Multi-agent-based multi-level cooperative control method for integrated energy system

Abstract

The invention belongs to the technical field of integrated energy systems, and particularly relates to a multi-agent-based multi-level cooperative control method for an integrated energy system. Comprising the following steps: constructing a multi-level intelligent agent cooperative control framework of the comprehensive energy system; carrying out mass distributed flexible resource virtual aggregation controllable-adjustable characteristic analysis and modeling; establishing an efficient and rapid cooperative control model of the agent layer distributed controllable resources; according to the method, intelligent agents which are in one-to-one correspondence with physical system level nodes and have autonomous intelligent decision making functions are established, and a cross-platform integrated energy system multi-level intelligent agent cooperative control framework which adapts to distributed resource efficient access is constructed; analyzing a space-time uncertainty change rule, and constructing a mass distributed source-load-storage controllable resource virtual aggregation'controllable-adjustable 'characteristic model; on the basis, a control normal form of hierarchical interaction and autonomous cooperation among edge, edge and end, and end is established, and efficient suppression of power flow fluctuation under'plug and play 'energy impact is realized in real time.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated energy systems, and specifically relates to a multi-level collaborative control method for integrated energy systems based on multiple intelligent agents. Background Art

[0002] With the rapid economic and social development of my country, total energy demand has increased dramatically, and the contradiction between energy supply and demand has become increasingly prominent. In recent years, energy issues have become a hot topic of discussion across various sectors. The emergence of concepts such as integrated energy systems and the recent promotion of the "Internet + Smart Energy" concept in my country have ushered in a wave of energy reform, providing a new perspective for energy analysis and promoting integration and innovation across various fields and disciplines. Integrated energy systems integrate energy production, transmission, and consumption, meeting the diverse energy needs of the system, including electricity, cooling, heating, and gas. They contribute to the implementation of supply-side reforms in the energy industry and enable its transition towards a low-carbon, efficient, and sustainable future. While integrated energy systems offer numerous advantages, they also present unprecedented challenges and challenges. The hybrid AC / DC system significantly increases the complexity of the power grid structure, and the frequent changes in the operating modes of various energy networks lead to greater uncertainty in system operation. The flexible integration of emerging loads, such as electric vehicles, smart buildings, and smart homes, and their two-way interaction with the integrated energy system (IES), have increased the complexity of system operation. Currently, unified modeling, multi-energy complementary scheduling, and multi-level coordinated control of IESs face significant challenges. IESs are the primary carrier of a new type of power system, integrating multiple energy sources, including electricity, gas, heating / cooling, and hydrogen. Distributed energy sources within IESs can be autonomously plugged and played as long as they meet access standards. However, this can generate disruptive power flow fluctuations distributed across multiple time and space, jeopardizing system operation. Within this open, plug-and-play access environment, it is crucial to flexibly activate and coordinate the control of massive distributed controllable resources to rapidly mitigate energy fluctuations and ensure high-quality system energy supply. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a multi-level collaborative control method for an integrated energy system based on multi-agents.

[0004] The object of the present invention is achieved as follows: a multi-level collaborative control method for an integrated energy system based on a multi-agent system, comprising:

[0005] Construct a multi-level intelligent collaborative control architecture for integrated energy systems;

[0006] Conduct controllable and adjustable feature analysis and modeling for massive distributed flexible resource virtual aggregation;

[0007] Establish an efficient and rapid collaborative control model for distributed controllable resources at the agent layer.

[0008] Furthermore, the construction of a multi-level intelligent collaborative control architecture for an integrated energy system includes:

[0009] Analyze the dynamic characteristics of each energy flow aggregation node in the integrated energy system at different time and spatial scales, determine the electrical coupling and information coupling relationship between each node, and further use topological graph theory to describe the physical and information interaction relationship between each energy flow aggregation node;

[0010] For each physical system level node in the integrated energy system, based on the dynamic characteristics of each node, an intelligent agent with intelligent environmental perception, intelligent knowledge acquisition, and autonomous intelligent decision-making is constructed. Based on the physical information interaction relationship between each node, complex system and networked control theory are used to build physical information associations between intelligent agents.

[0011] By analyzing the functional characteristics of different intelligent agents and the logical relationships between them, a scheduling system with a three-layer architecture of terminal layer, agent layer and master layer is constructed based on the idea of ​​hierarchical coordination and peer collaboration.

[0012] Furthermore, the construction of a multi-level intelligent collaborative control architecture for an integrated energy system also includes:

[0013] Establish a simplified integrated energy system steady-state model that considers the combined operation of the power system and the natural gas system, including a DC power flow model for the power system, a steady-state model for the natural gas pipeline without compressors, and a gas generator model;

[0014] The power system adopts the DC power flow model, as shown below:

[0015]

[0016]

[0017]

[0018] Where: P i is the active power injected by node i; θ j is the voltage phase angle at node j; n is the number of nodes; is the susceptance matrix element in the DC power flow equation; x ij is the reactance of branch ij;

[0019] The pipeline flow equation of the natural gas system pipeline steady-state model without considering the compressor is as follows:

[0020]

[0021]

[0022] Where: Fbd is the natural gas flow rate of pipeline bd; k bd is the parameter of the pipeline; s bd is a parameter indicating the direction of natural gas flow; p b and pd are the pressures at nodes b and d, respectively;

[0023] The gas generator model is as follows:

[0024]

[0025]

[0026] Where: Q GG is the heat consumed by the gas generator; P GG is the electric power generated by the gas generator; α GG , β GG and γ GG is the energy conversion efficiency coefficient; Γ GG is the natural gas consumption; L HV It is the lower calorific value of natural gas.

[0027] Furthermore, the controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation include:

[0028] Based on the regulation characteristics and time scales of various flexible loads, and using the K-means clustering method, we determine the changing patterns of user-side electricity consumption behavior under social environmental factors, and statistically analyze data such as the adjustable range and timing of user-side load regulation.

[0029] Determine the control range and margin of massive distributed controllable loads and their correlation with regulation compensation costs. Analyze the correlation between user electricity consumption behavior and electricity comfort based on the inductive statistics of historical data.

[0030] According to the regulation capability, uncertainty and dynamic characteristics of renewable energy power generation resources, a controllable and adjustable control strategy for the aggregation of massive flexible resources is established, thereby providing reliable data support for the coordinated control of source-load aggregation nodes.

[0031] Furthermore, the controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation also includes:

[0032] Establish a detailed integrated energy system steady-state model, taking into account the combined operation of the power system, natural gas system, and thermal system. The coupling elements between the systems are gas generators and CHP units. This includes an AC power flow model for the power system, a steady-state model for the natural gas pipeline including compressors, and a CHP unit model.

[0033] The power system adopts the AC power flow model, as shown below:

[0034]

[0035] Where: P i , Q i are the active power and reactive power injected into node i respectively; e i 、e j and f i 、f j are the real and imaginary parts of the voltage at nodes i and j, respectively; G ij and B ij are the real and imaginary parts of the node admittance matrix elements, respectively;

[0036] Consider the steady-state model of a natural gas pipeline with a compressor as shown below,

[0037] The power consumption of the compressor is,

[0038]

[0039] Where: H com is the power consumed by the compressor; E and W are constants, E is determined by the compressor temperature and efficiency, and W is determined by the compression factor; F com is the natural gas flow rate flowing through the compressor; p o is the pressure at node o;

[0040] The natural gas flow consumed by the gas turbine is

[0041]

[0042] Where: τ com is the natural gas flow consumed by the gas turbine; α com , β com and γ com is the energy conversion efficiency constant;

[0043] CHP units generate heat and electricity simultaneously by consuming natural gas. There are two types: fixed heat-to-electricity ratio and variable heat-to-electricity ratio. The model is as follows:

[0044]

[0045]

[0046] Where: c m 、 and They are the constant heat-to-electricity ratio, output heat power and electric power of the constant heat-to-electricity ratio CHP unit; c z 、 and They are the variable heat-to-electricity ratio, output thermal power and electric power of the variable heat-to-electricity ratio CHP unit; is the efficiency coefficient of the variable heat-to-power ratio CHP unit; F in is the natural gas input flow rate.

[0047] Furthermore, the establishment of an efficient and rapid collaborative control model for distributed controllable resources at the agent layer includes:

[0048] The multi-agent deep reinforcement learning method is used to determine the autonomous collaborative interaction mechanism between the edge-layer virtual aggregation nodes and the end-layer distributed flexible controllable units triggered by power fluctuation limit conditions.

[0049] Based on the integration of adaptive dynamic programming and mechanism model, a distributed operation control paradigm based on security event triggering among various intelligent agents is established;

[0050] Based on a multi-agent control architecture with vertical coordination and horizontal collaboration, a control paradigm with hierarchical interaction and autonomous collaboration among edge, end, and end is established to suppress power fluctuations under plug-and-play energy shocks in real time and efficiently.

[0051] Furthermore, the establishment of an efficient and rapid collaborative control model for distributed controllable resources at the proxy layer also includes:

[0052] A transient model of the integrated energy system is established. Considering the differences in the dynamic characteristics of the power system and the natural gas system on a time scale, the transient model of the natural gas system is adopted. The model is as follows:

[0053]

[0054]

[0055] Where: ρ is the density of natural gas; v is the axial velocity of natural gas; l is the length variable of the natural gas pipeline; t is the time variable; p is the pressure of natural gas; g0 is the acceleration of gravity; H is the elevation, κ is the friction factor, and D is the inner diameter of the pipeline. When the pipelines are at the same level, it can be simplified to the following formula:

[0056]

[0057] The natural gas pressure p and pipeline flow F are as follows,

[0058] p=ρZRT

[0059]

[0060] Where: Z is the average compressibility factor of natural gas; R is the gas constant, T is the average temperature of natural gas, and ρ0 is the density of natural gas under standard conditions.

[0061] Beneficial effects of the present invention: The multi-level collaborative control method of the integrated energy system based on multi-agents of the present invention includes: constructing a multi-level intelligent agent collaborative control architecture for the integrated energy system; conducting controllable-adjustable characteristic analysis and modeling of massive distributed flexible resources virtual aggregation; establishing an efficient and rapid collaborative control model for distributed controllable resources at the agent layer; the multi-level collaborative control method of the integrated energy system based on multi-agents of the present invention establishes intelligent agents with autonomous intelligent decision-making that correspond one-to-one to the hierarchical nodes of the physical system, and constructs a multi-level intelligent agent collaborative control architecture for the integrated energy system that adapts to efficient access of distributed resources and cross-platform; analyzes the spatiotemporal uncertainty change laws of renewable energy power generation and load power consumption behaviors under the influence of social environmental factors, and constructs a "controllable-adjustable" characteristic model of virtual aggregation of massive distributed source-load-storage controllable resources; on this basis, establishes a control paradigm of hierarchical interaction and autonomous collaboration between edges, ends, and ends, and effectively suppresses tidal fluctuations under "plug and play" energy impact in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 Flowchart of the multi-level collaborative control method for an integrated energy system based on multi-agents.

[0064] Figure 2 Schematic diagram of the technical route of the multi-level collaborative control method for integrated energy systems based on multi-agents. DETAILED DESCRIPTION

[0065] The present invention will be further described below with reference to the accompanying drawings.

[0066] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0068] It should be noted that all directional indications in the embodiments of the present invention (such as up-down-left-right-front-back...) are only used to explain the relative position relationship - movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. The connection can be a direct connection or an indirect connection.

[0069] like Figure 1-2 As shown, the multi-level collaborative control method of the integrated energy system based on multi-agent of the present invention includes:

[0070] Construct a multi-level intelligent collaborative control architecture for integrated energy systems;

[0071] Conduct controllable and adjustable feature analysis and modeling for massive distributed flexible resource virtual aggregation;

[0072] Establish an efficient and rapid collaborative control model for distributed controllable resources at the agent layer.

[0073] Furthermore, the construction of a multi-level intelligent collaborative control architecture for an integrated energy system includes:

[0074] Analyze the dynamic characteristics of each energy flow aggregation node in the integrated energy system at different time and spatial scales, determine the electrical coupling and information coupling relationship between each node, and further use topological graph theory to describe the physical and information interaction relationship between each energy flow aggregation node;

[0075] For each physical system level node in the integrated energy system, based on the dynamic characteristics of each node, an intelligent agent with intelligent environmental perception, intelligent knowledge acquisition, and autonomous intelligent decision-making is constructed. Based on the physical information interaction relationship between each node, complex system and networked control theory are used to build physical information associations between intelligent agents.

[0076] By analyzing the functional characteristics of different intelligent agents and the logical relationships between them, a scheduling system with a three-layer architecture of terminal layer, agent layer and master layer is constructed based on the idea of ​​hierarchical coordination and peer collaboration.

[0077] Furthermore, the construction of a multi-level intelligent collaborative control architecture for an integrated energy system also includes:

[0078] Establish a simplified integrated energy system steady-state model that considers the combined operation of the power system and the natural gas system, including a DC power flow model for the power system, a steady-state model for the natural gas pipeline without compressors, and a gas generator model;

[0079] The power system adopts the DC power flow model, as shown below:

[0080]

[0081]

[0082]

[0083] Where: P i is the active power injected by node i; θ j is the voltage phase angle at node j; n is the number of nodes; is the susceptance matrix element in the DC power flow equation; x ij is the reactance of branch ij;

[0084] The pipeline flow equation of the natural gas system pipeline steady-state model without considering the compressor is as follows:

[0085]

[0086]

[0087] Where: F bd is the natural gas flow rate of pipeline bd; k bd is the parameter of the pipeline; s bd is a parameter indicating the direction of natural gas flow; p b and p d are the pressures at nodes b and d, respectively;

[0088] The gas generator model is as follows:

[0089]

[0090]

[0091] Where: Q GG is the heat consumed by the gas generator; P GG is the electric power generated by the gas generator; α GG , β GG and γ GG is the energy conversion efficiency coefficient; Γ GG is the natural gas consumption; L HV It is the lower calorific value of natural gas.

[0092] Furthermore, the controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation include:

[0093] Based on the regulation characteristics and time scales of various flexible loads and the K-means clustering method, we determined how electricity consumption behavior changes under social environmental factors, analyzed the dynamic impact mechanism between massive distributed adjustable loads and electricity prices, and statistically analyzed data such as the adjustable range and timing of user-side load regulation.

[0094] Based on historical data, a neural network deep learning method is used to determine the control range and margin of massive distributed controllable loads and their correlation with regulation compensation costs. Based on the inductive statistics of historical data, the correlation between user electricity consumption behavior and electricity comfort is analyzed.

[0095] According to the regulation capability, uncertainty and dynamic characteristics of renewable energy power generation resources, and taking into account the constraints such as upper and lower output limits, ramp rate and power generation reserve, a controllable and adjustable control strategy for the aggregation of massive flexible resources is established, thereby providing reliable data support for the coordinated control of source-load aggregation nodes.

[0096] Furthermore, the controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation also includes:

[0097] Establish a detailed integrated energy system steady-state model, taking into account the combined operation of the power system, natural gas system, and thermal system. The coupling elements between the systems are gas generators and CHP units. This includes an AC power flow model for the power system, a steady-state model for the natural gas pipeline including compressors, and a CHP unit model.

[0098] The power system adopts the AC power flow model, as shown below:

[0099]

[0100] Where: P i , Q i are the active power and reactive power injected into node i respectively; e i 、e j and f i 、f j are the real and imaginary parts of the voltage at nodes i and j, respectively; G ij and B ij are the real and imaginary parts of the node admittance matrix elements, respectively;

[0101] Consider the steady-state model of a natural gas pipeline with a compressor as shown below,

[0102] The power consumption of the compressor is,

[0103]

[0104] Where: H com is the power consumed by the compressor; E and W are constants, E is determined by the compressor temperature and efficiency, and W is determined by the compression factor; F com is the natural gas flow rate flowing through the compressor; p o is the pressure at node o;

[0105] The natural gas flow consumed by the gas turbine is

[0106]

[0107] Where: τ com is the natural gas flow consumed by the gas turbine; α com , β com and γ com is the energy conversion efficiency constant;

[0108] CHP units generate heat and electricity simultaneously by consuming natural gas. There are two types: fixed heat-to-electricity ratio and variable heat-to-electricity ratio. The model is as follows:

[0109]

[0110]

[0111] Where: c m 、 and They are the constant heat-to-electricity ratio, output heat power and electric power of the constant heat-to-electricity ratio CHP unit; c z 、 and They are the variable heat-to-electricity ratio, output thermal power and electric power of the variable heat-to-electricity ratio CHP unit; is the efficiency coefficient of the variable heat-to-power ratio CHP unit; F in is the natural gas input flow rate.

[0112] Furthermore, the establishment of an efficient and rapid collaborative control model for distributed controllable resources at the agent layer includes:

[0113] Aiming at the impact power fluctuations in multiple spatiotemporal distributions caused by distributed energy plug-and-play in integrated energy systems, a multi-agent deep reinforcement learning method is used to determine the autonomous collaborative interaction mechanism of intelligent agents between edge-layer virtual aggregation nodes and between them and end-layer distributed flexible controllable units triggered by power fluctuation exceeding limit conditions.

[0114] Based on the integration of adaptive dynamic programming and mechanism model, a distributed operation control paradigm based on safety event triggering is established to achieve autonomous awakening and flexible collaboration between intelligent agents when there is risk and autonomy when there is no risk.

[0115] Based on a multi-agent control architecture with vertical coordination and horizontal collaboration, a control paradigm with hierarchical interaction and autonomous collaboration among edge, end, and end is established to suppress power fluctuations under plug-and-play energy shocks in real time and efficiently.

[0116] Furthermore, the establishment of an efficient and rapid collaborative control model for distributed controllable resources at the proxy layer also includes:

[0117] A transient model of the integrated energy system is established. Considering the differences in the dynamic characteristics of the power system and the natural gas system on a time scale, the transient model of the natural gas system is adopted. The model is as follows:

[0118]

[0119]

[0120] Where: ρ is the density of natural gas; v is the axial velocity of natural gas; l is the length variable of the natural gas pipeline; t is the time variable; p is the pressure of natural gas; g0 is the acceleration of gravity; H is the elevation, κ is the friction factor, and D is the inner diameter of the pipeline. When the pipelines are at the same level, it can be simplified to the following formula:

[0121]

[0122] The natural gas pressure p and pipeline flow F are as follows,

[0123] p=ρZRT

[0124]

[0125] Where: Z is the average compressibility factor of natural gas; R is the gas constant, T is the average temperature of natural gas, and ρ0 is the density of natural gas under standard conditions.

[0126] In summary, the multi-level collaborative control method of the integrated energy system based on multiple agents of the present invention first clarifies the electrical coupling relationship and spatiotemporal mapping relationship of each energy flow aggregation node of the physical system, and studies the establishment of an intelligent agent that corresponds one-to-one to the hierarchical nodes of the physical system, can intelligently perceive the environment, intelligently acquire knowledge, and has autonomous intelligent decision-making; further, the three-layer scheduling framework of the integrated energy system "terminal layer-agent layer-master station layer" that adapts to efficient access to distributed resources and cross-platform is studied.

[0127] Secondly, the changing laws of spatiotemporal uncertainty of renewable energy power generation and load electricity consumption behavior under the influence of social environmental factors are analyzed; considering the spatiotemporal uncertainty of both source and load ends, time-of-use electricity prices, user satisfaction and incentive-type load demand-side responses at different time scales, the study constructs the "controllable-adjustable" characteristics of virtual aggregation of massive distributed source, load and storage controllable resources.

[0128] Finally, in response to the multi-temporal and spatially distributed power fluctuations caused by the "plug and play" of distributed energy, based on the multi-agent control architecture, the autonomous collaborative interaction mechanism and distributed collaborative dynamic control method of the agents between the edge-layer virtual aggregation nodes and the end-layer distributed flexible controllable units triggered by the power fluctuation limit conditions are studied, and a control paradigm of hierarchical interaction and autonomous collaboration between edge-edge, edge-end, and end-end is established to suppress the power fluctuations under the "plug and play" energy impact in real time and efficiently.

[0129] In terms of integrated energy coordinated control, the integrated energy system is the primary carrier of a new type of power system that integrates multiple energy sources, including electricity, gas, heating / cooling, and hydrogen. While these distributed energy sources can be autonomously plugged and played as long as they meet access standards, this can also generate disruptive power flow fluctuations distributed across multiple time and space, potentially endangering system operation. Within this open, plug-and-play access environment, it is crucial to flexibly activate and coordinate the control of massive distributed controllable resources to rapidly mitigate energy fluctuations and ensure high-quality system energy supply. The multi-level collaborative control method for an integrated energy system based on multiple agents of the present invention includes the following steps: constructing a multi-level intelligent agent collaborative control architecture for an integrated energy system; conducting controllable and adjustable characteristic analysis and modeling of massive distributed flexible resources virtual aggregation; establishing an efficient and rapid collaborative control model for distributed controllable resources at the agent layer; through the above steps, the multi-level collaborative control method for an integrated energy system based on multiple agents of the present invention establishes intelligent agents with autonomous intelligent decision-making that correspond one-to-one to the hierarchical nodes of the physical system, and constructs a multi-level intelligent agent collaborative control architecture for an integrated energy system that is adaptable to efficient access of distributed resources and cross-platform; analyzing the spatiotemporal uncertainty variation laws of renewable energy power generation and load power consumption behaviors under the influence of social environmental factors, and constructing a "controllable-adjustable" characteristic model for the virtual aggregation of massive distributed source-load-storage controllable resources; on this basis, establishing a control paradigm for hierarchical interaction and autonomous collaboration between edges, ends, and ends, and effectively suppressing tidal fluctuations under "plug and play" energy impacts in real time.

[0130] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0131] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0132] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The above embodiments are only used to illustrate the present invention, wherein the structure, connection mode and manufacturing process of each component can be changed. Any equivalent transformation and improvement based on the technical solution of the present invention should not be excluded from the scope of protection of the present invention.

[0134] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0135] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. A multi-level collaborative control method for an integrated energy system based on multi-agents, characterized in that: include: Construct a multi-level intelligent collaborative control architecture for integrated energy systems; Conduct controllable and adjustable feature analysis and modeling for massive distributed flexible resource virtual aggregation; Establish an efficient and rapid collaborative control model for distributed controllable resources at the agent layer.

2. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 1, characterized in that: The multi-level intelligent collaborative control architecture for the integrated energy system includes: Analyze the dynamic characteristics of each energy flow aggregation node in the integrated energy system at different time and spatial scales, determine the electrical coupling and information coupling relationship between each node, and further use topological graph theory to describe the physical and information interaction relationship between each energy flow aggregation node; For each physical system level node in the integrated energy system, based on the dynamic characteristics of each node, an intelligent agent with intelligent environmental perception, intelligent knowledge acquisition, and autonomous intelligent decision-making is constructed. Based on the physical information interaction relationship between each node, complex system and networked control theory are used to build physical information associations between intelligent agents. By analyzing the functional characteristics of different intelligent agents and the logical relationships between them, a scheduling system with a three-layer architecture of terminal layer, agent layer and master layer is constructed based on the idea of ​​hierarchical coordination and peer collaboration.

3. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 2, characterized in that: The construction of a multi-level intelligent collaborative control architecture for an integrated energy system also includes: Establish a simplified integrated energy system steady-state model that considers the combined operation of the power system and the natural gas system, including a DC power flow model for the power system, a steady-state model for the natural gas pipeline without compressors, and a gas generator model; The power system adopts the DC power flow model, as shown below: Where: P i is the active power injected by node i; θ j is the voltage phase angle at node j; n is the number of nodes; is the susceptance matrix element in the DC power flow equation; x ij is the reactance of branch ij; The pipeline flow equation of the natural gas system pipeline steady-state model without considering the compressor is as follows: Where: F bd is the natural gas flow rate of pipeline bd; k bd is the parameter of the pipeline; s bd is a parameter indicating the direction of natural gas flow; p b and p d are the pressures at nodes b and d, respectively; The gas generator model is as follows: Where: Q GG is the heat consumed by the gas generator; P GG is the electric power generated by the gas generator; α GG , β GG and γ GG is the energy conversion efficiency coefficient; Γ GG is the natural gas consumption; L HV It is the lower calorific value of natural gas.

4. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 1, characterized in that: The controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation include: Based on the regulation characteristics and time scales of various flexible loads, and using the K-means clustering method, we determine the changing patterns of user-side electricity consumption behavior under social environmental factors, and statistically analyze data such as the adjustable range and timing of user-side load regulation. Determine the control range and margin of massive distributed controllable loads and their correlation with regulation compensation costs. Analyze the correlation between user electricity consumption behavior and electricity comfort based on the inductive statistics of historical data. According to the regulation capability, uncertainty and dynamic characteristics of renewable energy power generation resources, a controllable and adjustable control strategy for the aggregation of massive flexible resources is established.

5. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 4, characterized in that: The controllable and adjustable characteristics analysis and modeling of massive distributed flexible resource virtual aggregation also includes: Establish a detailed integrated energy system steady-state model, taking into account the combined operation of the power system, natural gas system, and thermal system. The coupling elements between the systems are gas generators and CHP units. This includes an AC power flow model for the power system, a steady-state model for the natural gas pipeline including compressors, and a CHP unit model. The power system adopts the AC power flow model, as shown below: Where: P i , Q i are the active power and reactive power injected into node i respectively; e i 、e j and f i 、f j are the real and imaginary parts of the voltage at nodes i and j, respectively; G ij and B ij are the real and imaginary parts of the node admittance matrix elements, respectively; Consider the steady-state model of a natural gas pipeline with a compressor as shown below, The power consumption of the compressor is, Where: H com is the power consumed by the compressor; E and W are constants, E is determined by the compressor temperature and efficiency, and W is determined by the compression factor; F com is the natural gas flow through the compressor; p o is the pressure at node o; The natural gas flow consumed by the gas turbine is Where: τ com is the natural gas flow consumed by the gas turbine; α com , β com and γ com is the energy conversion efficiency constant; CHP units generate heat and electricity simultaneously by consuming natural gas. There are two types: fixed heat-to-electricity ratio and variable heat-to-electricity ratio. The model is as follows: Where: c m 、 and They are the constant heat-to-electricity ratio, output heat power and electric power of the constant heat-to-electricity ratio CHP unit; c z 、 and They are the variable heat-to-electricity ratio, output thermal power and electric power of the variable heat-to-electricity ratio CHP unit; is the efficiency coefficient of the variable heat-to-power ratio CHP unit; F in is the natural gas input flow rate.

6. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 1, characterized in that: The establishment of an efficient and rapid collaborative control model for distributed controllable resources at the agent layer includes: The multi-agent deep reinforcement learning method is used to determine the autonomous collaborative interaction mechanism between the edge-layer virtual aggregation nodes and the end-layer distributed flexible controllable units triggered by power fluctuation limit conditions. Based on the integration of adaptive dynamic programming and mechanism model, a distributed operation control paradigm based on security event triggering among various intelligent agents is established; Based on a multi-agent control architecture with vertical coordination and horizontal collaboration, a control paradigm of hierarchical interaction and autonomous collaboration among edges, ends, and ends is established.

7. The multi-level collaborative control method for an integrated energy system based on multi-agents according to claim 1, characterized in that: The establishment of an efficient and rapid collaborative control model for decentralized controllable resources at the proxy layer further includes: A transient model of the integrated energy system is established. Considering the differences in the dynamic characteristics of the power system and the natural gas system on a time scale, the transient model of the natural gas system is adopted. The model is as follows: Where: ρ is the density of natural gas; v is the axial velocity of natural gas; l is the length variable of the natural gas pipeline; t is the time variable; p is the pressure of natural gas; g0 is the acceleration of gravity; H is the elevation, k is the friction factor, and D is the inner diameter of the pipeline. When the pipelines are at the same level, it can be simplified to the following formula: The natural gas pressure p and pipeline flow F are as follows, p=ρZRT Where: Z is the average compressibility factor of natural gas; R is the gas constant, T is the average temperature of natural gas, and ρ0 is the density of natural gas under standard conditions.

Images