Self-discipline optimization control method based on micro-grid group aggregation resources

Abstract

The invention discloses a self-discipline optimization control method based on micro-grid group aggregation resources, and the method comprises the steps: constructing a self-discipline decentralized system based on the micro-grid group aggregation resources, and obtaining the global information of the system through employing a consistency algorithm; constructing a multi-region distributed power supply self-discipline decentralized control strategy based on system global information; and according to the self-discipline decentralized control strategy, establishing a distributed power supply intra-day optimization scheduling model, and adjusting the output of each distributed power supply. According to the invention, the output of each distributed power supply is adjusted through a self-discipline decentralized control strategy.

Description

Technical Field

[0001] The present invention relates to the field of optimal dispatching of power systems, and more particularly, to an autonomous optimization control method based on aggregated resources of a microgrid group. Background Art

[0002] In recent years, due to environmental problems caused by factors such as global warming, as well as the increasingly serious situation of energy shortage, the demand for clean energy such as wind energy and solar energy has also become greater. In addition, new energy generation has unique advantages in islands and remote areas where the power grid is difficult to reach.

[0003] Large-scale wind and photovoltaic power generation will bring some new problems to the operation of the power system. At the same time, its characteristics such as weak randomness and adjustability also bring new challenges to the stable operation and optimal dispatching of the power system.

[0004] Therefore, it is necessary to develop an autonomous optimization control method based on aggregated resources of a microgrid group.

[0005] The information disclosed in the background art part of the present invention is only intended to deepen the understanding of the general background art of the present invention, and should not be regarded as an admission or any form of suggestion that this information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0006] The present invention proposes an autonomous optimization control method based on aggregated resources of a microgrid group, which can adjust the output of each distributed power source through an autonomous decentralized control strategy.

[0007] An autonomous optimization control method based on aggregated resources of a microgrid group includes:

[0008] Construct an autonomous decentralized system based on aggregated resources of a microgrid group, and use a consensus algorithm to obtain the global information of the system;

[0009] Based on the global information of the system, construct a multi-region distributed power source autonomous decentralized control strategy;

[0010] According to the autonomous decentralized control strategy, establish an intraday optimal dispatching model for distributed power sources to adjust the output of each distributed power source.

[0011] Preferably, constructing an autonomous decentralized system based on aggregated resources of a microgrid group and using a consensus algorithm to obtain the global information of the system includes:

[0012] Determine the output expressions of each subsystem in the autonomous decentralized system, and then determine the effective output of each subsystem;

[0013] Through the consensus algorithm, after multiple iterations, the information of each subsystem is converged to the same value, so as to obtain the global information of the system.

[0014] Preferably, the subsystem matrix for the (k + 1)-th iteration is:

[0015]

[0016] where n is the number of subsystem units participating in information exchange, A is the information exchange coefficient matrix, a ij is the coefficient of information exchange between subsystems i and j, N i is the set of subsystems adjacent to subsystem i; d i is the number of subsystems adjacent to subsystem i.

[0017] Preferably, the converged system matrix after iteration is:

[0018]

[0019] In the formula, X0 is the column vector formed by the initial information obtained by each subsystem unit; I is the n-dimensional column vector with all elements being 1.

[0020] Preferably, the multi-region distributed power autonomous decentralized control strategy includes:

[0021] When the actual output of the uncontrollable units within the multi-region distributed power as a whole is less than the planned output, the consensus algorithm is used to solve the output fluctuation values that each distributed power needs to bear. If the adjustable cost or adjustable capacity of a certain distributed power reaches its own limit, it is withdrawn from the network topology, and the adjacent distributed powers update the network state after receiving the information;

[0022] When the actual output of the uncontrollable units within the multi-region distributed power as a whole is greater than the planned output, it is absorbed according to the types of aggregated distributed energy of each region's distributed power;

[0023] When the multi-region distributed power as a whole cannot absorb the power fluctuations caused by the internal distributed energy, the overall power balance is achieved by balancing the power purchase and sale in the market.

[0024] Preferably, absorbing according to the types of aggregated distributed energy of each region's distributed power includes:

[0025] If the distributed power aggregates energy storage devices, the energy storage is charged according to the operating state of the energy storage devices;

[0026] If the distributed power aggregates electric vehicles, price incentives are used to guide the charging of electric vehicles;

[0027] If the distributed power contains small hydropower, the small hydropower is charged through pumped storage.

[0028] Achieving the balance of the overall power through the way of balancing the market purchase and sale of electricity includes:

[0029] The purchase cost of distributed power sources in the balancing market is borne by the distributed power sources in each region with power deficits according to the proportion of the deficit electricity quantity, and the electricity sales revenue is obtained by the distributed power sources in each region corresponding to the power generation according to their power generation quantity.

[0030] Preferably, the intra-day optimal scheduling model of distributed power sources:

[0031] Determine the power regulation cost as the consistency variable, and calculate the power regulation cost of the i-th distributed power source at time t;

[0032] Establish an objective function within each distributed power source to minimize the average regulation cost of each distributed power source, optimize to obtain the income and cost of each distributed power source in the balancing market, and then calculate the overall income of multi-region distributed power sources participating in the electricity market.

[0033] Preferably, the power regulation cost of the i-th distributed power source at time t is:

[0034]

[0035] The objective function is:

[0036]

[0037] In the formula: refers to the regulation cost of the controllable unit of the i-th distributed power source at time t; refers to the regulation cost of the energy storage of the i-th distributed power source at time t; refers to the regulation cost of the interruptible load of the i-th distributed power source at time t.

[0038] Preferably, the income and cost of each distributed power source in the balancing market are optimized through the following formula:

[0039]

[0040] In the formula refers to the purchase cost or electricity sales revenue of the overall distributed power sources in the real-time market at time t; ΔP t rt refers to the overall purchase or sale electricity quantity of the distributed power sources in the real-time market at time t; refers to the purchase and sale electricity quantity provided by the i-th distributed power source in the real-time market at time t; refers to the electricity sales price in the real-time market at time t; refers to the electricity purchase price in the real-time market at time t; ΔP j,trefers to the power regulated by the j-th distributed power source at time t; ΔP j,i,t refers to the power that the i-th distributed power source helps the j-th distributed power source regulate at time t.

[0041] Preferably, the overall revenue of multi-region distributed power sources participating in the power market is calculated by the following formula:

[0042]

[0043] The method of the present invention has other characteristics and advantages, which will be obvious from the accompanying drawings incorporated herein and the subsequent detailed description, or will be described in detail in the accompanying drawings incorporated herein and the subsequent detailed description, and these drawings and detailed description are jointly used to explain the specific principles of the present invention. Description of the Drawings

[0044] By describing the exemplary embodiments of the present invention in more detail in conjunction with the accompanying drawings, the above and other objects, features, and advantages of the present invention will become more obvious, wherein, in the exemplary embodiments of the present invention, the same reference numerals generally represent the same components.

[0045] Figure 1 Shows a flowchart of the steps of an autonomous optimization control method based on aggregated resources of a microgrid group according to an embodiment of the present invention.

[0046] Figure 2 Shows a structural diagram of regional distributed power sources according to an embodiment of the present invention. Detailed Description

[0047] The preferred embodiments of the present invention will be described in more detail below. Although the preferred embodiments of the present invention are described below, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0048] To facilitate understanding of the solution and its effects of the embodiments of the present invention, a specific application example is given below. Those skilled in the art should understand that this example is only for facilitating understanding of the present invention, and any specific details are not intended to limit the present invention in any way.

[0049] Example 1

[0050] Figure 1 Shows a flowchart of the steps of an autonomous optimization control method based on aggregated resources of a microgrid group according to an embodiment of the present invention.

[0051] As Figure 1 shown, the autonomous optimization control method based on aggregated resources of a microgrid group includes:

[0052] Step 101: Construct an autonomous decentralized system based on the aggregated resources of the microgrid cluster, and use the consensus algorithm to obtain the global information of the system;

[0053] Step 102: Based on the global information of the system, construct an autonomous decentralized control strategy for multi-region distributed power sources;

[0054] Step 103: Establish an intraday optimal scheduling model for distributed power sources according to the autonomous decentralized control strategy, and adjust the output of each distributed power source.

[0055] In an example, constructing an autonomous decentralized system based on the aggregated resources of the microgrid cluster and using the consensus algorithm to obtain the global information of the system includes:

[0056] Determine the output expressions of each subsystem in the autonomous decentralized system, and then determine the effective output of each subsystem;

[0057] Through the consensus algorithm, after multiple iterations, the information of each subsystem is converged to the same value, so as to obtain the global information of the system.

[0058] In an example, the subsystem matrix of the (k + 1)-th iteration is:

[0059]

[0060] where n is the number of subsystem units participating in information exchange, A is the information exchange coefficient matrix, a ij is the coefficient of information exchange between subsystem i and j, N i is the set of subsystems adjacent to subsystem i; d i is the number of subsystems adjacent to subsystem i.

[0061] In an example, the converged system matrix after iteration is:

[0062]

[0063] In the formula, X0 is the column vector composed of the initial information obtained by each subsystem unit; I is the n-dimensional column vector with all elements being 1.

[0064] In an example, the autonomous decentralized control strategy for multi-region distributed power sources includes:

[0065] When the actual output of the uncontrollable units inside the multi-region distributed power sources as a whole is less than the planned output, use the consensus algorithm to solve the output fluctuation values that each distributed power source needs to bear. If the adjustable cost or adjustable capacity of a certain distributed power source reaches its own limit, withdraw it from the network topology, and the distributed power sources adjacent to it will update the network state after receiving the information;

[0066] When the actual output of uncontrollable units within the overall multi - area distributed power source is greater than the planned output, consumption is carried out according to the types of aggregated distributed energy in each area's distributed power source;

[0067] When the overall multi - area distributed power source is unable to absorb the power fluctuations caused by internal distributed energy, the overall power balance is achieved through the purchase and sale of electricity in the balancing market.

[0068] In one example, consumption according to the types of aggregated distributed energy in each area's distributed power source includes:

[0069] If the distributed power source aggregates energy storage devices, the energy storage is charged according to the operating state of the energy storage device;

[0070] If the distributed power source aggregates electric vehicles, price incentives are used to guide the charging of electric vehicles;

[0071] If the distributed power source includes small - scale hydropower, the small - scale hydropower is charged through pumped - storage;

[0072] Achieving the overall power balance through the purchase and sale of electricity in the balancing market includes:

[0073] The power purchase cost of the distributed power source in the balancing market is borne by each area's distributed power source with a power deficit in proportion to the deficit electricity volume, and the power sale revenue is obtained by each area's distributed power source corresponding to the power generation according to its power generation volume.

[0074] In one example, the day - ahead optimal scheduling model of the distributed power source:

[0075] Determine the power regulation cost as a consistency variable, and calculate the power regulation cost of the \(i\) - th distributed power source at time \(t\);

[0076] Within each distributed power source, establish an objective function to minimize the average regulation cost of each distributed power source, optimize the revenue and cost of each distributed power source in the balancing market, and then calculate the overall revenue of the multi - area distributed power source participating in the power market.

[0077] In one example, the power regulation cost of the \(i\) - th distributed power source at time \(t\) is:

[0078]

[0079] The objective function is:

[0080]

[0081] In the formula: refers to the regulation cost of the controllable unit of the \(i\) - th distributed power source at time \(t\); Refers to the regulation cost of energy storage of the i-th distributed power source at time t; Refers to the regulation cost of interruptible load of the i-th distributed power source at time t.

[0082] In one example, the revenue and cost of each distributed power source in the balancing market are optimized by the following formula:

[0083]

[0084] In the formula Refers to the power purchase cost or power sales revenue of the overall distributed power sources in the real-time market at time t; ΔP t rt Refers to the overall power purchase or power sales volume of the distributed power sources in the real-time market at time t; Refers to the power purchase and sales volume provided by the i-th distributed power source in the real-time market at time t; Refers to the power sales price in the real-time market at time t; Refers to the power purchase price in the real-time market at time t; ΔP j,t Refers to the power regulated by the j-th distributed power source at time t; ΔP j,i,t Refers to the power that the i-th distributed power source helps the j-th distributed power source regulate at time t.

[0085] In one example, the overall revenue of multi-region distributed power sources participating in the power market is calculated by the following formula:

[0086]

[0087] Specifically, the autonomous decentralized system is composed of subsystem units with controllability and coordinability, and each subsystem unit contains data nodes and data domains. In the autonomous decentralized system, the non-operating state of any individual subsystem unit, such as faults, maintenance, etc., is a normal state for the entire autonomous decentralized system, that is, the structure of the autonomous decentralized system is not fixed, but determined by the states of each subsystem unit. In the autonomous decentralized system, each subsystem unit has the characteristics of equality, locality, and homogeneity. First, the ability of each subsystem to manage itself and interact with other subsystems is controllable by itself; second, each subsystem unit is equal and can complete its own control and management without being managed by other subsystem units, and there is no master-slave relationship between them; in addition, each subsystem unit can manage itself only relying on local information and complete the task of coordinated operation with other subsystem units. In the autonomous decentralized system, each subsystem unit only exchanges information with adjacent subsystem units, and there is only a loose coupling structure between each subsystem unit through data exchange.

[0088] Since each subsystem unit in the self-organizing decentralized system is independent and equal, and there is no subordination relationship among the subsystem units, in order to ensure the coordinated and effective operation of the entire system, the subsystem units need to jointly select and generate a leader. If any subsystem unit fails, the other subsystem units can still normally complete the tasks they undertake, and the subsystem units can coordinate with each other and redistribute the tasks that the system should complete, so as to achieve the overall operation control goal of the system; when the leader fails, the other subsystem units will elect a new leader again to ensure the continuous coordinated operation of the system.

[0089] Therefore, the self-organizing decentralized system is composed of individual independent subsystem units and is driven by the information generated by the outside world and other subsystem units. If the subsystem units are denoted as S i , it can be defined by the function f i . For any input u i , the output can be expressed as:

[0090] y i,k+1 = f i (u i,k+1 , y 1,k , y 2,k ,... y i,k ,... y n,k )

[0091] In the formula: y i,k refers to the k-th output of the subsystem unit i; u i,k refers to the k-th input of the subsystem unit i; n refers to the number of subsystem units in the self-organizing decentralized system. When the subsystem unit S i is driven by the input u i , its output result will normally drive other subsystem units, and a series of effective outputs will be generated by each subsystem unit in the system. Its specific expression is as follows:

[0092] Y S = H S ·(y1, y2,... y i ,... y n ) T

[0093] When the control goal of the self-organizing decentralized system is given, the feedback control u i,k+1 = k i (Y S,i ) can be designed to make the actual output of the system as close as possible to the set target output. In the self-organizing decentralized system, when a certain subsystem unit S iWhen it fails, the overall topology of the system will also change accordingly. The system structure becomes S', and the output results of other subsystem units no longer drive this subsystem unit. Then, other subsystem units continue to achieve the system goal under the structure of system S'.

[0094] In the process of optimizing a distributed system based on an autonomous decentralized system, the information acquisition and information exchange of each distributed unit are the key points and difficulties in the entire system's optimization process. When the consensus algorithm is applied to an autonomous decentralized system, its role is to enable each subsystem unit to obtain the global information of the system only through communication with adjacent subsystem units, which is the basis for realizing autonomous decentralized control. Its basic idea is to make the information of each subsystem unit converge to the same value through multiple iterations. The specific algorithm formula is as follows:

[0095]

[0096] In the formula: x i,k refers to the information obtained by the iterative subsystem unit i at the k-th time; a ij refers to the information exchange coefficient between subsystem units i and j; n refers to the number of subsystem units participating in information exchange. From the perspective of the entire system, it can be written in the following matrix form to obtain:

[0097]

[0098] In the formula: A refers to the information exchange coefficient matrix. Since the distributed system is connected, when the communication between subsystem units in the distributed system is bidirectional and equivalent, then the elements in the information exchange coefficient matrix A are:

[0099]

[0100] In the formula: N i refers to the set of subsystem units adjacent to subsystem unit i; d i refers to the number of subsystem units adjacent to subsystem unit i. After multiple iterations, the information obtained by each subsystem unit finally reaches consistency, and the matrix X converges to X AVE , and its expression is as follows:

[0101]

[0102] In the formula: X0 refers to the column vector composed of the initial information obtained by each subsystem unit; I refers to the n-dimensional column vector with all elements being 1. From the formula of X AVE , it can be seen that when the system reaches convergence, the elements in each subsystem unit are equal and only related to the initial information they obtain respectively.

[0103] Figure 2Shows the structure diagram of regional distributed power supply according to an embodiment of the present invention.

[0104] Distributed power supply can aggregate distributed energy of different types, capacities and even regions through communication and control technologies, enabling it to participate in the electricity market competition and power system operation as a whole. This not only reduces the grid connection risks of each distributed energy source, but also can obtain certain additional profits compared with the individual operation of distributed energy sources. The present invention assumes that each regional distributed power supply contains controllable traditional power generation units (such as thermal power), uncontrollable power generation units (such as wind power or photovoltaic power), energy storage devices, interruptible loads and fixed loads. In the multi-regional distributed power supply network framework, each region is interconnected through an energy and information network, and the distributed power supplies in different regions are independent individuals with their own interests. During the process of multi-regional distributed power supply participating in the power system operation, when a certain region has energy saturation or energy deficit, the distributed power supplies in other regions can help it absorb or supply energy to it, specifically as Figure 2 shown.

[0105] In the self-regulating and decentralized system for aggregating resources in a microgrid cluster, the multi-regional distributed power supply as a whole constitutes a self-regulating and decentralized control system. Each regional distributed power supply is a subsystem unit in the self-regulating and decentralized control system. Each distributed power supply obtains information by communicating with the distributed power supplies in adjacent regions, so as to achieve the coordinated interaction of the multi-regional distributed power supply and ensure the balance of the overall power of the system.

[0106] Distributed power supply self-regulating and decentralized control strategy

[0107] Each distributed power supply aggregates the distributed energy within its own region to make it a whole, and forms an interconnected network structure of multi-regional distributed power supplies with independent interests but mutual communication with the distributed power supplies in other regions. Since each regional distributed power supply is an individual with independent interests, each distributed power supply cannot obtain the network structure information of other regions and all the power generation information of other distributed power supplies. Therefore, during the day-ahead optimization stage of the distributed power supply, each distributed power supply only optimizes with the goal of maximizing its own interests, obtains the day-ahead output plan of the distributed power supply as a whole (that is, the day-ahead planned power and electricity price of the distributed power supply as a whole) and reports it; after receiving the day-ahead output plan information reported by each distributed power supply, the grid trading center combines the network information to conduct market clearing with the goal of maximizing the overall social benefits, so as to determine the overall bidding result of the market and issue it to each distributed power supply.

[0108] However, considering that there are uncontrollable units such as wind and light aggregated inside the distributed power source, and its output has a certain degree of randomness and volatility, there will be certain errors in only using the model to predict the output of uncontrollable units on the day-ahead. Therefore, the distributed power source needs to perform secondary adjustment of the output of each distributed energy source inside it during the day. To ensure the stability and economy of the overall operation of the multi-region distributed power source, when the output of uncontrollable power generation units fluctuates within a certain region, the distributed power source with the lowest average adjustment cost is used to undertake more output, so as to ensure that the overall adjustment cost of the multi-region distributed power source is relatively low. Therefore, the present invention adopts the method of autonomous decentralized control to adjust the output fluctuations of each distributed power source during the day, selects the power adjustment cost as the consistency variable for the day-ahead adjustment, and to maintain the balance of the overall operation of the multi-region distributed power source, the day-ahead autonomous decentralized control strategy for the multi-region distributed power source is as follows:

[0109] 1) When the actual output of uncontrollable units inside the multi-region distributed power source as a whole is less than the planned output, the output fluctuation is jointly borne by the multi-region distributed power source. The consistency algorithm is used to solve the output fluctuation value that each distributed power source needs to bear. If the adjustable cost or adjustable capacity of a certain distributed power source reaches its own limit value, it is withdrawn from the network topology. After the adjacent distributed power sources receive the information, they update the network state, so that the remaining distributed power sources coordinate with each other to continue to complete the power adjustment target, thereby ensuring the stability of the multi-region distributed power source as a whole while ensuring its economy as much as possible.

[0110] 2) When the actual output of uncontrollable units inside the multi-region distributed power source as a whole is greater than the planned output, targeted consumption is carried out according to the types of distributed energy sources aggregated by the distributed power sources in each region. If the distributed power source aggregates energy storage devices, the energy storage can be charged according to the operating state of the energy storage device; if the distributed power source aggregates electric vehicles, price incentives can be used to guide electric vehicle charging; if the distributed power source includes small-scale hydropower, the small-scale hydropower can be charged through pumped storage.

[0111] 3) When the multi-region distributed power source as a whole cannot absorb the power fluctuation caused by the distributed energy inside it, the overall power balance is achieved by purchasing and selling electricity in the balancing market, so as to ensure the safety and stability of the overall operation of the multi-region distributed power source. At this time, the power purchase cost of the distributed power source in the balancing market is borne by each region's distributed power source with a power deficit according to the proportion of the deficit electricity, while the power selling income of the distributed power source in the balancing market is obtained by each region's distributed power source corresponding to the power generation according to its power generation volume.

[0112] Since the self-organizing decentralized control method is adopted to schedule the intra-day fluctuations of multi-region distributed power sources, during the scheduling process, when the distributed power source acting as the leader reaches the upper limit of its own adjustable cost or adjustable capacity, according to the control rules, it needs to be withdrawn from the system network, and a new leader is re-elected by the remaining distributed power sources. Therefore, to simplify the solution process of the consensus algorithm, the distributed power source with the largest adjustable cost is selected as the initial leader of the algorithm, so as to ensure that there is no need to replace the leader.

[0113] Intra-day Optimal Scheduling Model of Distributed Power Sources

[0114] Due to the randomness of the output of uncontrollable units such as wind and light, on the basis of the day-ahead optimization of each distributed member within each distributed power source, it is also necessary to perform secondary adjustment and control on the output of each distributed member according to the error between its actual output and the planned output within the day. The present invention adopts the self-organizing decentralized control method to perform intra-day adjustment on multi-region distributed power sources, and selects the power adjustment cost as the consensus variable. Then, the power adjustment cost of the i-th distributed power source at time t is:

[0115]

[0116] In the formula: refers to the adjustment cost of the controllable unit of the i-th distributed power source at time t; refers to the adjustment cost of the energy storage of the i-th distributed power source at time t; refers to the adjustment cost of the interruptible load of the i-th distributed power source at time t. To ensure the stability and economy of the overall intra-day scheduling of multi-region distributed power sources, an objective function is established within each distributed power source to minimize the average adjustment cost of each distributed power source:

[0117]

[0118] During the optimization process, the power balance constraint conditions need to be satisfied:

[0119]

[0120] In the formula: refers to the adjustment power of the controllable unit within the i-th distributed power source at time t; refers to the adjustment power of the energy storage within the i-th distributed power source at time t; refers to the adjustment power of the interruptible load within the i-th distributed power source at time t; refers to the actual output of the uncontrollable unit within the i-th distributed power source at time t; refers to the planned output of the uncontrollable unit within the i-th distributed power source at time t; P t rtRefers to the transaction electricity volume of the i-th distributed power source as a whole in the real-time market at time t. After optimization, the revenue and cost of each distributed power source in the balancing market are as follows:

[0121]

[0122] In the formula Refers to the power purchase cost or power sale revenue of the distributed power source as a whole in the real-time market at time t; ΔP t rt Refers to the total power purchase or power sale electricity volume of the distributed power source in the real-time market at time t; Refers to the power purchase and sale electricity volume provided by the i-th distributed power source in the real-time market at time t; Refers to the power sale price in the real-time market at time t; Refers to the power purchase price in the real-time market at time t; ΔP j,t Refers to the power adjusted by the j-th distributed power source at time t; ΔP j,i,t Refers to the power that the i-th distributed power source helps the j-th distributed power source to adjust at time t. Then the overall revenue of multi-region distributed power sources participating in the power market is:

[0123]

[0124] Through the overall revenue of multi-region distributed power sources participating in the power market, adjust the output of each distributed power source to achieve self-disciplined optimal control.

[0125] The present invention conducts research on the self-disciplined optimal control method for distributed resource clusters, designs a self-disciplined decentralized system based on the consensus algorithm. In this system, each subsystem unit is equal, can complete its own control and management, without being managed by other subsystem units, and there is no master-slave relationship between them. In addition, each subsystem unit can manage itself only relying on local information and complete the task of coordinating with other subsystem units. In the self-disciplined decentralized system, each subsystem unit only exchanges information with adjacent subsystem units, and the subsystem units only maintain a loose coupling structure through data exchange. On this basis, the power regulation cost is selected as the consensus variable, and a self-disciplined decentralized control strategy for multi-region distributed power sources is proposed.

[0126] Those skilled in the art should understand that the purpose of the above description of the embodiments of the present invention is only to exemplarily illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any example given.

[0127] The embodiments of the present invention have been described above. The above description is exemplary and not exhaustive, and is also not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A self-discipline optimization control method based on aggregated resources of a microgrid group, characterized in that Including: Construct an autonomous decentralized system based on the aggregated resources of the microgrid cluster, and use the consensus algorithm to obtain the global information of the system; Based on the global information of the system, construct an autonomous decentralized control strategy for multi-region distributed power sources; Establish an intraday optimal scheduling model for distributed power sources according to the autonomous decentralized control strategy, and adjust the output of each distributed power source.

2. The self-discipline optimization control method based on the aggregated resources of the microgrid group according to claim 1, wherein, Constructing an autonomous decentralized system based on the aggregated resources of the microgrid cluster and using the consensus algorithm to obtain the global information of the system includes: Determine the output expressions of each subsystem in the autonomous decentralized system, and then determine the effective output of each subsystem; Through the consensus algorithm, the information of each subsystem is converged to the same value after multiple iterations, so as to obtain the global information of the system.

3. The self-discipline optimization control method based on aggregated resources of a microgrid group according to claim 2, wherein, The subsystem matrix of the (k + 1)-th iteration is: where n is the number of subsystem units participating in information exchange, A is the information exchange coefficient matrix, a ij is the coefficient of information exchange between subsystem i and j, N i is the set of subsystems adjacent to subsystem i; d i is the number of subsystems adjacent to subsystem i.

4. The self-discipline optimization control method based on aggregated resources of a microgrid group according to claim 2, wherein, The converged system matrix after iteration is: In the formula, X0 is the column vector composed of the initial information obtained by each subsystem unit; I is the n-dimensional column vector with all elements being 1.

5. The self-discipline optimization control method based on aggregated resources of a microgrid group according to claim 1, wherein, The autonomous decentralized control strategy for multi-region distributed power sources includes: When the actual output of the uncontrollable units inside the multi-region distributed power sources as a whole is less than the planned output, use the consensus algorithm to solve the output fluctuation values that each distributed power source needs to bear. If the adjustable cost or adjustable capacity of a certain distributed power source reaches its own limit, remove it from the network topology, and the adjacent distributed power sources will update the network state after receiving the information; When the actual output of the uncontrollable units inside the multi-region distributed power sources as a whole is greater than the planned output, absorb it according to the types of distributed energy aggregated by the distributed power sources in each region; When the multi-region distributed power sources as a whole cannot absorb the power fluctuations caused by the internal distributed energy, achieve the overall power balance through the way of purchasing and selling electricity in the balancing market.

6. The self-discipline optimization control method based on the aggregated resources of a microgrid group according to claim 5, wherein, Absorbing according to the types of distributed energy aggregated by the distributed power sources in each region includes: If the distributed power source aggregates energy storage devices, charge the energy storage according to the operating state of the energy storage devices; If the distributed power source aggregates electric vehicles, adopt a price discount method to guide the charging of electric vehicles; If the distributed power source contains small hydropower, charge the small hydropower through pumped storage; Achieving the overall power balance through the way of purchasing and selling electricity in the balancing market includes: The power purchase cost of the distributed power source in the balancing market is borne by each distributed power source with a power deficit in proportion to the deficit electricity quantity, and the power selling income is obtained by each distributed power source corresponding to the power generation according to its power generation quantity.

7. The self-discipline optimization control method based on aggregated resources of a microgrid cluster according to claim 1, wherein, The intraday optimal scheduling model for distributed power sources: Determine the power regulation cost as the consensus variable, and calculate the power regulation cost of the i-th distributed power source at time t; Establish an objective function inside each distributed power source to minimize the average regulation cost of each distributed power source, optimize the income and cost of each distributed power source in the balancing market, and then calculate the overall income of the multi-region distributed power sources participating in the electricity market.

8. The self-discipline optimization control method based on aggregated resources of a microgrid group according to claim 7, wherein, The power regulation cost of the i-th distributed power source at time t is: The objective function is: In the formula: refers to the regulation cost of the controllable unit of the i-th distributed power source at time t; refers to the regulation cost of the energy storage of the i-th distributed power source at time t; refers to the regulation cost of the interruptible load of the i-th distributed power source at time t.

9. The self-discipline optimization control method based on aggregated resources of a microgrid cluster according to claim 7, wherein, Optimize the income and cost of each distributed power source in the balancing market through the following formula: where refers to the electricity purchase cost or electricity sales revenue of the overall distributed power source in the real-time market at time t; ΔP t rt refers to the overall electricity purchase or electricity sales volume of the distributed power source in the real-time market at time t; refers to the electricity purchase and sales volume provided by the i-th distributed power source in the real-time market at time t; refers to the electricity sales price in the real-time market at time t; refers to the electricity purchase price in the real-time market at time t; ΔP j,t refers to the power regulated by the j-th distributed power source at time t; ΔP j,i,t refers to the power that the i-th distributed power source helps the j-th distributed power source to regulate at time t.

10. The self-discipline optimization control method based on the aggregated resources of the microgrid group according to claim 7, wherein, Calculate the overall income of the multi-region distributed power sources participating in the electricity market through the following formula:

Images