first objective function with the goal of minimizing the operating cost of a virtual power plant utilizing wind energy storage; Virtual power plant using wind energy storage
number
number
number
number
number
[0033]
number
number
number
number
number
number
number
number
number
number
[0034] Therefore, the expression of the first objective function is shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0035] The power balance constraint for the virtual wind farm is shown as follows:
number
number
number
number
number
number
[0036] The upper and lower limit constraints on the charge and discharge power of the energy storage battery are shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0037] In the rolling optimization submodel, we need to solve a single-stage optimization problem, so the constraints are a set of constraints that are restricted within a time period t, and the whole problem is a linear program. Therefore, the following constraints are adaptively constructed for the first objective function: For the first objective function, a power balance constraint
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0038] The rolling optimization model established in the present invention aims at the lowest operating cost and achieves economic optimization for real-time scheduling in a single time period.
[0039] In addition, the single-stage objective function actually adopted by the rolling optimization sub-model of the present invention may be the minimum weighted value of the operating cost of the virtual power station using wind energy storage and the life loss of the energy storage battery, which is based on the structural model of the virtual power station using wind energy storage and also takes into account the operating life of the energy storage. The first objective function is expressed as follows:
number
number
number
number
number
[0040] The constraints are made to match a first objective function that targets the lowest operating costs of a virtual power plant utilizing wind energy storage.
[0041] Based on the above embodiments, as an optional embodiment, determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the basic feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model includes: determining an operating cost of a virtual power plant utilizing wind energy storage corresponding to a scheduling plan of the virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; The operating cost of the virtual power plant utilizing wind energy storage is input into a lower-level optimization sub-model, and the simulation algorithm is employed to obtain the lower-level optimization sub-model to obtain a scheduling plan for the virtual power plant utilizing wind energy storage during the target time period.
[0042] In addition, when running the simulation algorithm, the basic feasible plan is used
number
number
number
[0043] In the present invention, the operation method of the simulation algorithm is different from that of dynamic programming used in solving MDP. In dynamic programming, forward or backward recursion and backtracking are required to obtain the final result. Since the basic feasible plan provides an approximation for cost calculation, the simulation algorithm needs to recurse only one step to avoid the dimensionality problem of recursion. Therefore, the simulation algorithm can effectively and in real time solve the multi-energy scheduling problem and avoid the dimensionality problem caused by the state and decision space being too large. Compared with the conventional technology, the performance and calculation time of the algorithm proposed in the present invention are superior to the existing methods.
[0044] Based on the above embodiments, as an optional embodiment, the process of constructing the lower optimization sub-model includes: constructing a second objective function with a goal of minimizing the sum of an operating cost of a virtual power plant utilizing wind energy storage during the target time period and an expected value of an operating cost of a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; For the second objective function, establish a power balance constraint, a charge state constraint of an energy storage battery, a second energy storage battery charge / discharge power upper and lower limit constraints, a second power network exchange power upper and lower limit constraints, and a second cogeneration equipment output power upper and lower limit constraints; generating the lower-level optimization sub-model based on the second objective function and its corresponding constraints.
[0045] The objective function of the single-stage optimization problem is the single-step cost function in the sub-optimization model, and the expression of the second objective function (minimization of the overall cost function) is shown as follows:
number
number
number
number
number
number
number
number
[0046] The adaptively constructed constraint for the second objective function is shown as follows: The power balance constraint is expressed as follows:
number
number
number
number
number
number
number
number
number
number
number
[0047] The present invention considers the real-time scheduling problem of a virtual power plant utilizing wind energy storage, and aims to obtain a group scheduling plan to meet the power load and heat demand with the minimum cost. Considering the influence of wind power uncertainty in the system and the real-time requirements of the joint scheduling problem, a rolling period MDP is formulated to describe the joint scheduling problem. Then, a simulation algorithm is adopted to solve the rolling level MDP problem. The performance and calculation time of the algorithm proposed in the present invention are superior to existing methods because the simulation algorithm can solve the multi-energy scheduling problem in real time and avoid the dimensionality problem caused by the state and decision space being too large.
[0048] That is, the present invention can effectively deal with a large decision space, and can obtain a cost-effective scheduling plan while guaranteeing real-time performance.
[0049] The online scheduling device for the virtual power plant utilizing wind energy storage provided by the present invention will be described below. The online scheduling device for the virtual power plant utilizing wind energy storage described below can be referred to correspondingly with the online scheduling method for the virtual power plant utilizing wind energy storage described above. Figure 3 shows a structural schematic diagram of the online scheduling of the virtual power plant utilizing wind energy storage. As shown in Figure 3, the device includes: a target rolling period determination module 21 for configuring a target rolling period in a target time period and a plurality of subsequent time periods; a first optimization obtaining module 22 for inputting the state of a virtual power plant utilizing wind energy storage in a target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling cycle into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model and obtain a basic feasible plan consisting of a scheduling plan of the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling cycle; a second optimization module 23 for determining a scheduling plan of a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a lower optimization sub-model of the rolling period MDP model; Wherein, the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with a power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with the power network, and the operating power of an energy storage battery.
[0050] The online scheduling device for a virtual power station utilizing wind energy storage provided by the present invention considers the randomness of wind power generation and the requirements for real-time scheduling, formulates a rolling period MDP model including a rolling optimization submodel and a sub-optimization submodel to describe the joint scheduling problem of a virtual power station utilizing wind energy storage, and obtains a rolling optimization submodel during application to obtain a basic feasible plan corresponding to the rolling period to which the target time slot belongs, and then employs a simulation algorithm to obtain a sub-optimization submodel based on the basic feasible plan to obtain a scheduling plan for the target time slot. The scheduling plan for the target time slot obtained based on the simulation algorithm is necessarily better than the scheduling plan inferred by the rolling optimization submodel for the target time slot, and the optimization performance is improved. In addition, the simulation algorithm can effectively solve the multi-energy real-time scheduling problem, and can avoid the dimensionality problem caused by the state and decision space being too large, so that the difficulty of obtaining the optimization and the calculation overhead are significantly reduced.
[0051] Based on the above embodiments, as an optional embodiment, the first optimization module 22 specifically comprises: A greedy algorithm is employed to solve the rolling optimization sub-model and obtain a basic feasible plan, which is a scheduling plan for a virtual power plant utilizing wind energy storage during non-target time periods of the target rolling period.
[0052] Based on the above embodiments, as an optional embodiment, the apparatus further includes a construction module, and the construction module specifically includes a first construction unit for constructing a rolling optimization sub-model, and the first construction unit includes: a first objective function construction subunit for constructing a first objective function with a goal of minimizing the operating cost of the virtual power plant utilizing wind energy storage; A first objective function constraint construction subunit for constructing a power balance constraint, a first energy storage battery charge / discharge power upper and lower limit constraint, a first power network exchange power upper and lower limit constraint, and a first cogeneration equipment output power upper and lower limit constraint for the first objective function; a first generating subunit for generating the rolling optimization sub-model based on the first objective function and its corresponding constraints.
[0053] Based on the above embodiments, in one alternative embodiment, the expression of the first objective function is shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0054] Based on the above embodiments, as an optional embodiment, the second optimization module 22 specifically comprises: a determining unit for determining an operating cost of a virtual power plant with wind energy storage corresponding to a scheduling plan of the virtual power plant with wind energy storage during a non-target time period of the target rolling period; and a simulation algorithm optimization obtaining unit for inputting the operating cost of the virtual power plant utilizing wind energy storage into a lower-level optimization sub-model, employing the simulation algorithm to obtain the lower-level optimization sub-model, and obtaining a scheduling plan of the virtual power plant utilizing wind energy storage in the target time period.
[0055] Based on the above embodiments, as an optional embodiment, the construction module includes a second construction unit for constructing a lower-level optimization sub-model, and the second construction unit includes: a second objective function construction subunit for constructing a second objective function with a goal of minimizing an additive sum of an operating cost of a virtual power plant utilizing wind energy storage in the target time period and an expected value of an operating cost of a virtual power plant utilizing wind energy storage in a non-target time period of the target rolling period; a second objective function constraint construction subunit for constructing a power balance constraint, a charge state constraint of an energy storage battery, a second energy storage battery charge / discharge power upper and lower limit constraints, a second power network exchange power upper and lower limit constraints, and an output power upper and lower limit constraints of a second cogeneration facility for the second objective function; and a second generating subunit for generating the lower-level optimization sub-model based on the second objective function and its corresponding constraints.
[0056] Based on the above embodiments, as an optional embodiment, the expression of the second objective function is shown as follows:
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
number
[0057] In a third aspect, FIG. 4 shows a schematic diagram of an entity structure of an electronic device, and as shown in FIG. 4, the electronic device may include a processor 410, a communications interface 420, a memory 430, and a communications bus 440, where the processor 410, the communications interface 420, and the memory 430 complete communication between each other via the communications bus 440. The processor 410 can call logic instructions in the memory 430 to execute an online scheduling method for a virtual power plant utilizing wind energy storage, the method including: configuring a target rolling period in a target time period and a plurality of subsequent time periods; inputting the state of the virtual power plant utilizing wind energy storage in the target time period and the information predicted by the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling period into a rolling optimization sub-model of a pre-constructed rolling period MDP model to obtain the rolling optimization sub-model, and obtaining a basic feasible plan consisting of a scheduling plan for the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling period; and synthesizing the basic feasible plan, a simulation algorithm, and the and determining a scheduling plan for the virtual power plant utilizing wind energy storage for the target time period based on a lower-level optimization sub-model of a rolling period MDP model, wherein the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power network exchange power and a wind power output power in a previous time period, the information predicted by the virtual power plant utilizing wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing wind energy storage, and the scheduling plan for the virtual power plant utilizing wind energy storage includes an output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, a power network exchange power and an operating power of an energy storage battery.
[0058] In addition, the logic instructions in the memory 430 described above may be stored in a computer-readable storage medium when implemented in the form of a software functional unit and sold or used as an independent product. Based on this understanding, the technical solution of the present invention may be essentially or partly embodied in the form of a software product, and the computer software product includes some instructions stored in a storage medium to cause a computer device (such as a personal computer, a server, or a network device) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as a USB disk, a removable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0059] In a fifth aspect, the present invention further provides a computer program product including a computer program, the computer program being capable of being stored in a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is capable of executing an online scheduling method for a virtual power plant utilizing wind energy storage provided by each of the above methods, the method including: configuring a target rolling period in a target time period and a plurality of time periods thereafter; inputting information on a state of the virtual power plant utilizing wind energy storage in the target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling period into a rolling optimization sub-model of a pre-constructed rolling period MDP model to obtain the rolling optimization sub-model; and scheduling the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling period. and determining a scheduling plan for the virtual power plant utilizing wind energy storage in the target time zone based on the basic feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model, wherein the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power network switching power, and a wind power output power in a previous time zone, the information predicted by the virtual power plant utilizing wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing wind energy storage, and the scheduling plan for the virtual power plant utilizing wind energy storage includes an output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, a power network switching power, and an operating power of an energy storage battery.
[0060] In a sixth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an online scheduling method for a virtual power plant utilizing wind energy storage provided by each of the above methods, the method including: configuring a target rolling period in a target time period and a plurality of time periods thereafter; inputting information on a state of the virtual power plant utilizing wind energy storage in the target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling period into a rolling optimization sub-model of a pre-constructed rolling period MDP model to obtain the rolling optimization sub-model, and obtaining a basic feasible plan consisting of a scheduling plan for the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling period; and determining a scheduling plan for the virtual power plant utilizing wind energy storage in the target time period based on the basic feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model, wherein the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power network exchange power, and a wind power output power in a previous time period, the information predicted by the virtual power plant utilizing wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing wind energy storage, and the scheduling plan for the virtual power plant utilizing wind energy storage includes an output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, a power network exchange power, and an operating power of an energy storage battery.
[0061] The above-mentioned device embodiments are merely exemplary, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., located in one place or distributed among multiple network units. The objective of the solution according to the present embodiment can be achieved by selecting some or all of the modules according to actual needs. Those skilled in the art can understand and implement it without any creative effort.
[0062] From the description of the above embodiments, a person skilled in the art can clearly understand that each embodiment may be realized by adding a common hardware platform required for software, and of course may be realized by hardware. Based on such understanding, some of the above technical solutions that essentially or contribute to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes some instructions for making a computer device (such as a personal computer, a server, or a network device) execute the method described in some parts of each embodiment or example.
[0063] Finally, it should be understood that the above embodiments are for illustrating the technical solutions of the present invention, but are not limiting thereof. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art may still modify the technical solutions described in the above embodiments or make equivalent substitutions for some or all of the technical features thereof, and such modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the scope of the technical solutions of the embodiments of the present invention.
[0064] This application claims priority to a Chinese patent application bearing application number 202211176980.1 and entitled "Online scheduling method and apparatus for virtual power plant utilizing wind energy storage", filed with the China Patent Office on September 26, 2022, the entire contents of which are incorporated herein by reference.
Claims
1. 1. A method for online scheduling of a virtual power plant utilizing wind energy storage, comprising: configuring a target rolling period by a target time period and a plurality of time periods following the target time period; Inputting the state of a virtual power plant utilizing wind energy storage in a target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling cycle into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model, and obtaining a basic feasible plan consisting of a scheduling plan of the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling cycle; determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model; Wherein, the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with a power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The online scheduling method for a virtual power plant utilizing wind energy storage, characterized in that the scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with a power network, and the operating power of an energy storage battery.
2. Determining the rolling optimization sub-model to obtain a master feasible plan comprising a scheduling plan for a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period, comprising:
2. The method for online scheduling of a virtual power plant utilizing wind energy storage according to claim 1, further comprising: adopting a greedy algorithm to obtain the rolling optimization sub-model to obtain a basic feasible plan comprising a scheduling plan of the virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period.
3. The process of constructing the rolling optimization sub-model includes: Developing a first objective function targeting a minimum operating cost of a virtual power plant utilizing wind energy storage; Constructing a power balance constraint, a first energy storage battery charge / discharge power upper and lower limit constraint, a first power network exchange power upper and lower limit constraint, and a first cogeneration facility output power upper and lower limit constraint for the first objective function; and generating the rolling optimization sub-model based on the first objective function and its corresponding constraints.
4. The expression of the first objective function is shown as follows: [0010] [0025] [0030] [0045] The power balance constraint expression constructed for the first objective function is shown as follows: [0050] The expression of the upper and lower limit constraints of the charge / discharge power of the first energy storage battery is shown as follows: [006] [0070] [0080] The expressions for the first power network exchange power upper and lower bound constraints are shown as follows: [0097] [0089] ##EQU00011## The expression of the upper and lower limit constraints on the output power of the first cogeneration facility is expressed as follows: ##EQU00012## ##EQU00013## ##EQU00014## Where: ##EQU00015## teeth, ##EQU00016## The operating costs of a hypothetical power plant with wind energy storage during the time period; ##EQU00017## teeth, [0018] The state of a virtual power plant with wind energy storage during the time period; [0019] teeth, [0020] Scheduling plan for a virtual power plant with wind energy storage over time, ##EQU00021## teeth, [0022] the charge state of the energy storage battery in a hypothetical power plant utilizing wind energy storage during the time period; [0023] teeth, ##EQU00024## The wind power output of the virtual power plant utilizing the wind energy storage during the time period; [0025] teeth, [0026] The power exchanged by the power grid of a virtual power plant with wind energy storage during the time period; [0027] teeth, [0028] The power exchanged by the power grid of a virtual power plant with wind energy storage during the time period; [0029] teeth, [0030] The first time point for a virtual power plant using wind energy storage [0031] Output power of each cogeneration facility, [0032] teeth, [Equation 33] The operating power of the energy storage battery of a virtual power plant utilizing wind energy storage during the time period; [0034] teeth, [Equation 35] The first time point for a virtual power plant using wind energy storage [0036] Natural gas consumption of each cogeneration facility, [Equation 37] teeth, [Equation 38] The first time point for a virtual power plant using wind energy storage [0039] Heat supply load of each cogeneration facility, [0040] teeth, [0041] The load of a virtual power plant with wind energy storage during the time period, [0042] is the duration of the time period, [0043] is the electricity network power exchange price, [0044] is the price of natural gas used, [0045] teeth, [0046] The first time point for a virtual power plant using wind energy storage [0047] The slope of the linear function between the output power and the natural gas consumption of a cogeneration facility, [0048] teeth, [0049] The first time point for a virtual power plant using wind energy storage [Number 50] The intercept of the linear function between the output power and the amount of natural gas consumed by a cogeneration facility, [0051] is the charging efficiency of the energy storage battery, [0052] is the discharge efficiency of the energy storage battery, [0,53] is the lower limit of the state of charge of the energy storage battery, [0,54] is the upper limit of the charge state of the energy storage battery, [0.55] is the upper limit of the operating power of the energy storage battery, [0,56] teeth, [0.57] The theoretical lower bound of the operating power of the energy storage battery during the time period, [0,58] teeth, [0,59] The theoretical upper bound of the operating power of the energy storage battery during the time period, [0.60] is the ramp rate, [0061] is the lower bound of the power exchange capacity, [0062] is the upper limit of power exchange capacity, [0063] teeth, [0,64] The theoretical lower bound of the power exchanged by a hypothetical power plant with wind energy storage in a given time period; [0.65] teeth, [0.66] The theoretical upper bound of the power exchanged by a virtual power plant with wind energy storage in a given time period; [0.67] teeth, [0068] The first time point for a virtual power plant using wind energy storage [0,69] The theoretical lower bound of the output power of a cogeneration facility, [Number 70] teeth, [Equation 71] The first time point for a virtual power plant using wind energy storage [Equation 72] The theoretical upper bound of the output power of a cogeneration facility, [Equation 73] is the lower limit of the conversion rate of electrical energy to thermal energy, [74] is the upper limit of the conversion rate of electrical energy to thermal energy, [Number 75] This is the first study on a virtual power plant using wind energy storage. [76] The upper limit of the output power of each cogeneration facility, [77] The method for online scheduling of a virtual power plant with wind energy storage according to claim 3, characterized in that: ##EQU1## is the total number of cogeneration facilities included in the virtual power plant with wind energy storage.
5. determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a sub-optimization sub-model of the rolling period MDP model, determining an operating cost of a virtual power plant utilizing wind energy storage corresponding to a scheduling plan of the virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; The online scheduling method for the virtual power plant with wind energy storage according to claim 1, further comprising: inputting the operating cost of the virtual power plant with wind energy storage into a lower-level optimization sub-model; employing the simulation algorithm to obtain the lower-level optimization sub-model to obtain a scheduling plan for the virtual power plant with wind energy storage in the target time period.
6. The process of constructing the lower optimization sub-model comprises: constructing a second objective function with a goal of minimizing the sum of an operating cost of a virtual power plant utilizing wind energy storage during the target time period and an expected value of an operating cost of a virtual power plant utilizing wind energy storage during a non-target time period of the target rolling period; For the second objective function, establish a power balance constraint, a charge state constraint of an energy storage battery, a second energy storage battery charge / discharge power upper and lower limit constraints, a second power network exchange power upper and lower limit constraints, and a second cogeneration facility output power upper and lower limit constraints; and generating the lower-level optimization sub-model based on the second objective function and its corresponding constraints.
7. The expression of the second objective function is shown as follows: [78] [79] [Number 80] [Number 81] [0082] The expression of the power balance constraint is shown as follows: [Number 83] The expression for the state of charge constraint of the energy storage battery is shown as follows: [Number 84] [Number 85] [Number 86] The expression of the upper and lower limit constraints of the second energy storage battery charge / discharge power is expressed as follows: [Number 87] The expressions of the second power network exchange power upper and lower bound constraints are shown as follows: [Number 88] [Number 89] The expression of the upper and lower limit constraints on the output power of the second cogeneration facility is expressed as follows: [Number 90] [Number 91] Where: [Number 92] teeth, [Number 93] The operating costs of a hypothetical power plant with wind energy storage during the time period; [Number 94] teeth, [Number 95] an operating cost of a virtual power plant with wind energy storage corresponding to a scheduling plan of the virtual power plant with wind energy storage during a time period; [Number 96] teeth, [Number 97] The state of a virtual power plant with wind energy storage during the time period; [Number 98] teeth, [Number 99] Scheduling plan for a virtual power plant with wind energy storage over time, [Number 100] teeth, [Number 101] the charge state of the energy storage battery in a hypothetical power plant utilizing wind energy storage during the time period; [00102] teeth, [00103] The wind power output of the virtual power plant utilizing the wind energy storage during the time period; [Equation 104] teeth, [Number 105] The power exchanged by the power grid of a virtual power plant with wind energy storage during the time period; [Number 106] teeth, [Equation 107] The power exchanged by the power grid of a virtual power plant with wind energy storage during the time period; [Number 108] teeth, [Number 109] The first time point for a virtual power plant using wind energy storage [Equation 110] Output power of each cogeneration facility, [Equation 111] teeth, [Equation 112] The operating power of the energy storage battery of a virtual power plant utilizing wind energy storage during the time period; [Equation 113] teeth, [Equation 114] The first time point for a virtual power plant using wind energy storage [Number 115] Natural gas consumption of each cogeneration facility, [Equation 116] teeth, [Equation 117] The first time point for a virtual power plant using wind energy storage [Number 118] Heat supply load of each cogeneration facility, [Equation 119] teeth, [Number 120] The load of a virtual power plant with wind energy storage during the time period, [Number 121] is the duration of the time period, [Number 122] is the electricity network power exchange price, [Number 123] is the price of natural gas used, [Number 124] teeth, [Number 125] The first time point for a virtual power plant using wind energy storage [Number 126] The slope of the linear function between the output power and the natural gas consumption of a cogeneration facility, [Number 127] teeth, [Number 128] The first time point for a virtual power plant using wind energy storage [Number 129] The intercept of the linear function between the output power and the amount of natural gas consumed by a cogeneration facility, [Number 130] is the charging efficiency of the energy storage battery, [Number 131] is the discharge efficiency of the energy storage battery, [Number 132] is the lower limit of the state of charge of the energy storage battery, [Number 133] is the upper limit of the charge state of the energy storage battery, [Number 134] is the upper limit of the operating power of the energy storage battery, [Number 135] is the ramp rate, [Number 136] is the lower bound of the power exchange capacity, [Number 137] is the upper limit of power exchange capacity, [Number 138] is the lower limit of the conversion rate of electrical energy to thermal energy, [Number 139] is the upper limit of the conversion rate of electrical energy to thermal energy, [Number 140] This is the first study on a virtual power plant using wind energy storage. [Number 141] The upper limit of the output power of each cogeneration facility, [Equation 142] 7. The method for online scheduling of a virtual power plant with wind energy storage according to claim 6, wherein: ##EQU1## is the total number of cogeneration facilities included in the virtual power plant with wind energy storage.
8. An online scheduling device for a virtual power plant utilizing wind energy storage, comprising: a target rolling period determination module for configuring a target rolling period including a target time period and a plurality of time periods following the target time period; a first optimization obtaining module for inputting the state of a virtual power plant utilizing wind energy storage in a target time period and information predicted by the virtual power plant utilizing wind energy storage in a non-target time period of the target rolling cycle into a rolling optimization sub-model of a pre-constructed rolling cycle MDP model to obtain the rolling optimization sub-model, and obtaining a basic feasible plan consisting of a scheduling plan of the virtual power plant utilizing wind energy storage in the non-target time period of the target rolling cycle; a second optimization module for determining a scheduling plan for a virtual power plant utilizing wind energy storage during the target time period based on the master feasible plan, a simulation algorithm, and a lower-level optimization sub-model of the rolling period MDP model; Wherein, the state of the virtual power plant utilizing wind energy storage includes a charge state of an energy storage battery of the virtual power plant utilizing wind energy storage, a power exchanged with a power grid in a previous time period, and a wind power output power; The information predicted by the virtual power plant utilizing the wind energy storage includes a wind power output power forecast value and a load forecast value of the virtual power plant utilizing the wind energy storage, The online scheduling device for a virtual power plant utilizing wind energy storage, characterized in that the scheduling plan of the virtual power plant utilizing wind energy storage includes the output power of a cogeneration facility of the virtual power plant utilizing wind energy storage, the power exchanged with a power network, and the operating power of an energy storage battery.
9. 8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein when the processor executes the computer program, the method for online scheduling of a virtual power plant utilizing wind energy storage according to any one of claims 1 to 7 is implemented.
10. 8. A non-transitory computer readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, a method for online scheduling of a virtual power plant with wind energy storage according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Virtual power plant optimization scheduling method and device
CN113902227A
Large-scale wind power electric power system energy storage power supply optimization equipment
CN114362142A
Method and apparatus for controlling wind power generation
JP2002027679A
Wind power generator system and wind power generator extension method in the system
JP2012249447A
Wind power generator, method for restricting fluctuation of wind power generator and program for restricting fluctuation of wind power generator
JP2015090145A
