An optimization dispatching method and system for weakly linked island microgrid groups

Abstract

The present invention relates to the technical field of microgrid optimization and scheduling, and discloses an optimization and scheduling method and system for a weakly linked island microgrid group. A typical electrical system model including a plurality of independent island microgrid units is constructed, and a communication and distributed control architecture based on a multi-agent is designed according to the typical electrical system model. Based on the architecture, through the island microgrid resource aggregation layer of the agent in the double-layer optimization and scheduling strategy, the load, photovoltaic power generation and other resources in the microgrid are aggregated with the charging characteristics of the charging pile, and the charging and discharging model is fitted to a single microgrid power function. In the upper consistency algorithm interaction layer, interaction variables are generated based on the power mutual assistance potential of each microgrid, and are transmitted to adjacent agents in real time through a communication link, and the microgrid output plan is continuously iterated and adjusted until the global optimum is reached. The power scheduling of the island microgrid group is optimized, which is of great significance to improving the robustness of the island energy system.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid optimization scheduling, and in particular to an optimization scheduling method and system for a weakly linked island microgrid group. Background Art

[0002] With the intensification of the global energy crisis and the improvement of environmental protection awareness, the development and utilization of renewable energy has become an important direction for global energy development. Due to the remote geographical location of island areas, electricity supply often relies on traditional fossil energy, which not only aggravates energy consumption, but also brings serious environmental pollution problems. Therefore, the development of island microgrids has become an effective way to solve the energy problem in island areas. Island microgrids are small power systems composed of distributed power sources, energy storage equipment, loads, control devices and protection devices, which can achieve autonomy and self-sufficiency in power supply in island areas. However, the construction and operation of island microgrids face many challenges, one of which is the difficulty of power mutual assistance between islands. Since islands are usually connected by submarine cables, this connection method has problems such as fragile links and susceptibility to environmental influences, resulting in limited power mutual assistance capabilities between island microgrids.

[0003] In order to solve the problem of power mutual assistance between island microgrids, researchers have proposed a variety of distributed optimization scheduling strategies, such as:

[0004] In the two-layer distributed energy scheduling of multi-microgrids based on cloud-edge collaboration, it studies the distributed energy scheduling problem of multi-microgrids and virtual power plants containing hybrid energy supply stations, and proposes a two-layer distributed energy scheduling framework based on cloud-edge collaboration to optimize the scheduling of heterogeneous equipment such as electric vehicles, energy storage equipment and clean energy. The upper and lower layers of the optimization framework respectively consider the energy optimization scheduling of virtual power plants and microgrids, and use the distributed optimization algorithm of the target cascade method to achieve problem decoupling, so as to achieve decentralized autonomy of virtual power plants and microgrids. However, the actual application scenario analysis of this method is insufficient.

[0005] In the optimization design of offshore island cluster microgrid system based on island interconnection, the load distribution and renewable energy reserve status are studied according to the special location of offshore island clusters, and an optimization scheduling model of offshore island cluster microgrid system based on island interconnection is established. The goal is to minimize the average daily operating cost and maximize the renewable energy consumption ratio. The improved particle swarm grey wolf hybrid algorithm is used to solve the optimization scheduling model. Compared with the traditional particle swarm algorithm, the algorithm has better convergence performance and has significant advantages in improving the renewable energy consumption ratio and reducing the system operation cost. However, this literature lacks more detailed consideration of the characteristics of different islands and cannot be used in weak link environments.

[0006] In the study of the optimal operation of island microgrid clusters taking into account demand response, an optimization method for connecting electric vehicles to island microgrid clusters was proposed. Taking demand response into consideration, an optimization model for microgrid clusters was established with the goal of optimizing the economic and environmental benefits of island microgrid clusters. The model was solved using the spider bee optimization algorithm. Although it can effectively reduce system costs and be more economical in scenarios with high penetration rates of renewable energy, it lacks more detailed consideration of the characteristics of different islands and cannot be used in weak link environments.

[0007] In a hierarchical coordination method for island microgrids based on electric-hydrogen hybrid energy storage, a hierarchical coordination method for island microgrids is proposed, with special attention paid to the electric-hydrogen hybrid energy storage system. The island microgrid consists of an AC subsystem, a DC subsystem and a hydrogen energy subsystem, and is connected through a bidirectional converter, an electrolyzer and a hydrogen fuel cell. The stratification of this method includes year-round daytime scheduling and daily day-ahead scheduling. Although it can solve the seasonal imbalance problem of renewable energy and ensure power balance, it does not consider the response strategy of the island microgrid under extreme conditions.

[0008] In the two-layer optimization scheduling method of multi-microgrids considering interactive power control and bilateral bidding transactions, it focuses on the optimal scheduling of multi-microgrid systems, especially interactive power control and bilateral bidding transactions. The system includes a microgrid central controller, a dispatching center, a buyer aggregator and a seller aggregator. Through upper and lower layer optimization, interactive power control of multi-microgrids and distribution networks is realized. At the same time, a bilateral bidding transaction strategy is adopted to achieve power trading. Although it can achieve the purpose of reducing the impact of multi-microgrid grid connection on the operation of the distribution network and reasonably allocating power generation resources of the multi-microgrid system, it does not consider the communication and coordination challenges of interactive power control between multi-microgrids in practical applications.

[0009] In the distributed control method and system of island microgrid based on multi-agent communication network, the fully distributed control of the microgrid is realized by constructing a communication network and designing a weight matrix. Although it can ensure that the controllable DG of the microgrid outputs power in proportion to the maximum power generation power and maintain the power balance of the system, it does not give sufficient consideration to the stability and security of the communication network, and does not consider verification in actual weak link scenarios.

[0010] In the deep learning multi-agent microgrid collaborative control method based on dual neural networks, it uses reinforcement learning framework training to find the optimal Q-value network and realize frequency deviation adjustment of distributed power sources. Although it can realize the collaborative control of microgrids, it does not consider how to ensure the real-time and accuracy of the deep learning model in the actual microgrid.

[0011] It can be seen that how to solve the problem that the existing technology does not fully consider the stability and robustness of the island microgrid under weak link conditions, and lacks high requirements for real-time and adaptability, has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0012] The present invention provides an optimization scheduling method and system for a weakly linked island microgrid group, which solves the problem that the prior art does not fully consider the stability and robustness of the island microgrid under weak link conditions, and lacks high requirements for real-time performance and adaptability.

[0013] In order to solve the above technical problems, the present invention provides an optimization scheduling method for a weakly linked island microgrid group, comprising:

[0014] Constructing a typical electrical system model of an island microgrid group, wherein the typical electrical system model includes a plurality of independent island microgrid units;

[0015] An intelligent agent is provided for each island microgrid unit, and each island microgrid unit is connected through a semi-mesh topology structure to obtain a weak link island microgrid group communication model;

[0016] Based on the weak-link island microgrid group communication model, a weak-link island microgrid group optimization scheduling model is constructed, and the weak-link island microgrid group optimization scheduling model includes an intra-island microgrid resource aggregation layer and an inter-island consistency algorithm interaction layer;

[0017] In the intra-island microgrid resource aggregation layer, the power of the island microgrid units is aggregated by each of the intelligent agents to obtain the power mutual assistance potential interval and the power resource function, and in the inter-island consistency algorithm interaction layer, the power mutual assistance potential interval and the power resource function of the island microgrid units are analyzed by each of the intelligent agents to obtain the interaction variables;

[0018] Each of the intelligent agents receives the interaction variables sent by the adjacent intelligent agents, and sends the interaction variables of the island microgrid unit to which it belongs to the adjacent intelligent agents, so as to update the optimized dispatching output plan and interaction variables of the island microgrid unit to which it belongs;

[0019] The updating process of the interactive variables is iterated until all consistency protocols converge, and the optimal output plan for the optimized scheduling of the island microgrid group is obtained.

[0020] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0021] (1) A typical electrical system model containing multiple independent island microgrid units is constructed, and a multi-agent based communication and distributed control architecture is designed based on the typical electrical system model. Based on this architecture, the load, photovoltaic power generation and other resources in the microgrid are aggregated with the charging characteristics of the charging piles through the agent in the island microgrid resource aggregation layer in the two-layer optimization scheduling strategy, and the charging and discharging model is fitted into a single microgrid power function. In the upper consistency algorithm interaction layer, interaction variables are generated based on the power mutual assistance potential of each microgrid, and transmitted to adjacent agents in real time through communication links. The microgrid output plan is continuously iterated and adjusted until the global optimality is achieved. The power scheduling of the island microgrid group is optimized, which is of great significance to improving the robustness of the island energy system.

[0022] (2) In response to the weak link problem unique to island microgrid clusters, this application proposes a distributed two-layer optimization scheduling strategy. Through a multi-agent consensus algorithm, it achieves stable operation in the event of communication link interruption. It does not rely on a centralized control structure and can respond more flexibly to link failures and dynamically changing energy demands, enhancing the robustness of the system. It is suitable for potential interruption threat scenarios in the context of weak-link island microgrid clusters, to ensure safe, stable resource utilization and optimal operation of island microgrid clusters in weak link interruption scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 It is a flow chart of an optimization scheduling method for a weakly linked island microgrid group provided by a certain embodiment of the present invention;

[0025] Figure 2 is a flowchart of step S1 provided in one embodiment of the present invention;

[0026] Figure 3 It is an overall architecture diagram of a typical electrical system model provided by an embodiment of the present invention;

[0027] Figure 4 is a flowchart of step S2 provided by an embodiment of the present invention;

[0028] Figure 5 is a diagram of a process of constructing an inter-island communication link provided by an embodiment of the present invention;

[0029] Figure 6 A microgrid group and a semi-mesh topology connection architecture diagram provided by an embodiment of the present invention;

[0030] Figure 7 is a diagram of a generation process of an interaction variable provided by a certain embodiment of the present invention;

[0031] Figure 8 is a flowchart of step S5 provided by an embodiment of the present invention;

[0032] Fig. 9 is a flowchart of step S6 provided in one embodiment of the present invention;

[0033] Fig.10 is an updating process diagram of an optimized dispatching output plan provided by a certain embodiment of the present invention;

[0034] Fig.11 It is a distributed two-layer optimization scheduling strategy diagram of an island microgrid group provided by a certain embodiment of the present invention;

[0035] Fig.12 It is a structural diagram of an optimization dispatching system for a weakly linked island microgrid group provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0038] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0039] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0040] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides an optimization scheduling method for a weakly linked island microgrid group, comprising:

[0041] S1. Construct a typical electrical system model of an island microgrid group, wherein the typical electrical system model includes multiple independent island microgrid units; specifically, the typical electrical system model includes a distribution-micro transformer, a microgrid line grid, a diesel generator, a distributed photovoltaic system, a centralized energy storage system, a distributed household energy storage system, an orderly charging pile and an island load user;

[0042] Wherein, step S1 is as follows: Figure 2 As shown, including:

[0043] S11. When the distribution networks connected to the plurality of island microgrid units are at the same voltage level, the island microgrid units are connected in parallel via a common bus;

[0044] S12. When the distribution networks connected to the plurality of island microgrid units are not at the same voltage level, the island microgrid units at the distribution network of the same voltage level are connected in parallel through a common bus, and the distribution networks are connected through a connecting transformer to perform voltage level conversion;

[0045] S13. According to the characteristics of each island, the load, new energy capacity and power generation power of each island microgrid unit are configured differently to obtain the typical electrical system model.

[0046] Specifically, the typical electrical system model of the island microgrid group is composed of several equipment unit models. The overall architecture of the typical electrical system model is as follows: Figure 3 As shown in the figure, according to the voltage level of the distribution network connected to the island microgrid unit, each island microgrid unit is connected in a corresponding manner. Due to the influence of different island geographical environments, climate factors, number of users and other characteristics, there are differences in load levels, new energy installation capacity and maximum power generation between island microgrid units. According to their configuration characteristics, they can be divided into three types: main load microgrid, secondary load microgrid and main resource microgrid. The classification basis is as follows:

[0047] Main load microgrid: There are a large number of residential and commercial users on the island, the orderly charging pile installation capacity and utilization rate are high, the overall load level of the microgrid is high, and the installation capacity of the photovoltaic power generation system and the energy storage system is low. Under most normal operating conditions, the island load cannot be independently supplied by local distributed energy, which is an under-power operation state where supply is less than demand;

[0048] Secondary load microgrid: The number of residents and commercial users on the island is moderate, the installed capacity of photovoltaic power generation systems and energy storage systems is moderate, the overall load level of the microgrid is relatively balanced with the distributed energy generation level, and in most normal operating conditions, the island load can be independently supplied by local resources to achieve energy autonomy of the microgrid. In an emergency, power replenishment from the external grid may be required, or power support can be provided to the external grid, which is a power operation state of quasi-balanced supply and demand;

[0049] Main resource microgrid: There are fewer residents and commercial users on the island, and photovoltaic resources are abundant. The installed capacity of photovoltaic power generation systems and energy storage systems is high. The distributed energy generation level of the microgrid is greater than the overall load level. Under most normal operating conditions, there is a large amount of power redundancy and abandonment. It can provide greater power support to the external power grid, which is an over-power operation state where supply exceeds demand.

[0050] The power generation model and resource model of the typical equipment unit model in the island microgrid unit include:

[0051] The diesel generator model is represented by the following formula:

[0052]

[0053] In the formula, is the dispatched power generation of the diesel generator at time t; and They are the upper and lower limits of the power generation of the diesel generator respectively; and They are the upper and lower limit climbing power of the diesel generator respectively; is the power generation resource function of the diesel generator; , are the quadratic, linear and constant coefficients of the resource function respectively.

[0054] Photovoltaic power generation system model, which is expressed by the following formula:

[0055]

[0056] In the formula, is the dispatched power generation of the photovoltaic power generation system at time t; and They are the upper and lower limits of the photovoltaic power generation system, where the upper limit of the power generation is related to the photovoltaic panel equipment parameters and light conditions, taking hourly scheduling as an example; X is the difference in sunlight intensity on a typical day; S is the area of ​​the photovoltaic array; is the photoelectric efficiency; is the power generation resource function of the photovoltaic power generation system; , and are the quadratic, linear and constant coefficients of the resource function respectively.

[0057] The energy storage system model is expressed by the following formula:

[0058]

[0059] In the formula, is the dispatching charging and discharging power of the energy storage system at time t; and They are the upper and lower limits of charging and discharging power of the energy storage system respectively; is the remaining capacity of the energy storage system at time t; and They are the upper and lower capacity limits of the energy storage system respectively; is the charging and discharging resource function of the energy storage system; , and are the quadratic, linear and constant coefficients of the resource function respectively.

[0060] The orderly charging pile model is expressed by the following formula:

[0061]

[0062] In the formula, is the downward charging power of the orderly charging pile at time t; and They are the upper and lower power limits of the orderly charging pile charging power reduction; It is the function of downward charging resource for orderly charging piles; , and are the quadratic, linear and constant coefficients of the resource function respectively.

[0063] This application uses different voltage levels to connect island microgrid units in different ways, which not only ensures the interconnectivity of the system, but also retains the original voltage level characteristics of each distribution network, and enhances the scalability of the system; according to the characteristics of each island (such as geographical location, climatic conditions, load demand, etc.), the load, new energy capacity and power generation power of the island microgrid unit are differentiated to better match the actual situation of each island, improve the utilization rate of new energy, reduce energy waste, balance the supply and demand relationship of each island microgrid unit, reduce dependence on traditional power grids, and improve the self-sufficiency of the system.

[0064] S2, setting an intelligent agent for each island microgrid unit, and connecting each island microgrid unit through a semi-mesh topology structure to obtain a weak link island microgrid group communication model;

[0065] In one embodiment, step S2 is as follows: Figure 4 As shown, including:

[0066] S21. Deploy an intelligent entity for each island microgrid unit to serve as the downlink of the energy management system of the island microgrid unit, and connect to the various energy equipment terminals and smart meters of the island microgrid unit to which it belongs by wireless means; specifically, in order to realize the distributed optimization scheduling of the island microgrid group, it is necessary to build a communication architecture that supports the local information collection, local information interaction and control command issuance of the island microgrid unit. To this end, the present application deploys an intelligent entity as a microgrid energy management system for each island microgrid unit, and its downlink is connected to the various energy equipment terminals and smart meters in the microgrid by RS485, optical fiber or wireless communication. At the same time, its two-way communication can realize the collection of local electrical equipment and the issuance of optimization scheduling instructions.

[0067] S22, constructing communication links between intelligent entities on adjacent islands through submarine cables, and connecting inter-island submarine cables of the island microgrid group through a semi-mesh topology structure to establish inter-island communication links in the island microgrid group, and generating the weak link island microgrid group communication model;

[0068] In one embodiment, the island microgrid unit includes a main load microgrid, a secondary load microgrid and a main resource microgrid; wherein the process of constructing the inter-island communication link in step S22 is as follows: Figure 5 As shown, including:

[0069] S221, using the main load microgrid as the starting end of the weak link island microgrid group communication model, directly connecting to the land power grid through the inter-island submarine cable, and directly connecting each of the main load microgrids to at least one of the main resource microgrids;

[0070] S222. The secondary load microgrid is set as the middle or end of the weak link island microgrid group communication model, and is connected in a semi-mesh topology structure through the inter-island submarine cable to obtain an inter-island communication link in the island microgrid group.

[0071] Specifically, the present application connects adjacent island intelligent entities through submarine cables to build communication links and support information interaction for power coordination and mutual assistance between islands. The inter-island submarine cable connection of the island microgrid group is constructed using a semi-mesh topology structure, which reduces the installation, operation and maintenance costs of the submarine cable compared to the full mesh topology structure.

[0072] First, the main load microgrid is set at the starting end of the electrical system of the island microgrid group, that is, it is directly connected to the land power grid. Under normal operation, the main grid power, local distributed resources and adjacent island microgrid unit resources can be flexibly dispatched for load supply. However, in the weak link interruption scenario, a power interruption with the land power grid may occur, and the island microgrid group enters the state of isolated operation. At this time, it is urgent to dispatch the distributed resources of the adjacent islands to ensure load supply. Therefore, in order to ensure operational stability, each main load microgrid should be directly connected to at least one main resource microgrid for topological redundancy.

[0073] Next, the secondary load microgrid is usually located in the middle or end of the electrical system of the island microgrid group. In a weak link scenario, it may face electrical and communication interruptions with other island microgrid units. However, since its resource configuration is in a quasi-balanced state of supply and demand, it can achieve short-term local energy autonomy, so there is no need for redundant links with the main resource microgrid.

[0074] In this way, the inter-island communication link in the island microgrid group is formed, and combined with the communication link between the intelligent bodies of adjacent islands constructed by the submarine cable, a weak link island microgrid group communication model can be obtained; among them, the microgrid group and its semi-mesh topology link architecture are as follows Figure 6 As shown, Figure 6 It consists of 4 island microgrid units (1 main load microgrid, 1 secondary load microgrid and 2 main resource microgrids). Compared with the traditional full connection or star structure, the semi-mesh topology adopted in this application is more flexible and adaptable, allowing the system to maintain a certain communication capability when some connections are interrupted, thereby improving the robustness of the entire system.

[0075] S3. Based on the weak-link island microgrid group communication model, a weak-link island microgrid group optimization scheduling model is constructed. The weak-link island microgrid group optimization scheduling model includes an intra-island microgrid resource aggregation layer and an inter-island consistency algorithm interaction layer; specifically, the communication architecture of the island microgrid group is the basis for constructing a distributed two-layer optimization scheduling architecture, which includes multiple island microgrid units, each unit has a complete power generation, energy storage, distribution and control system, and at the same time, each unit interacts and shares information through a communication network. The communication architecture of the island microgrid group needs to meet the requirements of real-time, reliability and scalability to ensure the coordinated operation and optimized scheduling of the microgrid group. This application is based on the island microgrid group communication architecture, and comprehensively considers the internal characteristics and external interaction requirements of the island microgrid units to construct a distributed two-layer optimization scheduling architecture for the microgrid group.

[0076] S4, in the island microgrid resource aggregation layer, the power of the island microgrid unit is aggregated by each of the intelligent agents to obtain the power mutual assistance potential interval and power resource function, and in the inter-island consistency algorithm interaction layer, the power mutual assistance potential interval and power resource function of the island microgrid unit are analyzed by each of the intelligent agents to obtain the interaction variable;

[0077] Specifically, the present application is responsible for resource aggregation and optimal scheduling within each island microgrid unit through the island microgrid resource aggregation layer, and optimizes the configuration of local microgrid's power generation, energy storage, load and other resources through intelligent algorithms to achieve efficient operation and supply and demand balance within the microgrid; in the lower-level island microgrid resource aggregation layer, each intelligent agent aggregates and generates the power mutual assistance potential interval of the local microgrid based on the collected load power data, energy terminal equipment operation data and real-time environmental condition data, etc., which is used to coordinate other island microgrid units to provide or receive mutual assistance power to them; at the same time, the overall power resource function of an island microgrid unit is fitted to achieve global optimization in the upper-level consistency protocol interaction.

[0078] The inter-island consistency algorithm interaction layer is responsible for the information exchange and execution of the consistency protocol between each island microgrid unit; the scheduling results and real-time data of each unit are summarized and analyzed through the communication network to achieve cross-island collaborative scheduling and optimization; in the inter-island consistency protocol interaction layer, each intelligent agent analyzes the power mutual aid potential interval and resource function of the local island microgrid unit, generates the interaction variables of the consistency protocol, and transmits the data packet to the intelligent agent directly connected to it through the submarine cable. Each intelligent agent updates the microgrid power generation plan and interaction variables according to the received data, and iterates the interaction until all consistency protocols converge, that is, to achieve distributed optimization scheduling of island microgrid groups under the optimal global resource utilization.

[0079] In one embodiment, the power aggregation process in step S4 includes:

[0080] In the island microgrid resource aggregation layer, the power mutual aid potential interval is obtained by aggregating the sum of the power generation of all distributed resources in the island microgrid unit through each of the intelligent agents; and the power generation power of all distributed resources in the island microgrid unit is proportionally allocated according to the real-time maximum power generation power of each energy device terminal through each of the intelligent agents to obtain the power resource function; wherein,

[0081] The power mutual assistance potential range of the island microgrid unit can be calculated by the following formula:

[0082]

[0083] In the formula, and They are the upper and lower ranges of the power mutual assistance potential of island microgrid unit i; and are the sum of the maximum power generation and the sum of the minimum power generation of all distributed resources in the microgrid, respectively; is the sum of the loads of the microgrid; ER is the set of distributed resources in the microgrid, including the energy storage system set , photovoltaic power generation system collection , orderly charging pile collection and diesel generator set wait.

[0084] The power mutual assistance potential of island microgrid units includes the following three types:

[0085] Underpower status: ,At this time, the maximum power generation power of all distributed resources in the island microgrid unit cannot meet the local power load demand, and the main grid / other island microgrid units need to provide power supply;

[0086] Quasi-equilibrium state: ,At this time, the distributed resources in the island microgrid unit can meet the local power load demand through optimized scheduling, and can achieve energy autonomy of the island microgrid unit, and can also provide and receive a small amount of power to other island microgrid units;

[0087] Over power state: At this time, the minimum power generation power of all distributed resources in the island microgrid unit has exceeded the local power load demand, and the main grid / other island microgrid units need to receive power supply.

[0088] The above power mutual aid potential interval information can be used as a constraint condition for the output plan of the island microgrid unit in the consistency protocol interaction.

[0089] The fitting of the overall power-resource function of the island microgrid unit is to distribute the output power of the island microgrid unit in proportion to the real-time maximum power generation power of each energy equipment terminal to ensure the continuity of the resource function within the definition domain. The fitted power resource function is expressed by the following formula:

[0090]

[0091]

[0092] In the formula, is the power resource function fitted for the entire island microgrid unit i; is the proportion of the power generation of energy device j to the overall power generation processing plan of the microgrid; is the power generation of all distributed resources in the island microgrid unit; , and are the quadratic, linear and constant coefficients of the power resource function corresponding to energy device j, respectively.

[0093] The above proportional allocation quadratic cost single microgrid optimization function fitting strategy has the following characteristics:

[0094] Overfitting characteristics: The above strategy assumes that under any output plan, the overall power generation of the microgrid will be decomposed into each energy device in the form of proportional distribution. In the actual economic optimization scheduling, the device with the highest utilization rate of boundary resources should be used for power generation first. This leads to the fitting resource function being higher than the actual power-resource optimal function of the island microgrid unit under a lower output plan; Continuity characteristics: Due to the differences in resource characteristics between different energy devices, the actual power-resource optimal function of the island microgrid unit and its derivative are not smooth, so it is difficult to fit the curve and extract its expression coefficient (non-constant, varies with power). The power-resource suboptimal function formed by the proportional distribution method in this application can solve the problem of uneven curves, ensure that the coefficient of its resource function is an unchanging constant with simple calculation in the entire definition domain, and ensure that each energy device is reasonably distributed in the definition domain of the overall microgrid output plan to avoid the phenomenon that some devices are overloaded and some devices are underloaded.

[0095] In one embodiment, the generation process of the interaction variable in step S4 is as follows: Figure 7 As shown, including:

[0096] S41. In the inter-island consensus algorithm interaction layer, based on the power resource function of the island microgrid unit to which each of the intelligent agents belongs, a centralized optimization scheduling model is constructed by introducing the Lagrangian operator; specifically, the traditional centralized optimization scheduling problem can be expressed by the following formula:

[0097]

[0098] In the formula, It is the collection of all distributed resource devices in the island microgrid group; is the power resource function of each resource device; is the sum of the power loads of the island microgrid group; , and are the power generation of each resource equipment and its upper and lower limits. To solve the above optimization problem, this application introduces the Lagrangian operator , according to the first-order Lagrangian optimization condition, the optimization problem based on the Lagrangian function can be expressed as follows:

[0099]

[0100] In the formula, is the objective function in the initial optimization scheduling problem; is the constraint condition in the initial optimization scheduling problem. According to the KKT (Karush-Kuhn-Tucker) condition, for convex programming problems, the global (local) optimal point of the above Lagrangian function should satisfy the following relationship:

[0101]

[0102] According to the above formula, we can further deduce:

[0103]

[0104] In the formula, and They are all auxiliary variables.

[0105] According to the above conditions, under the optimal KKT conditions, the Lagrangian operator, as the consistency protocol variable between distributed resource devices in the island microgrid group, should converge to the same value under the resource optimal strategy, and its value is equal to the resource increment of each energy device, that is, the resource increment of all energy devices is the same. Under this condition, increasing / decreasing the power generation of any energy device (while reducing / increasing the power generation of any other energy device) will cause the overall cost to increase while the power generation output plan remains unchanged, resulting in a decrease in resource utilization.

[0106] According to the above conditions, the optimal strategy Lagrangian operator can be solved The value of is expressed as follows:

[0107]

[0108] The above optimal power generation plan strategy also needs to consider the upper and lower limits of the equipment power generation. Therefore, it is necessary to adjust the equipment power generation plan under the above optimal conditions, and adjust the power generation plan of the equipment that exceeds the upper and lower limits of the power generation to the upper and lower limits respectively (that is, if it exceeds the upper limit, it is adjusted to the upper limit power, and if it exceeds the lower limit, it is adjusted to the lower limit power). Incremental resource balancing calculations are performed in the remaining distributed resource devices to obtain the suboptimal power generation strategy under the power constraint conditions, and iterate in sequence until the power generation of all energy equipment under the output plan meets the constraints. At this time, the optimization scheduling scheme, that is, the centralized optimization scheduling model can be expressed as follows:

[0109]

[0110] In the formula, To optimize the dispatching plan, it can also be considered as an output plan; is the power in the output plan; It is the set of over-limit island microgrid units, that is, the set of distributed resource devices in the island microgrid group whose power generation plan exceeds the upper and lower limits.

[0111] S42, based on the connection topology information in the weak link island microgrid group communication model, converting the centralized optimization scheduling model into a distributed optimization model according to a consistency interaction protocol to generate the interaction variable;

[0112] Specifically, the above-mentioned centralized optimization scheduling problem based on Lagrangian function and the calculation and processing of its optimal strategy require a centralized control center to collect the operating data, power load data and upper and lower limits of power generation of all equipment to achieve optimization. However, in the context of weakly linked island microgrid groups, potential weak communication link problems may cause the interruption of a certain communication link or multiple communication links in a local area, resulting in the control center being unable to obtain distributed resource equipment information, and thus unable to perform optimization scheduling.

[0113] To solve this problem, this application adopts the idea of ​​distributed optimization scheduling, which can rewrite the centralized optimization problem into a distributed optimization problem based on connection topology information and consistency interaction protocol, in which the power generation power of each distributed resource only depends on the communication interaction information of local adjacent devices and local power load data for consistency optimization. The calculation method of its power generation power can be expressed by the following formula:

[0114]

[0115] In the formula, and is the interactive variable of power mutual assistance; and is the coefficient of the island microgrid group connection topology structure, where indicates that there is a direct submarine cable connection between island microgrid units n and m, otherwise Indicates that there is no direct link; represents the number of links directly connected to the island microgrid unit n; is the initial power generation. According to the above distributed power optimization scheduling method, the overall power supply and demand balance constraint of the island microgrid group can be directly met, which is specifically expressed as follows:

[0116]

[0117] Therefore, the distributed power optimization scheduling problem of island microgrid can be expressed as follows:

[0118]

[0119] This application constructs a centralized optimization scheduling model based on the power resource function of each island microgrid unit, which can comprehensively consider the power generation capacity, energy storage status and load demand of each unit to achieve optimal resource allocation; after converting it into a distributed model, each unit can be dynamically adjusted according to its own actual situation and global optimization goals, further improving resource utilization efficiency, reducing dependence on a single central control point, and reducing the risk of single point failure. Even if some communication links fail, as long as sufficient connection topology information is maintained, the system can still maintain operation through a consistent interaction protocol, thereby enhancing the robustness of the entire island microgrid group.

[0120] S5, each of the intelligent agents receives the interaction variables sent by the adjacent intelligent agents, and sends the interaction variables of the island microgrid unit to which it belongs to to the adjacent intelligent agents, so as to update the optimized dispatching output plan and interaction variables of the island microgrid unit to which it belongs;

[0121] In one embodiment, step S5 is as follows: Figure 8 As shown, including:

[0122] S51, constructing a link matrix between each of the island microgrid units according to the weak link island microgrid group communication model, determining a connection link of each of the island microgrid units according to the link matrix, and initializing the interaction variables and the Lagrangian operator;

[0123] S52. Each of the intelligent agents sends its own corresponding initialized interaction variables to the directly linked adjacent intelligent agents according to the connection link of the island microgrid unit to which it belongs, and updates the optimized scheduling output plan, interaction variables and initialized Lagrangian operator of the island microgrid unit to which it belongs according to the interaction variables sent by the received adjacent intelligent agents.

[0124] Specifically, according to the semi-mesh topology of the island microgrid group, a link matrix is ​​constructed. , and calculate the number of connection links of each microgrid , and initialize the power mutual interaction variables and Lagrangian operators of each island microgrid; each island microgrid intelligent agent sends its interaction variable data packet to its directly linked adjacent agent, and updates its power generation plan according to the received interaction variables, that is, optimizes the dispatch output plan, interaction variables and Lagrangian operator , the specific update strategy is expressed as follows:

[0125]

[0126] In the formula, and The interaction variables and Lagrangian operator The update increment during each iteration; Indicates positive and negative sign operations, that is, when the expression in the brackets is a positive value, ; When the expression in the brackets is a negative value, ; When the operation expression is 0, .

[0127] By constructing a link matrix and determining the connection links of each island microgrid unit, the present application can clearly define the communication paths between the units, thereby ensuring the efficient transmission of information in the microgrid group, reducing the redundancy and delay of information transmission, and improving the overall scheduling efficiency; by continuously updating and optimizing the scheduling output plan, interaction variables and Lagrangian operators, each unit can better adapt to changes in the system state and achieve a globally optimal or near-globally optimal scheduling plan, so as to improve the real-time and accuracy of decision-making, and help the system better respond to emergencies and rapidly changing needs.

[0128] S6, iterating the updating process of the interactive variables until all the consistency protocols converge, and obtaining the optimal output plan for the optimized scheduling of the island microgrid group;

[0129] In one embodiment, step S6 is as follows: Fig. 9 As shown, including:

[0130] S61, iterating the updating process of the interaction variables until the updated Lagrangian operators of each of the intelligent agents are the same, so as to achieve convergence of the consistency protocol, and determining the set of over-limit island microgrid units according to the power mutual assistance potential intervals of each of the island microgrid units;

[0131] S62, introducing auxiliary variables in each of the intelligent agents to perform auxiliary consistency protocol interaction, each of the intelligent agents receives the auxiliary variables sent by the adjacent intelligent agents, and sends the auxiliary variables of the island microgrid unit to which it belongs to to the adjacent intelligent agents, so as to update the auxiliary variables of the island microgrid unit to which it belongs;

[0132] S63, iterate the updating process of the auxiliary variables until the auxiliary variables of each of the intelligent agents converge to be consistent, and based on the set of off-limit island microgrid units, each non-off-limit island microgrid unit updates the optimized scheduling output plan through the updated auxiliary variables to obtain the optimized scheduling optimal output plan of the island microgrid group.

[0133] Specifically, the update process of the interaction variables is iterated until the consensus protocol converges. Under the convergence condition, the Lagrangian operator of each agent is The same, and equal to the incremental resources of the fitting resource function of each island microgrid unit; then according to the power mutual assistance potential interval of the island microgrid unit, the feasibility of each microgrid output plan under the optimal resource utilization condition is checked, and the set of over-limit island microgrid units is obtained. , and the power in the output plan of the island microgrid unit that exceeds the limit Adjust to the maximum / minimum value of the potential range, as shown in the following formula:

[0134]

[0135] Then introduce two auxiliary variables into each island agent and , used for power consistency protocol interaction of other non-over-limit microgrids, the auxiliary variables are initialized as follows:

[0136]

[0137] Subsequently, the intelligent agent of each island microgrid unit sends its auxiliary variable data packet to its directly linked adjacent agent, and updates its own auxiliary variables based on the received auxiliary variable data. The specific update rule is as follows:

[0138]

[0139] In the formula, and is the updated auxiliary variable.

[0140] The updating process of auxiliary variables is iterated until the auxiliary consistency protocol converges. In the convergence case, the auxiliary of each island agent converges to the same, as shown in the following formula:

[0141]

[0142] The Agent of each non-over-limit island microgrid unit updates its power output plan according to the convergence value of the auxiliary variable, and obtains the optimal output plan for the optimal scheduling of the island microgrid group. , as follows:

[0143] .

[0144] This application iterates the updating process of interactive variables until the Lagrangian operator of each intelligent agent is the same, which ensures that the consistency protocol in the distributed optimization algorithm reaches a convergence state so that all island microgrid units reach a consensus on the optimal scheduling, thereby achieving the global optimization goal; after reaching the convergence of the consistency protocol, the set of off-limit island microgrid units is determined according to the power mutual assistance potential interval of each island microgrid unit, which can timely discover and identify those units that may cause problems due to insufficient or excessive power; auxiliary variables are introduced in each intelligent agent to assist in the interaction of the consistency protocol, which provides the system with additional degrees of freedom, so that in processing power It is more flexible in complex situations such as over-limit. By iteratively updating auxiliary variables, the system can more accurately adjust the output plan of each unit. Based on the over-limit island microgrid unit set, each non-over-limit island microgrid unit updates its optimized scheduling output plan through the updated auxiliary variables, which can be adjusted in real time according to changes in the system status, thereby ensuring that the optimized scheduling output plan of the entire island microgrid group is always maintained at the optimal or near-optimal state. This scheme not only achieves global optimization, but also can effectively deal with potential problems such as power over-limit, greatly improving the stability and reliability of the island microgrid group, and reducing the risk of power outages caused by power imbalance or equipment failure.

[0145] In one embodiment, the updating process of the optimized dispatching output plan in step S63 is as follows: Fig.10 As shown, including:

[0146] S631, determining whether the optimal output plan of the optimized scheduling of the island microgrid group exceeds the limit;

[0147] S6321, if yes, then update the over-limit island microgrid unit set, reconstruct the auxiliary variables through the updated over-limit island microgrid unit set, and adjust the optimal output plan of the optimized scheduling; repeat the over-limit phenomenon judgment step until there is no over-limit phenomenon in the optimal output plan of the optimized scheduling of the island microgrid group;

[0148] S6322. If not, the optimized scheduling optimal output plan is returned to each corresponding intelligent agent, so that each intelligent agent decomposes and sends the optimized scheduling optimal output plan to the corresponding energy equipment terminal according to the centralized optimized scheduling model.

[0149] Specifically, check whether there is a new over-limit situation in the output plan of the new island microgrid group. If so, return to the over-limit phenomenon judgment step, readjust the output plan and construct new auxiliary variables to interact with the auxiliary consistency protocol; if not, implement the global optimal output plan optimization scheduling of the island microgrid unit; and return the microgrid output plan to the Agent, so that each intelligent body adopts centralized optimization scheduling based on Lagrangian operator, and decomposes the microgrid output plan and sends it to the corresponding energy equipment terminals to realize the closed loop of distributed optimization scheduling.

[0150] Distributed two-layer optimization dispatch strategy for island microgrid clusters Fig.11 As shown, the present application can timely discover and deal with factors that may cause system instability by judging whether there is an over-limit phenomenon in the optimized dispatching optimal output plan of the island microgrid group, thereby ensuring that the finally generated dispatching plan will not fail due to power over-limit and other problems in actual operation, thereby improving the reliability and stability of the system; after discovering the over-limit phenomenon, by updating the over-limit island microgrid unit set, reconstructing auxiliary variables and adjusting the optimized dispatching output plan, the optimization process is continuously iterated, and the global optimal solution is gradually approached, so that the finally generated dispatching plan is more in line with the actual needs and operating conditions of the system, and the optimization degree of the dispatching plan is improved; after the optimized dispatching output plan does not have an over-limit phenomenon, the dispatching plan is returned to the corresponding each intelligent body, and the dispatching plan is decomposed and distributed to each energy equipment terminal according to the centralized optimization dispatching model, which simplifies the dispatching execution process, enables each energy device to respond quickly and execute the dispatching instruction, and improves the response speed and execution efficiency of the system.

[0151] In one embodiment, in order to deal with the problem of communication link interruption caused by the destruction of submarine communication cables and other possible situations in the context of weak links on islands, a communication monitoring module is deployed on each intelligent body, which can detect weak link interruption accidents based on the reception of real-time data packets, and promptly issue alarms and quickly achieve optimized steady-state convergence by updating communication link parameters, adjusting and recalculating the consistency algorithm. The above-mentioned distributed optimization scheduling strategy can deal with potential interruption threats in the context of weak-link island microgrid groups, and ensure the safe, stable and optimal resource utilization of island microgrid groups in weak-link interruption scenarios.

[0152] In the case of weak link interruption incidents occur in the communication link When the fault occurs, the communication monitoring module in the Agent of the island microgrid unit connected to the fault submarine cable can quickly detect the loss of the opposite Agent data packet. When the packet loss time reaches a certain threshold, it is determined to be a weak link interruption accident. At this time, the communication interruption module will alarm the interruption accident to the Agent module and dynamically adjust the interaction strategy of the consistency protocol, including the following two sub-steps:

[0153] (1) The corresponding Set to 0;

[0154] (2) Connect the corresponding microgrids of the two islands and Updated to and .

[0155] Through the above-mentioned dynamic adjustment method of the consistency protocol, it is only necessary to change the hyperparameters to achieve flexible adjustment of the power mutual assistance protocol between island microgrid groups, quickly adapt to the new inter-island topology, and achieve optimized scheduling under the new topology. According to the different locations where the fault occurs, the optimized scheduling under the weak link interruption scenario can be divided into the following three types:

[0156] Global optimization scenario: When a weak link interruption scenario occurs in the redundant lines in the topology, that is, the non-faulty lines can still interconnect the island microgrid group as a whole, then after dynamic adjustment of the consistency protocol, the output plan of the microgrid group will quickly stabilize at the same level as before the fault, thereby minimizing the impact of weak link interruption accidents on the operational stability and economy of the island microgrid group.

[0157] Local optimization scenario: When the weak link interruption scenario occurs in the middle line of the topology structure, that is, the island microgrid group is divided into two independent new microgrid groups, after dynamic adjustment of the consistency protocol, the two independent new microgrid groups will achieve a new power mutual assistance state through resource scheduling within their respective local areas, and realize sub-optimal optimization scheduling in their respective microgrid groups, thereby fully reducing the impact of weak link interruption accidents on the operating stability and resource utilization of the island microgrid group.

[0158] Forced island optimization scenario: When the weak link interruption scenario occurs at the end line of the topology structure, that is, the island microgrid group is divided into an independent island microgrid and a new microgrid group, after the dynamic adjustment of the consistency protocol, the independent island microgrid can only achieve optimal resource utilization in the energy autonomy scenario through local centralized optimization scheduling, while the new microgrid group will achieve a new power mutual assistance state through resource scheduling within the local area, and realize this optimized scheduling in the microgrid group. In this scenario, although the terminal microgrid is forced to enter the island operation state, its communication detection module can quickly identify the island accident, adjust the consistency protocol in time, realize the optimal allocation of resources for the island operation, and reduce the impact of the weak link interruption accident on the stability of the island microgrid group operation; at the same time, the new microgrid group achieves a new power mutual assistance state through resource scheduling within the local area, and realizes sub-optimal scheduling.

[0159] The present invention improves the scheduling efficiency of renewable energy and energy storage equipment through real-time monitoring and dynamic adjustment strategies of intelligent agents, can respond more accurately to changes in electricity supply and demand, and achieve more refined energy management; this data-driven approach has higher adaptability and accuracy than traditional rule-based or static model-based approaches.

[0160] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0161] In another embodiment, if Fig.12 As shown, the second aspect of the present invention provides an optimization dispatching system for a weakly linked island microgrid group, comprising:

[0162] A system model building module 10 is used to build a typical electrical system model of an island microgrid group, wherein the typical electrical system model includes a plurality of independent island microgrid units;

[0163] The communication model building module 20 is used to set an intelligent agent for each island microgrid unit, and connect each island microgrid unit through a semi-mesh topology structure to obtain a weak link island microgrid group communication model;

[0164] A scheduling model building module 30 is used to build a weak-link island microgrid group optimization scheduling model based on the weak-link island microgrid group communication model, and the weak-link island microgrid group optimization scheduling model includes an intra-island microgrid resource aggregation layer and an inter-island consistency algorithm interaction layer;

[0165] The function variable generation module 40 is used to aggregate the power of the island microgrid units through each of the intelligent agents in the island microgrid resource aggregation layer to obtain the power mutual assistance potential interval and the power resource function, and to analyze the power mutual assistance potential interval and the power resource function of the island microgrid units through each of the intelligent agents in the inter-island consistency algorithm interaction layer to obtain the interaction variable;

[0166] The planned variable updating module 50 is used for each of the intelligent agents to receive the interactive variables sent by the adjacent intelligent agents, and send the interactive variables of the island microgrid unit to which it belongs to to the adjacent intelligent agents, so as to update the optimized dispatching output plan and interactive variables of the island microgrid unit to which it belongs;

[0167] The update iteration module 60 is used to iterate the update process of the interactive variables until all consistency protocols converge to obtain the optimal output plan for the optimized scheduling of the island microgrid group.

[0168] It should be noted that each module in the above-mentioned optimization scheduling system based on a weakly linked island microgrid group can be fully or partially implemented through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of an optimization scheduling system for a weakly linked island microgrid group, please refer to the definition of an optimization scheduling method for a weakly linked island microgrid group in the above text. The two have the same functions and effects, which will not be repeated here.

[0169] In summary, the present invention relates to the technical field of microgrid optimization and scheduling, and discloses an optimization scheduling method and system for a weakly linked island microgrid group. A typical electrical system model including multiple independent island microgrid units is constructed, and a communication and distributed control architecture based on a multi-agent is designed according to the typical electrical system model. Based on the architecture, through the island microgrid resource aggregation layer of the agent in the two-layer optimization and scheduling strategy, the load, photovoltaic power generation and other resources in the microgrid are aggregated with the charging characteristics of the charging pile, and the charging and discharging model is fitted to a single microgrid power function. In the upper consistency algorithm interaction layer, interaction variables are generated based on the power mutual assistance potential of each microgrid, and are transmitted to adjacent agents in real time through communication links. The microgrid output plan is continuously iterated and adjusted until the global optimum is reached. The power scheduling of the island microgrid group is optimized, which is of great significance to improving the robustness of the island energy system.

[0170] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0171] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. An optimization dispatching method for a weakly linked island microgrid group, characterized in that: include: Constructing a typical electrical system model of an island microgrid group, wherein the typical electrical system model includes a plurality of independent island microgrid units; An intelligent agent is provided for each island microgrid unit, and each island microgrid unit is connected through a semi-mesh topology structure to obtain a weak link island microgrid group communication model; Based on the weak-link island microgrid group communication model, a weak-link island microgrid group optimization scheduling model is constructed, and the weak-link island microgrid group optimization scheduling model includes an intra-island microgrid resource aggregation layer and an inter-island consistency algorithm interaction layer; In the intra-island microgrid resource aggregation layer, the power of the island microgrid units is aggregated by each of the intelligent agents to obtain the power mutual assistance potential interval and the power resource function, and in the inter-island consistency algorithm interaction layer, the power mutual assistance potential interval and the power resource function of the island microgrid units are analyzed by each of the intelligent agents to obtain the interaction variables; Each of the intelligent agents receives the interaction variables sent by the adjacent intelligent agents, and sends the interaction variables of the island microgrid unit to which it belongs to the adjacent intelligent agents, so as to update the optimized dispatching output plan and interaction variables of the island microgrid unit to which it belongs; Iterate the updating process of the interactive variables until all the consistency protocols converge, and obtain the optimal dispatching and optimal output plan of the island microgrid group; In the island microgrid resource aggregation layer, the power of the island microgrid units is aggregated by each of the intelligent agents to obtain the power mutual assistance potential interval and power resource function, including: In the island microgrid resource aggregation layer, the power mutual aid potential interval is obtained by aggregating the sum of the power generation of all distributed resources in the island microgrid unit through each of the intelligent agents; and the power resource function is obtained by proportionally allocating the power generation of all distributed resources in the island microgrid unit according to the real-time maximum power generation of each energy device terminal through each of the intelligent agents; In the inter-island consensus algorithm interaction layer, each of the intelligent agents analyzes the power mutual assistance potential interval and power resource function of the island microgrid unit to which it belongs, and obtains the interaction variables, including: In the inter-island consensus algorithm interaction layer, based on the power resource function of the island microgrid unit to which each of the intelligent agents belongs, a centralized optimization scheduling model is constructed by introducing the Lagrangian operator; Based on the connection topology information in the weak-link island microgrid group communication model, the centralized optimization scheduling model is converted into a distributed optimization model according to a consistent interaction protocol to generate the interaction variables.

2. The optimization scheduling method for a weakly linked island microgrid group according to claim 1 is characterized in that: The typical electrical system model includes distribution-micro transformer, microgrid line grid, diesel generator, distributed photovoltaic system, centralized energy storage system, distributed household energy storage system, orderly charging pile and island load users; The typical electrical system model for constructing the island microgrid group includes: When the distribution networks connected to the plurality of island microgrid units are at the same voltage level, the island microgrid units are connected in parallel via a common bus; When the distribution networks connected to the multiple island microgrid units are not at the same voltage level, the island microgrid units under the distribution network of the same voltage level are connected in parallel through a common bus, and each of the distribution networks is connected through a connecting transformer to perform voltage level conversion; According to the characteristics of each island, the load, new energy capacity and power generation power of each island microgrid unit are configured differently to obtain the typical electrical system model.

3. The optimization scheduling method for a weakly linked island microgrid group according to claim 1 is characterized in that: The method of setting an intelligent agent for each island microgrid unit and connecting each island microgrid unit through a semi-mesh topology structure to obtain a weak link island microgrid group communication model includes: Deploy an intelligent agent for each island microgrid unit to serve as a downlink of the energy management system of the island microgrid unit and connect to each energy device terminal and smart meter of the island microgrid unit to which it belongs by wireless means; Communication links between intelligent entities on adjacent islands are constructed through submarine cables, and inter-island submarine cables of the island microgrid group are connected through a semi-mesh topology structure to establish inter-island communication links in the island microgrid group and generate the weak-link island microgrid group communication model.

4. The optimization scheduling method for a weakly linked island microgrid group according to claim 3 is characterized in that: The island microgrid unit includes a main load microgrid, a secondary load microgrid and a main resource microgrid; The inter-island submarine cables connecting the island microgrid group through the semi-mesh topology structure to establish an inter-island communication link in the island microgrid group include: The main load microgrid is used as the starting end of the weak link island microgrid group communication model, and is directly connected to the land power grid through the inter-island submarine cable, and each of the main load microgrids is directly connected to at least one of the main resource microgrids; The secondary load microgrid is set as the middle or end of the weak link island microgrid group communication model, and is connected in a semi-mesh topology through the inter-island submarine cable to obtain the inter-island communication link in the island microgrid group.

5. The optimization scheduling method for a weakly linked island microgrid group according to claim 1 is characterized in that: Each of the intelligent agents receives the interaction variables sent by the adjacent intelligent agents, and sends the interaction variables of the island microgrid unit to which it belongs to the adjacent intelligent agent, so as to update the optimized dispatching output plan and interaction variables of the island microgrid unit to which it belongs, including: According to the weak link island microgrid group communication model, a link matrix between each of the island microgrid units is constructed, and a connection link of each of the island microgrid units is determined according to the link matrix, and the interaction variable and the Lagrangian operator are initialized; Each of the intelligent agents sends its own corresponding initialized interaction variables to the directly linked adjacent intelligent agents according to the connection link of the island microgrid unit to which it belongs, and updates the optimized scheduling output plan, interaction variables and initialized Lagrangian operator of the island microgrid unit to which it belongs according to the interaction variables sent by the received adjacent intelligent agents.

6. The optimization scheduling method for a weakly linked island microgrid group according to claim 5 is characterized in that: The updating process of the iterative interactive variables, until all consistency protocols converge, obtains the optimal dispatching and optimal output plan of the island microgrid group, including: Iterate the updating process of the interaction variables until the updated Lagrangian operators of each of the intelligent agents are the same, so as to achieve convergence of the consistency protocol, and determine the set of off-limit island microgrid units according to the power mutual assistance potential intervals of each of the island microgrid units; Auxiliary variables are introduced into each of the intelligent agents to perform auxiliary consistency protocol interaction. Each of the intelligent agents receives the auxiliary variables sent by the adjacent intelligent agents, and sends the auxiliary variables of the island microgrid unit to which it belongs to to the adjacent intelligent agents, so as to update the auxiliary variables of the island microgrid unit to which it belongs; The updating process of the auxiliary variables is iterated until the auxiliary variables of each of the intelligent agents converge to be consistent, and based on the set of off-limit island microgrid units, each non-off-limit island microgrid unit updates the optimized scheduling output plan through the updated auxiliary variables to obtain the optimized scheduling optimal output plan of the island microgrid group.

7. The optimization scheduling method of a weakly linked island microgrid group according to claim 6 is characterized in that: The method of updating the optimized dispatching output plan by using the updated auxiliary variables to obtain the optimized dispatching optimal output plan of the island microgrid group includes: Determine whether the optimal output plan of the optimized scheduling of the island microgrid group exceeds the limit; If so, update the over-limit island microgrid unit set, reconstruct the auxiliary variables through the updated over-limit island microgrid unit set, and adjust the optimal output plan of the optimized scheduling; repeat the over-limit phenomenon judgment step until there is no over-limit phenomenon in the optimal output plan of the optimized scheduling of the island microgrid group; If not, the optimized scheduling optimal output plan is returned to each corresponding intelligent agent, so that each intelligent agent decomposes and sends the optimized scheduling optimal output plan to the corresponding energy equipment terminal according to the centralized optimized scheduling model.

8. An optimization dispatching system for weakly linked island microgrid groups, characterized in that: include: A system model building module, used to build a typical electrical system model of an island microgrid group, wherein the typical electrical system model includes a plurality of independent island microgrid units; A communication model building module is used to set an intelligent agent for each island microgrid unit, and connect each island microgrid unit through a semi-mesh topology structure to obtain a weak link island microgrid group communication model; A scheduling model construction module, used to construct a weak-link island microgrid group optimization scheduling model based on the weak-link island microgrid group communication model, wherein the weak-link island microgrid group optimization scheduling model includes an intra-island microgrid resource aggregation layer and an inter-island consistency algorithm interaction layer; A function variable generation module is used to aggregate the power of the island microgrid units through each of the intelligent agents in the island microgrid resource aggregation layer to obtain the power mutual assistance potential interval and power resource function, and to analyze the power mutual assistance potential interval and power resource function of the island microgrid units through each of the intelligent agents in the inter-island consistency algorithm interaction layer to obtain interaction variables; A planned variable updating module is used for each of the intelligent agents to receive the interactive variables sent by the adjacent intelligent agents, and send the interactive variables of the island microgrid unit to which it belongs to to the adjacent intelligent agents, so as to update the optimized dispatching output plan and interactive variables of the island microgrid unit to which it belongs; An update iteration module is used to iterate the update process of the interactive variables until all consistency protocols converge to obtain the optimal output plan for the optimized scheduling of the island microgrid group; In the island microgrid resource aggregation layer, the power of the island microgrid units is aggregated by each of the intelligent agents to obtain the power mutual assistance potential interval and power resource function, including: In the island microgrid resource aggregation layer, the power mutual aid potential interval is obtained by aggregating the sum of the power generation of all distributed resources in the island microgrid unit through each of the intelligent agents; and the power resource function is obtained by proportionally allocating the power generation of all distributed resources in the island microgrid unit according to the real-time maximum power generation of each energy device terminal through each of the intelligent agents; In the inter-island consensus algorithm interaction layer, each of the intelligent agents analyzes the power mutual assistance potential interval and power resource function of the island microgrid unit to which it belongs, and obtains the interaction variables, including: In the inter-island consensus algorithm interaction layer, based on the power resource function of the island microgrid unit to which each of the intelligent agents belongs, a centralized optimization scheduling model is constructed by introducing the Lagrangian operator; Based on the connection topology information in the weak-link island microgrid group communication model, the centralized optimization scheduling model is converted into a distributed optimization model according to a consistent interaction protocol to generate the interaction variables.

Images