Method and system for collaborative optimization of large-scale electrothermal hydrogen cross-regional integrated energy system

Abstract

This invention discloses a collaborative optimization method and system for a large-scale cross-regional integrated energy system of electricity, heat, and hydrogen. The method includes the following steps: upgrading network structure modeling to construct a multi-level multi-energy flow coupling model; improving the large-scale system-adaptive ADMM algorithm to optimize algorithm efficiency; adopting multi-factor weighted generalized Nash bargaining to innovate the distribution of benefits among multiple stakeholders; establishing a distributed robust optimization model to robusten uncertainty, including considering spatiotemporal correlation to define the uncertainty set and determining the distributed robust optimization objective; and performing equipment aggregation modeling under large-scale systems to adapt to equipment clusters. This invention effectively solves four major pain points in large-scale cross-regional integrated energy systems: a surge in computational complexity, intensified coupling of energy system network structures, imbalance in the distribution of benefits among multiple stakeholders, and accumulation of uncertainty. This enables the collaborative optimization of large-scale cross-regional integrated energy systems of electricity, heat, and hydrogen across regions and provinces.

Description

Technical Field

[0001] This invention belongs to the field of integrated energy system (IES) technology, specifically relating to a collaborative optimization method and system for a large-scale electrothermal hydrogen cross-regional integrated energy system. Background Technology

[0002] Compared to the decentralized development model of individual energy systems (IES), the collaborative control and unified management model of multiple IESs has significant advantages in terms of operational efficiency, economy, and low carbon emissions. This is because this model can integrate diverse and flexible regional resources, enabling interaction, sharing, and mutual support among various energy systems. However, with the accelerated development of IESs, the complexity of multi-energy interactions is increasing, further complicating the coordination of relationships among the various stakeholders in the system.

[0003] For the problem of multi-IES collaborative optimization, existing research mostly focuses on the low-carbon economic optimization operation of a single IES. Research on multi-regional multi-energy collaborative optimization is still in its early stages, especially regarding the following limitations in the research on multi-IES collaborative optimization mechanisms: 1) Insufficient in-depth exploration of hydrogen energy sharing in multi-IES systems; 2) Existing research mostly focuses on the comparative analysis of economic indicators before and after the cooperative game model, without systematically revealing the impact mechanism of cooperative game on the internal demand response and hydrogen blending model optimization, and also lacking quantitative research on the evolution characteristics of low-carbon benefits. To address these shortcomings, some scholars have proposed a cross-regional electricity-heat-hydrogen multi-energy sharing collaborative optimization model based on cooperative game theory, using generalized Nash bargaining theory as its foundation. However, in this model, they only explored the collaborative optimization problem of multi-IES composed of three IESs, without exploring multi-energy sharing optimization methods for larger-scale cross-regional IESs to improve the model's adaptability to complex network structures. In response to the needs of future large-scale new energy systems, there is an urgent need for a collaborative optimization method for large-scale cross-regional integrated energy systems involving electricity, heat, and hydrogen across regions and provinces. This method aims to address four major challenges that large-scale cross-regional integrated energy systems must face: a surge in computational complexity, increased coupling of energy system network structures, imbalance in the distribution of interests among multiple stakeholders, and the accumulation of uncertainties. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a collaborative optimization method and system for a large-scale electrothermal hydrogen cross-regional integrated energy system. It mainly addresses four major pain points: the surge in computational complexity of large-scale cross-regional integrated energy systems, the intensified coupling of energy system network structures, the imbalance in the distribution of interests among multiple stakeholders, and the accumulation of uncertainties.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In a first aspect, the present invention provides a collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system, comprising:

[0007] A multi-level multi-energy flow coupling model is constructed for the large-scale electricity-heat-hydrogen cross-regional integrated energy system to realize network structure modeling; the multi-level multi-energy flow coupling model includes hierarchical partitioning and correlation matrix, refined modeling of multi-energy flow transmission, and multi-level coordination constraints;

[0008] An improved adaptive ADMM algorithm for large-scale electric-thermal-hydrogen cross-regional integrated energy systems is adopted to reduce the order of the large-scale electric-thermal-hydrogen cross-regional integrated energy system.

[0009] The large-scale electricity-heat-hydrogen cross-regional integrated energy system adopts multi-factor weighted generalized Nash bargaining to innovate the distribution of interests among multiple stakeholders;

[0010] A distributed robust optimization model is established for the large-scale electricity-heat-hydrogen cross-regional integrated energy system to robusten the uncertainty;

[0011] Equipment aggregation modeling is performed on the large-scale electricity-heat-hydrogen cross-regional integrated energy system to adapt the equipment cluster.

[0012] Secondly, this invention provides a method and system for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system, comprising:

[0013] The network structure modeling upgrade construction module is used to build multi-level multi-energy flow coupling models for large-scale cross-regional integrated energy systems of electricity, heat, and hydrogen, including:

[0014] 1) The hierarchical division and association matrix submodule is used to define hierarchical sets, establish region-hierarchical association matrices and network-region association matrices;

[0015] 2) A refined modeling submodule for multi-energy flow transmission, used to establish power grid transmission models, hydrogen pipeline transmission models, and heat network transmission models;

[0016] 3) Multi-level coordination constraint submodule, used to establish upper-level power allocation constraints on lower-level layers and lower-level layers feedback constraints to upper-level layers;

[0017] The algorithm efficiency optimization construction module reduces the order of the large-scale electricity-heat-hydrogen inter-regional integrated energy system; it includes:

[0018] 1) The block-based ADMM submodule is used to describe the problem decomposition logic, block objective function, coupling constraint handling method, and iterative update rules;

[0019] 2) Parallel computing acceleration submodule, used for thread allocation, data communication optimization and accelerated convergence;

[0020] 3) Model reduction submodule, used to filter key regions and simplify aggregated region variables;

[0021] A multi-stakeholder interest distribution innovation module is used to innovate the multi-stakeholder interest distribution by employing multi-factor weighted generalized Nash bargaining for the large-scale electricity-heat-hydrogen inter-regional integrated energy system, including:

[0022] 1) Multi-dimensional contribution quantification submodule, used to quantify resource endowment contribution, network contribution, investment cost contribution and comprehensive contribution;

[0023] 2) An improved generalized Nash bargaining model submodule, used to describe the objective function, define the constraints, and calculate the Nash equilibrium solution;

[0024] 3) The inter-level benefit-sharing mechanism submodule is used to describe the cost-sharing method of the upper-level network and the cross-level subsidy mechanism;

[0025] An uncertainty robustness construction module establishes a distributed robust optimization model for the large-scale electricity-heat-hydrogen inter-regional integrated energy system, robustening the uncertainty, including:

[0026] 1) Uncertainty set definition submodule, used to define the uncertainty set of wind and solar power output, the uncertainty set of load, and spatiotemporal correlation constraints;

[0027] 2) Distributed robust optimization objective submodule, used to establish the max-min objective function, calculate the penalty cost and solve the transformation, transforming the max-min problem into a single-level optimization through duality theory;

[0028] The equipment cluster adaptation construction module performs equipment aggregation modeling for the large-scale electricity-heat-hydrogen cross-regional integrated energy system, including:

[0029] 1) Electrolyzer cluster model submodule, used to describe polymerization power constraints, cluster ramp-up constraints and hydrogen production efficiency model;

[0030] 2) Energy storage cluster model submodule, used to describe aggregated capacity constraints, charge / discharge power constraints, and cluster self-discharge rate.

[0031] Compared with the prior art, the advantages of this invention are as follows:

[0032] 1) The network structure modeling was upgraded, and a multi-level multi-energy flow coupling model was established. A three-dimensional coupling model of hierarchy-region-pipeline was constructed, which accurately depicted the constraints of complex networks.

[0033] 2) The large-scale system-adaptive ADMM algorithm has been improved, and its efficiency has been optimized. Through block-based optimization, parallel processing, and order reduction, the solution efficiency can be significantly improved.

[0034] 3) It adopts a multi-factor weighted generalized Nash bargaining mechanism, which innovates the distribution of benefits among multiple stakeholders. It considers multiple factors such as resource endowment, network contribution, and investment costs to ensure the fairness of benefit distribution.

[0035] 4) A distributed robust optimization model was established to robusten uncertainties. A distributed robust + interval constraint model was constructed to improve the system's disturbance resistance.

[0036] 5) Perform device aggregation modeling in large-scale systems to adapt to device clusters. Aggregate similar devices into virtual clusters to simplify model complexity.

[0037] 6) It can solve four major pain points: the surge in computational complexity of large-scale cross-regional integrated energy systems, the intensified coupling of energy system network structures, the imbalance in the distribution of interests among multiple stakeholders, and the accumulation of uncertainty. Attached Figure Description

[0038] Figure 1 The main flowchart of the collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system provided in Embodiment 1 of this application;

[0039] Figure 2 A schematic diagram of the main structure of the large-scale electrothermal hydrogen cross-regional integrated energy system collaborative optimization method and system provided in the embodiments of this application. Detailed Implementation

[0040] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0041] Example 1

[0042] The definition of a large-scale cross-regional integrated energy system (hereinafter referred to as a large-scale system) is as follows: 1) In terms of the number of regions, there are no fewer than 10 IES (Integrated Energy Systems). This standard covers provincial / cross-provincial areas, such as the 15 municipal IES in the North China Power Grid. 2) In terms of network structure, it has multiple levels, such as provincial-municipal-industrial park levels; it has multiple energy pipelines, such as the interweaving of power grid, heating network, and hydrogen pipeline network. 3) It involves no fewer than 20 stakeholders, including regional operators, power grid companies, heating companies, hydrogen energy producers, user clusters, energy storage companies, etc. 4) In terms of equipment scale, it includes electrolyzers, fuel cells, photovoltaics, wind turbines, electric energy storage, thermal storage tanks, etc., with a total of no fewer than 50 units, and the installed capacity of wind power and photovoltaics is more than 10 GW. The core characteristics of this type of large-scale cross-regional integrated energy system are as follows: 1) In terms of energy flow coupling, electricity, heat, and hydrogen can flow bidirectionally across levels and regions, and the loss and blockage effects are significant in the energy flow and conversion process. 2) In terms of computational complexity, the number of variables has increased from thousands to tens of thousands to hundreds of thousands, while the number of constraints has grown exponentially. 3) Regarding conflicts of interest, the contributions of various stakeholders differ greatly, making it difficult for traditional bargaining mechanisms to guarantee fairness. 4) In terms of uncertainty, the uncertainty of wind and solar power output on the source side and the uncertainty of load demand on the load side accumulate on a large scale, requiring higher system resistance to disturbances.

[0043] Therefore, this embodiment provides a collaborative optimization method for a large-scale cross-regional integrated energy system of electricity, heat, and hydrogen. It offers a feasible technical solution from five aspects: network structure modeling upgrade, algorithm efficiency optimization, multi-stakeholder benefit distribution innovation, uncertainty robustness, and equipment cluster adaptation. The specific steps include:

[0044] S1 constructs a multi-level multi-energy flow coupling model for a large-scale cross-regional integrated energy system of electricity, heat and hydrogen to achieve network structure modeling.

[0045] Thus, in this step, by constructing a multi-level multi-energy flow coupling model for the large-scale system, the complex network structure of the large-scale system can be upgraded through network structure modeling, so as to facilitate the solution of the subsequent step S2.

[0046] Existing technologies only consider simple single-layer networks. The "large-scale system" mentioned in step S1 requires the construction of a three-dimensional coupled model of hierarchy, region, and network, as well as the accurate characterization of complex network constraints. Specifically, it includes the following sub-steps:

[0047] S1.1, Hierarchical Division and Association Matrix.

[0048] Define the hierarchical set as L1 is a provincial (level 1) IES, L2 is a municipal (level 2) IES, and L3 is a park-level (level 3) IES.

[0049] The region-hierarchical association matrix is ​​represented as follows Where N is the number of regions, M is the number of levels, and A ij =1 indicates that the i-th region belongs to the j-th level; otherwise, it is 0.

[0050] The pipeline network-region correlation matrix is ​​represented as follows: Where K is the number of pipeline types, including electric / heat / hydrogen pipelines, and B kl =1 indicates that the k-type pipeline network covers the l-th area; otherwise, it is 0.

[0051] S1.2, Refined Modeling of Multi-Energy Stream Transmission.

[0052] (1) Power grid transmission model

[0053] Considering losses and congestion, the power grid transmission model is as follows: (1)

[0054] in, Provide power output for region i; The power transmitted from region i to j is represented by Ω; positive values ​​indicate transmission and negative values ​​indicate reception. i Represents the set of neighboring regions of region i; Let be the load power of region i; The charging power for the regional i-energy storage system; The discharge power of the regional i energy storage system; The power consumption of the electrolytic cell in region i is denoted as .

[0055] Using the B-coefficient method, the transmission loss model is expressed as:

[0056] (2)

[0057] in, Resistance per unit length; This refers to the transmission distance.

[0058] The blocking constraint is: (3)

[0059] in, This represents the upper limit of the line's transmission capacity.

[0060] Inter-level power transmission constraints are: (4)

[0061] It means that the total power received by the provincial level from the municipal level is equal to the power allocated by the provincial level to each region.

[0062] (2) Hydrogen pipeline network transmission model

[0063] Considering hydrogen pressure loss and leakage, the hydrogen pipeline network transmission model is as follows: The hydrogen flow rate constraint is: (5)

[0064] in, Let be the hydrogen flow rate in region i→j; This refers to the pipeline flow coefficient. For node pressure difference; Let be the hydrogen pressure in region i; Let be the hydrogen pressure in region j.

[0065] The leakage loss model is as follows: (6)

[0066] in, The leakage rate per unit length is 0.001.

[0067] The pressure constraint is: (7)

[0068] in, Take 0.5 MPa, Take 35 MPa.

[0069] The regional hydrogen balance constraint is: (8)

[0070] in, The power consumption of the electrolytic cell in region i; The hydrogen production efficiency of the electrolyzer; The power generation capacity of the fuel cell in region i; For the power generation efficiency of fuel cells; The power of the hydrogen load in region i; Let be the hydrogen flow rate in region i→j; Let be the hydrogen leakage flow rate in region i→j.

[0071] (3) Heat network transmission model

[0072] Considering temperature decay, the heat transfer model for the heating network is as follows: The heat flow constraint is: (9)

[0073] in, The heat flow rate from region i to j; The heat transfer coefficient of the heating network; For water supply temperature; This refers to the return water temperature.

[0074] The temperature decay model is as follows: (10)

[0075] in, The temperature drop coefficient per unit length is taken as 0.005.

[0076] The regional thermal balance constraint is: (11)

[0077] in, The heating capacity of the gas-fired boiler (GB) in Zone i; The heating capacity of the regional i combined cooling, heating and power (CCHP) unit; The heat flow rate from region j to i; The heat load of region i; The thermal energy storage power of the regional i thermal energy storage system; The heat release power of the regional i thermal energy storage system.

[0078] S1.3, Multi-level Coordination Constraints.

[0079] The upper layer's power allocation constraint to the lower layer is: (12)

[0080] in, Let be the power allocation coefficient for region i. .

[0081] The feedback constraint from the lower layer to the upper layer is: (13)

[0082] It means that the surplus of the lower layer minus the deficit equals the transmission power of the upper layer.

[0083] In this way, the above steps can accurately characterize the complex network constraints of large-scale systems, thereby helping to reduce the degree of computational complexity.

[0084] S2 employs an improved Alternating Direction Method of Multipliers (ADMM) algorithm adapted for large-scale electric-thermal-hydrogen cross-regional integrated energy systems to reduce the order of the large-scale electric-thermal-hydrogen cross-regional integrated energy system and optimize the algorithm efficiency.

[0085] The existing adaptive ADMM algorithm converges slowly with tens of thousands of variables, requiring over 2000 iterations. This step S2 employs a triple optimization approach of "block partitioning + parallelism + order reduction" to improve solution efficiency, specifically including the following sub-steps:

[0086] S2.1, Decompose the problem by level / region and implement block-based ADMM.

[0087] Large-scale system optimization problem (where X is the total variable, X...) k The problem is decomposed into K subproblems (as a block variable), and the decomposition criterion is based on the division by level / region.

[0088] Taking region partitioning as an example, the objective function for partitioning is: (14)

[0089] in, The system operating cost for region i; Revenue from P2P transactions between regional integrated energy systems; Transaction costs when the regional integrated energy system trades with various markets (such as the electricity market, heat market, and hydrogen market); The operating costs of demand response are considered for the regional integrated energy system.

[0090] Introducing a global coupling variable Z=AX, where A is the coupling matrix, the augmented Lagrangian function is constructed as follows: (15)

[0091] The iterative update rules are as follows: 1) Regional sub-problem update; (16)

[0092] in, This represents the updated value of the optimization variable for region i during the (k+1)th iteration; Let be the Lagrange multiplier vector for the k-th iteration. Transpose it; Let A be the coupling constraint matrix for region i. i Let A be the i-th row; ρ is the global coupling variable for the k-th iteration; ρ is the augmented Lagrange penalty parameter.

[0093] 2) Global coupling variable update; (17)

[0094] It represents mean aggregation, ensuring global consistency.

[0095] 3) Lagrange multiplier update; (18)

[0096] S2.2 adopts a GPU + multi-threaded architecture to accelerate parallel computing.

[0097] Each level / region is assigned an independent computation thread, and subproblems are solved in parallel, reducing serial waiting time.

[0098] By adopting a "local computation + global synchronization" model, coupled variables are transmitted only during iterative updates, reducing communication volume by 70% and thus optimizing data communication.

[0099] Introducing dynamic inertia weight ω k The updated formula is as follows: (19)

[0100] Using this formula, convergence can be accelerated, and the number of iterations can be reduced from more than 2,000 to less than 500.

[0101] S2.3 achieves model order reduction through key region screening and variable aggregation.

[0102] The key area filtering method is as follows:

[0103] 1) Construct a regional importance evaluation matrix: (20)

[0104] in, Let be the maximum transmission power from region i to j.

[0105] 2) Calculate the regional importance score: 1(21)

[0106] Where α is the damping coefficient, taken as 0.85; 1 is an all-1 vector.

[0107] 3) Screening threshold: The top 80% of the regions are retained as key regions, and the rest are aggregated regions.

[0108] The method to simplify the variables in the aggregation region is to weight and aggregate similar variables (such as load and power output) from multiple aggregation regions into one equivalent variable, as shown in the formula; (twenty two)

[0109] in, The aggregate weights are allocated based on the regional load percentage.

[0110] Thus, in step S2, the improved ADMM algorithm is used to solve the multi-level multi-energy flow coupling model constructed in S1. Through the solution method of block-based + parallel + order reduction triple optimization, the ultra-large and ultra-complex problem is broken down into small blocks and calculated quickly, thereby significantly improving the solution efficiency.

[0111] S3 adopts multi-factor weighted generalized Nash bargaining for the large-scale electricity-heat-hydrogen cross-regional integrated energy system, and innovates the distribution of interests among multiple stakeholders.

[0112] Existing technologies calculate bargaining weights based solely on energy trading volume. Large-scale systems require consideration of multiple factors, including resource endowment, network contribution, and investment costs, to ensure fair benefit distribution. Therefore, step S3 employs a multi-factor weighted generalized Nash bargaining mechanism for large-scale electricity-heat-hydrogen cross-regional integrated energy systems, innovating the distribution of benefits among multiple stakeholders. This includes the following sub-steps:

[0113] S3.1, Multi-dimensional Quantification of Contributions.

[0114] Resource endowment contribution The quantification formula is: (twenty three)

[0115] Among them, it means The percentage of maximum wind and solar power output in the region; To achieve the maximum photovoltaic output for region i; The maximum wind power output of region i.

[0116] Network contribution The quantification formula is: (twenty four)

[0117] in, This represents the proportion of pipeline transmission capacity in region i. The maximum pipeline transmission capacity for region i→j; This represents the maximum pipeline transmission capacity for region j→i.

[0118] Investment cost contribution The quantification formula is: (25)

[0119] This represents the percentage of equipment investment costs in region i.

[0120] Overall contribution The quantification formula is: (26)

[0121] in, The weights can be determined using the analytic hierarchy process (AHP).

[0122] S3.2, Establish an improved generalized Nash bargaining model.

[0123] Considering the weights of multiple factors, the objective function is established as follows: (27)

[0124] in, Benefits after regional cooperation; Non-cooperative gains (i.e., the breaking point); For negotiation parameters; The overall contribution level.

[0125] The constraints are: (28)

[0126] This indicates that the total revenue of the alliance is conserved.

[0127] Then, by transforming the logarithmic objective into a linear objective, the distribution of benefits in each region, i.e., the Nash equilibrium solution, is obtained: (29)

[0128] S3.3, Establish a benefit-sharing mechanism among different levels.

[0129] The cost-sharing mechanism for the upper-layer network is as follows: first, calculate the sum of the construction and maintenance costs of the provincial-level network, and then allocate the costs according to the transmission power ratio of the lower-layer regions, as shown in the following formula: (30); (31)

[0130] in, This is the sum of the construction and maintenance costs of the pipeline network at the provincial level; The cost of pipeline construction and maintenance for region i; This refers to the cost allocation for provincial-level networks. This refers to the transmission power of the provincial-level pipeline network. This refers to the transmission power of the pipeline network at the city level.

[0131] The cross-level subsidy mechanism is as follows: resource-rich lower-level areas (such as rural areas rich in scenic beauty) receive subsidies from higher-level areas, and the formula is as follows; (32)

[0132] in, The subsidy coefficient is set to 0.3.

[0133] Thus, by considering multiple factors such as resource endowment, network contribution, and investment cost in step S3, the fairness of benefit distribution can be ensured, thereby enabling the network structure upgraded in step S1 and the algorithm optimized in step S2 to play their role, and allowing the entire large-scale system to operate stably for a long time.

[0134] S4. Establish a distributed robust optimization model for the large-scale electricity-heat-hydrogen cross-regional integrated energy system to robusten the uncertainty.

[0135] In large-scale systems, the uncertainties of wind and solar power fluctuations and load changes can accumulate and be amplified. Therefore, in step S4, a distributed robust model with interval constraints is constructed to improve disturbance resistance. Step S4 specifically includes the following sub-steps:

[0136] S4.1, Considering spatiotemporal correlation, define the uncertainty set.

[0137] The set of uncertainties in wind and solar power output is: (33)

[0138] in, The intraday volatility coefficient is set to 0.3. The daily total fluctuation limit is set to 2.0.

[0139] The set of load uncertainties is: (34)

[0140] The spatiotemporal constraints are: (35)

[0141] This indicates that the daytime photovoltaic power in region i and the evening photovoltaic power in region j are complementary.

[0142] S4.2, Determine the distributed robust optimization objective.

[0143] The objective function is: (36)

[0144] in, The formula for calculating the penalty cost is as follows; (37)

[0145] As punishment for abandoning wind and light, Penalty for load shortage.

[0146] Using duality theory, the max-min problem can be transformed into a single-level optimization problem. By introducing an auxiliary variable θ, it can be transformed into: (38)

[0147] Thus, step S4 constructs a distributed robust + interval constraint model to improve the anti-disturbance of large-scale systems, so that large-scale systems can remain stable and not crash no matter what unexpected events occur, while also controlling costs and preventing previous network, algorithm, and revenue sharing rules from being wasted.

[0148] S5, perform equipment aggregation modeling for the large-scale electricity-heat-hydrogen cross-regional integrated energy system and adapt the equipment cluster.

[0149] Existing technologies model individual devices. In large-scale systems, similar devices need to be aggregated into "virtual clusters" to simplify the model's complexity. Therefore, step S5 performs device aggregation modeling for large-scale electricity-heat-hydrogen cross-regional integrated energy systems to adapt to device clusters. This includes the following sub-steps:

[0150] S5.1, Establish an electrolytic cell cluster model.

[0151] The polymerization power constraints are as follows: (39)

[0152] in, This refers to the load power of a single electrolytic cell; This refers to the load power of the electrolytic cell cluster.

[0153] The minimum and maximum output of the cluster are as follows: (40)

[0154] in, This represents the minimum output of the electrolytic cell cluster. This is the maximum output of the electrolytic cell cluster.

[0155] The cluster ramping constraint is: (41)

[0156] in, The "maximum power-up rate" of the electrolyzer cluster is the maximum amount of hydrogen that the entire cluster can produce in one hour (corresponding to the power consumption). The "maximum power reduction rate" of the electrolyzer cluster is the maximum amount of hydrogen that the entire cluster can reduce by in one hour (corresponding to the power consumption). Let t be the total hydrogen production power of the electrolyzer cluster at time t; This represents the total hydrogen production power of the electrolyzer cluster at time t-1.

[0157] The cluster ramp rates are as follows: (42)

[0158] in, The "maximum power-up rate" for a single electrolyzer is the maximum amount of hydrogen that the electrolyzer can produce in one hour (corresponding to the power consumption). The "maximum power reduction rate" for a single electrolyzer is the maximum amount of hydrogen that can be reduced in one hour (corresponding to the power consumption).

[0159] Considering the cluster load factor, the hydrogen production efficiency model is as follows: (43)

[0160] in, The hydrogen production efficiency of the electrolyzer cluster; The rated hydrogen efficiency of the electrolyzer; The load factor of the electrolytic cell cluster; This represents the maximum load rate of the electrolytic cell cluster. The higher the load rate, the closer the efficiency is to the rated value.

[0161] S5.2, Establish an energy storage cluster model.

[0162] The aggregation capacity constraint is as follows: (44)

[0163] in, For the capacity of the energy storage cluster; The capacity of a single energy storage device; , These represent the minimum and maximum capacities of the energy storage cluster, respectively.

[0164] The charging and discharging power constraints are: (45)

[0165] in, The energy storage capacity of the energy storage cluster; The energy storage capacity of a single energy storage device; The energy release power of the energy storage cluster; This refers to the energy release power of a single energy storage device.

[0166] The self-discharge rate of the cluster is: (46)

[0167] in, Let be the self-discharge rate of the energy storage cluster at time t; Let be the self-discharge rate of the energy storage cluster at time t-1; To improve the charging efficiency of energy storage clusters; This refers to the energy release efficiency of the energy storage cluster. The maximum self-discharge rate of a single device is taken as a conservative design.

[0168] In this way, by treating hundreds of similar small devices as a few large devices, the amount of computation is greatly reduced without losing key physical characteristics, thus ensuring accurate results.

[0169] Example 2

[0170] This embodiment proposes a large-scale cross-regional integrated energy system and synergistic optimization system for electricity, heat and hydrogen, which includes the following structural modules:

[0171] (1) Network structure modeling upgrade construction module, used to construct a multi-level multi-energy flow coupling model for a large-scale cross-regional integrated energy system of electricity-heat-hydrogen, specifically including:

[0172] 1) The hierarchical division and association matrix submodule is used to define hierarchical sets, establish region-hierarchical association matrices and network-region association matrices;

[0173] 2) A refined modeling submodule for multi-energy flow transmission, used to establish power grid transmission models, hydrogen pipeline transmission models, and heat network transmission models;

[0174] 3) Multi-level coordination constraint submodule, used to establish upper-level power allocation constraints on lower-level layers and lower-level feedback constraints on upper-level layers.

[0175] (2) An algorithm efficiency optimization construction module, used to reduce the order of the large-scale electricity-heat-hydrogen cross-regional integrated energy system, specifically including:

[0176] 1) The block-based ADMM submodule is used to describe the problem decomposition logic, block objective function, coupling constraint handling method, and iterative update rules;

[0177] 2) Parallel computing acceleration submodule, used for thread allocation, data communication optimization and accelerated convergence;

[0178] 3) Model reduction submodule, used to filter key regions and simplify aggregated region variables;

[0179] (3) A multi-stakeholder interest distribution innovation module, used to innovate the multi-stakeholder interest distribution by adopting multi-factor weighted generalized Nash bargaining for the large-scale electricity-heat-hydrogen cross-regional integrated energy system, specifically including:

[0180] 1) Multi-dimensional contribution quantification submodule, used to quantify resource endowment contribution, network contribution, investment cost contribution and comprehensive contribution;

[0181] 2) An improved generalized Nash bargaining model submodule, used to describe the objective function, define the constraints, and calculate the Nash equilibrium solution;

[0182] 3) The inter-level benefit-sharing mechanism submodule is used to describe the cost-sharing method of the upper-level network and the cross-level subsidy mechanism;

[0183] (4) Uncertainty Robustness Construction Module, used to establish a distributed robust optimization model for the large-scale electricity-heat-hydrogen cross-regional integrated energy system, and to robustness the uncertainty, specifically including:

[0184] 1) Uncertainty set definition submodule, used to define the uncertainty set of wind and solar power output, the uncertainty set of load, and spatiotemporal correlation constraints;

[0185] 2) Distributed robust optimization objective submodule, used to establish the max-min objective function, calculate the penalty cost and solve the transformation, transforming the max-min problem into a single-level optimization through duality theory;

[0186] (5) Equipment cluster adaptation construction module, used for equipment aggregation modeling of the large-scale electricity-heat-hydrogen cross-regional integrated energy system, specifically including:

[0187] 1) Electrolyzer cluster model submodule, used to describe polymerization power constraints, cluster ramp-up constraints and hydrogen production efficiency model;

[0188] 2) Energy storage cluster model submodule, used to describe aggregated capacity constraints, charging and discharging power constraints, and cluster self-discharge rate; This embodiment describes a network structure modeling upgrade construction module and its functions for a cross-regional and cross-provincial large-scale electric-thermal-hydrogen cross-regional integrated energy system and collaborative optimization system, specifically including a hierarchical partitioning and correlation matrix module, a multi-energy flow transmission refined modeling module, and a multi-level coordination constraint module; describes an algorithm efficiency optimization construction module, specifically including a block-based ADMM module, a parallel computing acceleration module, and a model order reduction module; describes a multi-stakeholder interest distribution innovation construction module, specifically including a multi-dimensional contribution quantification module, an improved generalized Nash bargaining model module, and a hierarchical interest sharing mechanism module; describes an uncertainty robustness construction module, specifically including an uncertainty set definition module and a distributed robust optimization objective module; describes an equipment cluster adaptation construction module, specifically including an electrolyzer cluster model module and an energy storage cluster model module.

[0189] Example 3

[0190] Based on the same inventive concept, embodiments of the present invention also provide a computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system described in Embodiment 1.

[0191] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0192] Since the storage medium is the storage medium of a large-scale electrothermal hydrogen cross-regional integrated energy system collaborative optimization method according to an embodiment of the present invention, and the principle of the storage medium in solving the problem is similar to that of the method, the implementation of the storage medium can refer to the implementation process of the above-mentioned method embodiment one, and the repeated parts will not be described again.

[0193] In some possible implementations, various aspects of the methods of the embodiments of the present invention can also be implemented as a program product comprising program code that, when run on a computer device, causes the computer device to perform the steps of the sparse signal recovery method according to various exemplary embodiments of the present application described above. The executable computer program code or "code" for performing the various embodiments can be written in high-level programming languages ​​such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, Structured Query Language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0194] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above 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 one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0195] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system, characterized in that, include: A multi-level multi-energy flow coupling model is constructed for a large-scale cross-regional integrated energy system of electricity, heat and hydrogen to realize network structure modeling; the multi-level multi-energy flow coupling model includes hierarchical partitioning and correlation matrix, refined modeling of multi-energy flow transmission and multi-level coordination constraints; An improved adaptive ADMM algorithm for large-scale electric-thermal-hydrogen cross-regional integrated energy systems is adopted to reduce the order of the large-scale electric-thermal-hydrogen cross-regional integrated energy system. The large-scale electricity-heat-hydrogen cross-regional integrated energy system adopts multi-factor weighted generalized Nash bargaining to innovate the distribution of interests among multiple stakeholders; A distributed robust optimization model is established for the large-scale electricity-heat-hydrogen cross-regional integrated energy system to robusten the uncertainty; Equipment aggregation modeling is performed on the large-scale electricity-heat-hydrogen cross-regional integrated energy system to adapt the equipment cluster.

2. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 1, characterized in that, The hierarchical division and association matrix includes: Define the hierarchical set as L1 is a first-level IES, L2 is a second-level IES, and L3 is a third-level IES; The region-hierarchical association matrix is ​​represented as follows Where N is the number of regions, M is the number of levels, and A ij =1 indicates that the i-th region belongs to the j-th level, otherwise it is 0; The pipeline network-region correlation matrix is ​​represented as follows: Where K is the number of pipeline types, including electric / heat / hydrogen pipelines, and B kl =1 indicates that the k-type pipeline network covers the l-th area; otherwise, it is 0.

3. The collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 1, characterized in that, The refined modeling of multi-energy stream transmission includes: (1) Power grid transmission model (1) in, Provide power output for region i; The power transmitted from region i to j is represented by Ω; positive values ​​indicate transmission and negative values ​​indicate reception. i Represents the set of neighboring regions of region i; Let be the load power of region i; The charging power for the regional i-energy storage system; The discharge power of the regional i energy storage system; The power consumption of the electrolytic cell in region i; Using the B-coefficient method, the transmission loss model is expressed as: (2) in, Resistance per unit length; For transmission distance; The blocking constraint is: (3) in, This represents the upper limit of the line's transmission capacity. Inter-level power transmission constraints are: (4) This means that the total power received by the first-level layer from the second-level layer is equal to the power allocated by the first-level layer to each region; The hydrogen flow rate constraint is: (5) in, Let be the hydrogen flow rate in region i→j; This refers to the pipeline flow coefficient. For node pressure difference; Let be the hydrogen pressure in region i; Let be the hydrogen pressure in region j; The leakage loss model is as follows: (6) in, Leakage rate per unit length; The pressure constraint is: (7) in, Minimum pressure; For maximum pressure; The regional hydrogen balance constraint is: (8) in, The power consumption of the electrolytic cell in region i; The hydrogen production efficiency of the electrolyzer; The power generation capacity of the fuel cell in region i; For the power generation efficiency of fuel cells; The power of the hydrogen load in region i; Let be the hydrogen flow rate in region i→j; Let be the hydrogen leakage flow rate in region i→j; (3) Heat network transmission model Considering temperature decay, the heat transfer model for the heating network is as follows: The heat flow constraint is: (9) in, The heat flow rate from region i to j; The heat transfer coefficient of the heating network; For water supply temperature; The return water temperature; The temperature decay model is as follows: (10) in, The coefficient of temperature drop per unit length; The regional thermal balance constraint is: (11) in, The heating capacity of the gas-fired boiler (GB) in Zone i; The heating capacity of the regional i combined cooling, heating and power (CCHP) unit; The heat flow rate from region j to i; The heat load of region i; The thermal energy storage power of the regional i thermal energy storage system; The heat release power of the regional i thermal energy storage system.

4. The collaborative optimization method for a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 3, characterized in that, The multi-level coordination constraints include: The upper layer's power allocation constraint to the lower layer is: (12) in, Let be the power allocation coefficient for region i. ; The feedback constraint from the lower layer to the upper layer is: (13) This means that the surplus of the lower layer minus the deficit equals the transmission power of the upper layer.

5. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 3, characterized in that, The improved ADMM algorithm adapted for large-scale interregional integrated energy systems of electricity, heat, and hydrogen includes: Optimization problem of the large-scale interregional integrated energy system of electricity-heat-hydrogen The problem is decomposed into K subproblems to achieve block ADMM. The block objective function is: (14) in, The comprehensive energy system operating cost for region i; Revenue from inter-system transactions within the regional integrated energy system; Transaction costs when the regional integrated energy system engages in transactions with various markets; Consider the operating costs of demand response for the regional integrated energy system; Introducing a global coupling variable Z=AX, where A is the coupling matrix, the augmented Lagrangian function is constructed as follows: (15) The iterative update rules are as follows: 1) Regional subproblem update (16) in, This represents the updated value of the optimization variable for region i during the (k+1)th iteration; Let be the Lagrange multiplier vector for the k-th iteration. Transpose it; Let A be the coupling constraint matrix for region i. i Let A be the i-th row; Let be the global coupling variable for the k-th iteration; ρ is the augmented Lagrange penalty parameter; 2) Global Coupling Variable Update (17) It represents mean aggregation, ensuring global consistency; 3) Lagrange multiplier update (18)。 6. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 5, characterized in that, The improved ADMM algorithm adapted for large-scale interregional integrated energy systems of electricity, heat and hydrogen also includes: Each level / region is assigned an independent computation thread, and subproblems are solved in parallel. It adopts a local computation and global synchronization mode, and only transmits coupled variables during iterative updates; Introducing dynamic inertia weight ω k The updated formula is as follows: (19) By screening key regions and aggregating variables, the order of the large-scale cross-regional integrated energy system of electricity, heat and hydrogen can be reduced. The key region screening includes: 1) Construct a regional importance evaluation matrix: (20) in, The maximum transmission power from region i to j; 2) Calculate the regional importance score: (21) Where α is the damping coefficient; 1 is an all-1 vector; 3) Filtering threshold: The top 80% of the scores are retained as key areas, and the rest are aggregated areas; The variable aggregation includes: The formula for weighted aggregation of similar variables from multiple aggregation regions into a single equivalent variable is as follows: (22) in, The aggregate weights are allocated based on the regional load percentage.

7. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 1, characterized in that, The large-scale electricity-heat-hydrogen cross-regional integrated energy system adopts multi-factor weighted generalized Nash bargaining to innovate the distribution of interests among multiple stakeholders, including: Multi-dimensional quantitative contributions, including: Resource endowment contribution The quantification formula is: (23) Among them, it means The percentage of maximum wind and solar power output in the region; To achieve the maximum photovoltaic output for region i; The maximum wind power output of region i; Network contribution The quantification formula is: (24) in, This represents the proportion of pipeline transmission capacity in region i. The maximum pipeline transmission capacity for region i→j; The maximum pipeline transmission capacity for region j→i; Investment cost contribution The quantification formula is: (25) This represents the percentage of equipment investment costs in region i. Overall contribution The quantification formula is: (26) in, The weights can be determined using the analytic hierarchy process (AHP). Establish an improved generalized Nash bargaining model, including: Considering the weights of multiple factors, the objective function is established as follows: (27) in, Benefits after regional cooperation; Non-cooperative income; For negotiation parameters; Assess overall contribution; The constraints are: (28) This indicates that the total revenue of the alliance is conserved; Then, by transforming the logarithmic objective into a linear objective, the distribution of benefits in each region, i.e., the Nash equilibrium solution, is obtained: (29) Establish a mechanism for sharing benefits among different levels; The cost-sharing mechanism for the upper-layer network is as follows: first, calculate the sum of the construction and maintenance costs of the first-layer network, and then allocate the costs according to the transmission power ratio of the lower-layer areas, as shown in the following formula: (30) (31) in, This is the sum of the construction and maintenance costs of the first-level pipeline network; The cost of pipeline construction and maintenance for region i; This represents the cost allocation for the first-tier network. This refers to the transmission power of the first-level pipeline network. This refers to the transmission power of the second-level pipeline network. The cross-level subsidy mechanism is as follows: resource-rich lower-level regions receive subsidies from higher-level regions, and the formula is: (32) in, This is the subsidy coefficient.

8. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 1, characterized in that, A distributed robust optimization model is established for the large-scale inter-regional integrated energy system of electricity, heat, and hydrogen to robustly address uncertainties, including: Considering spatiotemporal correlation, we define an uncertainty set, including: The set of uncertainties in wind and solar power output is: (33) in, This refers to the intraday volatility coefficient. This is the upper limit of the total daily fluctuation. The set of load uncertainties is: (34) The spatiotemporal constraints are: (35) This indicates that daytime photovoltaic power in region i and evening photovoltaic power in region j are complementary; Define the distributed robustness optimization objectives, including: The objective function is: (36) in, The calculation formula for the penalty cost is as follows: (37) As punishment for abandoning wind and light, Penalty for load shortage; Using duality theory, the max-min problem is transformed into a single-level optimization problem. By introducing an auxiliary variable θ, it becomes: (38)。 9. The method for collaborative optimization of a large-scale electrothermal hydrogen cross-regional integrated energy system as described in claim 1, characterized in that, Equipment aggregation modeling is performed on the large-scale electricity-heat-hydrogen cross-regional integrated energy system to adapt the equipment cluster, including: Establish an electrolytic cell cluster model, including: The polymerization power constraint is as follows: (39) in, This refers to the load power of a single electrolytic cell; This refers to the load power of the electrolytic cell cluster; The minimum and maximum output of the cluster are as follows: (40) in, This represents the minimum output of the electrolytic cell cluster. This is the maximum output of the electrolytic cell cluster; The cluster ramping constraint is: (41) in, The maximum power-up rate of the electrolytic cell cluster; The maximum power reduction rate of the electrolytic cell cluster; Let t be the total hydrogen production power of the electrolyzer cluster at time t; The total hydrogen production power of the electrolyzer cluster at time t-1; The cluster ramp-up rates are as follows: (42) in, This refers to the "maximum power-up rate" of a single electrolytic cell; This refers to the "maximum power reduction rate" of a single electrolytic cell; Considering the cluster load factor, the hydrogen production efficiency model is as follows: (43) in, The hydrogen production efficiency of the electrolyzer cluster; The rated hydrogen efficiency of the electrolyzer; The load factor of the electrolytic cell cluster; This represents the maximum load rate of the electrolytic cell cluster. Establish an energy storage cluster model, including: The aggregation capacity constraint is as follows: (44) in, For the capacity of the energy storage cluster; The capacity of a single energy storage device; , These are the minimum and maximum capacities of the energy storage cluster, respectively. The charge / discharge power constraint is: (45) in, The energy storage capacity of the energy storage cluster; The energy storage capacity of a single energy storage device; The energy release power of the energy storage cluster; The energy release power of a single energy storage device; The self-discharge rate of the cluster is: (46) in, Let be the self-discharge rate of the energy storage cluster at time t; Let be the self-discharge rate of the energy storage cluster at time t-1; To improve the charging efficiency of energy storage clusters; The energy release efficiency of the energy storage cluster; Take the maximum self-discharge rate of a single device.

10. A collaborative optimization system for a large-scale electrothermal hydrogen cross-regional integrated energy system, characterized in that, include: The network structure modeling upgrade construction module is used to build multi-level multi-energy flow coupling models for large-scale cross-regional integrated energy systems of electricity, heat, and hydrogen, including: 1) The hierarchical division and association matrix submodule is used to define hierarchical sets, establish region-hierarchical association matrices and network-region association matrices; 2) A refined modeling submodule for multi-energy flow transmission, used to establish power grid transmission models, hydrogen pipeline transmission models, and heat network transmission models; 3) Multi-level coordination constraint submodule, used to establish upper-level power allocation constraints on lower-level layers and lower-level layers feedback constraints to upper-level layers; An algorithm efficiency optimization construction module is used to reduce the order of the large-scale electricity-heat-hydrogen inter-regional integrated energy system; it includes: 1) The block-based ADMM submodule is used to describe the problem decomposition logic, block objective function, coupling constraint handling method, and iterative update rules; 2) Parallel computing acceleration submodule, used for thread allocation, data communication optimization and accelerated convergence; 3) Model reduction submodule, used to filter key regions and simplify aggregated region variables; A multi-stakeholder interest distribution innovation module is used to innovate the multi-stakeholder interest distribution by employing multi-factor weighted generalized Nash bargaining for the large-scale electricity-heat-hydrogen inter-regional integrated energy system, including: 1) Multi-dimensional contribution quantification submodule, used to quantify resource endowment contribution, network contribution, investment cost contribution and comprehensive contribution; 2) An improved generalized Nash bargaining model submodule, used to describe the objective function, define the constraints, and calculate the Nash equilibrium solution; 3) The inter-level benefit-sharing mechanism submodule is used to describe the cost-sharing method of the upper-level network and the cross-level subsidy mechanism; An uncertainty robustness construction module establishes a distributed robust optimization model for the large-scale electricity-heat-hydrogen inter-regional integrated energy system, robustening the uncertainty, including: 1) Uncertainty set definition submodule, used to define the uncertainty set of wind and solar power output, the uncertainty set of load, and spatiotemporal correlation constraints; 2) Distributed robust optimization objective submodule, used to establish the max-min objective function, calculate the penalty cost and solve the transformation, transforming the max-min problem into a single-level optimization through duality theory; The equipment cluster adaptation construction module, used for equipment aggregation modeling of the large-scale electricity-heat-hydrogen cross-regional integrated energy system, includes: 1) Electrolyzer cluster model submodule, used to describe polymerization power constraints, cluster ramp-up constraints and hydrogen production efficiency model; 2) Energy storage cluster model submodule, used to describe aggregated capacity constraints, charge and discharge power constraints and cluster self-discharge rate.

Images