An electric-gas-heat integrated energy optimization scheduling and carbon flow calculation method and system

Abstract

This invention belongs to the field of integrated energy optimization scheduling technology, and provides a method and system for integrated energy optimization scheduling and carbon flow calculation of electricity-gas-heat, including: acquiring a physical power flow model and a carbon emission flow model of integrated energy (electricity-gas-heat); constructing a daily-level physical optimization scheduling model based on the acquired physical power flow model, and generating daily-level optimal scheduling samples; predicting the coupled scheduling of integrated energy based on the generated daily-level optimal scheduling samples and combining them with a daily-level graph convolutional network; obtaining the operational constraints of integrated energy based on the obtained coupled scheduling results, and calculating the carbon flow of integrated energy in combination with the acquired carbon emission flow model, thereby completing the optimized scheduling and carbon flow calculation of integrated energy (electricity-gas-heat).

Description

Technical Field

[0001] This invention belongs to the field of integrated energy optimization scheduling technology, specifically relating to a method and system for integrated energy optimization scheduling and carbon flow calculation of electricity-gas-heat. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the large-scale integration of renewable energy sources such as wind power and photovoltaics into integrated energy systems, the coupling between power systems, heating systems, and natural gas systems is deepening. The energy flow within integrated energy systems exhibits complex characteristics of multiple sources, nodes, paths, and time scales. Combined heat and power (CHP) equipment, power-to-gas (EPG) equipment, electric boilers, carbon capture devices, and energy storage devices form close energy coupling relationships between electricity, gas, and heat networks, making traditional single-energy system power flow calculation and optimization scheduling methods difficult to apply to scenarios with deep multi-energy coupling.

[0004] Most existing integrated energy optimization scheduling methods are based on physical mechanism models and mathematical programming models, solving scheduling schemes by constructing constraints on electricity, gas, and heat networks and equipment operation constraints. However, when it is necessary to repeatedly solve for a large number of wind and solar load disturbance scenarios, the computational burden is heavy, making it difficult to meet the fast computation requirements of real-time or near-real-time scheduling. Especially in scenarios with multiple coupled devices such as cogeneration, carbon capture, power-to-gas conversion, and energy storage, the optimization model contains a large number of continuous variables, coupling constraints, and time-series constraints, further increasing the time cost of traditional optimization solutions.

[0005] Furthermore, with the development of artificial intelligence technology, data-driven methods such as deep neural networks and graph convolutional networks are increasingly being applied to power system power flow calculations and optimal power flow calculations. Because graph convolutional networks can aggregate information from adjacent nodes using the system topology, they are suitable for handling objects with inherent network structures, such as power grids, heating networks, and gas networks. However, existing data-driven power flow calculation methods mostly focus on power systems or use neural networks merely as pure data fitting tools, lacking a unified characterization of the coupling relationships between multiple energy flows in the electricity-gas-heat integrated energy system, equipment operating boundaries, energy storage timing constraints, and carbon emission flow conservation relationships. Directly using the scheduling results or carbon flow results from neural networks can easily lead to problems such as power imbalance, equipment exceeding limits, infeasible energy storage states, or carbon flow transfer not conforming to the proportional sharing principle.

[0006] Furthermore, the calculation of integrated energy carbon emission flows has a clear physical meaning. Carbon emissions from the power system are transmitted with active power flow, carbon emissions from the heating system are transmitted with the heat medium flow, and carbon emissions from the natural gas system are transmitted with the gas flow. Combined heat and power (CHP), power-to-gas conversion, electric boilers, carbon capture and storage devices cause the transfer and redistribution of carbon emission responsibility among different energy networks. Therefore, carbon flow calculation should not be directly replaced by neural networks, but should be calculated based on scheduling and power flow results that meet physical constraints, in accordance with the principles of carbon emission flow conservation, proportional sharing, and upstream mixing. Summary of the Invention

[0007] To address the aforementioned issues, this invention proposes a method and system for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat. It introduces a daily-level graph convolutional network to achieve rapid prediction of the coupled electricity, gas, and heat scheduling results, and ensures that the data-driven model output meets the operational constraints of the integrated energy system through a physical loss function and a feasibility projection layer. Simultaneously, it decouples carbon flow calculation from neural network prediction, avoiding insufficient physical consistency caused by direct neural network prediction of carbon flow. This achieves collaborative modeling of economic scheduling, rapid power flow calculation, and unified carbon emission flow tracking for the integrated energy system, improving the system's rapid optimization scheduling capability and refined carbon responsibility tracking capability under multi-source uncertain scenarios.

[0008] According to some embodiments, the first aspect of the present invention provides a method for integrated energy optimization scheduling and carbon flow calculation of electricity, gas, and heat, employing the following technical solution: A method for integrated energy scheduling and carbon flow calculation combining electricity, gas, and heat includes: Physical flow model and carbon emission flow model for obtaining integrated electricity-gas-heat energy; Based on the obtained physical power flow model, a daily-level physical optimization scheduling model is constructed, and daily-level optimal scheduling samples are generated. Based on the generated daily optimal scheduling samples, combined with a daily graph convolutional network, the coupled scheduling of integrated energy is predicted. Based on the obtained coupled scheduling results, the operational constraints of integrated energy are obtained. Combined with the obtained carbon emission flow model, the carbon flow of integrated energy is calculated, and the optimal scheduling and carbon flow calculation of integrated energy (electricity, gas, and heat) are completed.

[0009] As a further technical limitation, the integrated electricity-gas-heat energy system includes an electric power system, a heating system, and a natural gas system; wherein, the electric power system includes wind turbine generators, photovoltaic generators, a power purchase interface with the upstream power grid, a combined heat and power (CHP) unit, an electric load node, and an electric energy storage device; the heating system includes a heat source, a heat load node, a heating network pipeline, an electric boiler, and a thermal energy storage device; the natural gas system includes a gas source, a gas load node, a gas network pipeline, an electric-to-gas conversion device, and a gas storage device; the CHP unit connects the natural gas system, the electric power-to-gas conversion device connects the electric power system and the natural gas system; the electric boiler connects the electric power system and the heating system; and both the CHP unit and the electric-to-gas conversion device are associated with carbon capture equipment.

[0010] As a further technical constraint, the physical flow model of the integrated electricity-gas-heat energy system obtained includes at least the power balance constraints of the power system, the power balance constraints of the thermal system, the balance constraints of the natural gas system, the operating constraints of the energy storage equipment, the coupling relationship of the cogeneration equipment, the coupling relationship of the power-to-gas equipment, the coupling relationship of the electric boiler equipment, and the energy consumption relationship of the carbon capture equipment.

[0011] As a further technical limitation, the constructed daily-level physical optimization scheduling model takes economic optimization as the objective function, that is, ;in, For time-of-use electricity pricing, For gas prices, For the power purchase capacity, To purchase gas volume, For equipment operation and maintenance costs, The penalty coefficient for wind and solar power curtailment. and These are the power curtailed from wind and the power curtailed from solar power, respectively.

[0012] As a further technical limitation, the day-level graph convolutional network includes a topological graph of the integrated energy conversion branches and its corresponding adjacency matrix, as well as a graph convolutional network model; wherein, the graph convolutional network model adopts a graph convolutional structure without attention weights.

[0013] As a further technical limitation, the operational constraints of the integrated energy system are physical constraints, and the loss function of the physical constraints is: ;in, For data fitting loss; Loss due to physical constraints; This is the regularization loss; represents the physical loss weights in the k-th training iteration.

[0014] According to some embodiments, the second aspect of the present invention provides an integrated energy optimization scheduling and carbon flow calculation system for electricity, gas, and heat, employing the following technical solution: A comprehensive energy optimization scheduling and carbon flow calculation system integrating electricity, gas, and heat, comprising: The acquisition module is configured to acquire the physical flow model of integrated electric-gas-heat energy and the carbon emission flow model; The generation module is configured to construct a daily-level physical optimization scheduling model based on the acquired physical power flow model and generate daily-level optimal scheduling samples. The prediction module is configured to predict the coupled scheduling of integrated energy resources based on the generated daily optimal scheduling samples and combined with a daily graph convolutional network. The optimization module is configured to obtain the operational constraints of integrated energy based on the obtained coupled scheduling results, calculate the carbon flow of integrated energy in combination with the obtained carbon emission flow model, and complete the optimized scheduling and carbon flow calculation of integrated energy of electricity, gas and heat.

[0015] According to some embodiments, a third aspect of the present invention provides a computer-readable storage medium, employing the following technical solution: A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity-gas-heat as described in the first aspect of the present invention.

[0016] According to some embodiments, the fourth aspect of the present invention provides an electronic device, which adopts the following technical solution: An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the integrated energy optimization scheduling and carbon flow calculation method for electricity, gas, and heat as described in the first aspect of the present invention.

[0017] According to some embodiments, the fifth aspect of the present invention provides a computer program product, which adopts the following technical solution: A computer program product includes software code, wherein the program in the software code performs the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity-gas-heat as described in the first aspect of the present invention.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention directly learns the mapping relationship from a day's wind, solar, and load price scenario to the day's optimal scheduling result through a daily graph convolutional network, avoiding insufficient temporal consistency caused by hourly independent predictions; through physical constraint loss and feasibility projection layer, the neural network output can meet the physical constraints of integrated energy (electricity, gas, and heat); through a data-driven method, the integrated energy carbon flow calculation result is calculated, ensuring that the carbon flow result meets the unified carbon emission flow model, thereby achieving the unification of rapid integrated energy carbon flow calculation and carbon emission responsibility tracking. Attached Figure Description

[0019] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0020] Figure 1 This is a flowchart of a method for integrated energy optimization scheduling and carbon flow calculation based on electricity, gas, and heat in Embodiment 1 of the present invention; Figure 2 This is an architectural diagram of an integrated electric-gas-heat energy system according to Embodiment 1 of the present invention; Figure 3 This is a flowchart of the daily training sample generation process in Embodiment 1 of the present invention; Figure 4 This is an architecture diagram of the physical constraint-based graph convolutional network scheduling prediction model in Embodiment 1 of the present invention; Figure 5 This is a structural block diagram of an integrated energy optimization scheduling and carbon flow calculation system based on the electricity-gas-heat model, as shown in Embodiment 2 of the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0024] In this invention, terms such as "upper," "lower," "left," "right," "front," "back," "vertical," "horizontal," "side," and "bottom" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used only to facilitate the description of the structural relationships of the various components or elements of this invention and do not specifically refer to any component or element in this invention. They should not be construed as limiting the invention.

[0025] In this invention, terms such as "fixed connection," "connected," and "linked" should be interpreted broadly, indicating a fixed connection, an integral connection, or a detachable connection; a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can determine the specific meaning of these terms in this invention based on the specific circumstances, and they should not be construed as limitations on the invention.

[0026] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0027] Example 1 Embodiment 1 of this invention introduces a method for integrated energy optimization scheduling and carbon flow calculation based on electricity, gas, and heat.

[0028] like Figure 1 and Figure 4 The method for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat, as shown, includes: Physical flow model and carbon emission flow model for obtaining integrated electricity-gas-heat energy; Based on the obtained physical power flow model, a daily-level physical optimization scheduling model is constructed, and daily-level optimal scheduling samples are generated. Based on the generated daily optimal scheduling samples, combined with a daily graph convolutional network, the coupled scheduling of integrated energy is predicted. Based on the obtained coupled scheduling results, the operational constraints of integrated energy are obtained. Combined with the obtained carbon emission flow model, the carbon flow of integrated energy is calculated, and the optimal scheduling and carbon flow calculation of integrated energy (electricity, gas, and heat) are completed.

[0029] As one or more implementation methods, this embodiment establishes an integrated energy physical flow model and a unified carbon emission flow model that includes wind power, photovoltaics, combined heat and power, carbon capture, electricity-to-gas conversion, electric boilers, and energy storage equipment; specifically: like Figure 2 As shown, the integrated energy system in this embodiment includes a power system, a heating system, and a natural gas system. The power system includes wind turbine generators, photovoltaic generators, a power purchase interface with the upstream power grid, a combined heat and power (CHP) unit, electrical load nodes, and electrical energy storage equipment. The heating system includes heat sources, heat load nodes, heating network pipelines, electric boilers, and thermal energy storage equipment. The natural gas system includes gas sources, gas load nodes, gas network pipelines, electricity-to-gas conversion equipment, and gas storage equipment. The CHP unit connects to the gas network, power grid, and heating network; the electricity-to-gas conversion equipment connects to the power grid and gas network; the electric boiler connects to the power grid and heating network; and the carbon capture equipment is associated with the CHP unit and the electricity-to-gas conversion equipment.

[0030] The power system power balance constraint in this embodiment is: ;in, For the power purchased; For the electrical output of combined heat and power units; and They provide power for wind power and solar power, respectively. and These are the energy storage discharge and charging power, respectively. This refers to the power consumption of the electricity-to-gas conversion process. The power consumption of the electric boiler; Power consumption of carbon capture equipment; For electrical load; This refers to power grid losses.

[0031] The power balance constraint of the thermal system in this embodiment is: ;in, Provides heating power to combined heat and power units; The electrothermal conversion efficiency of the electric boiler; and These are thermal energy storage heat release and thermal storage power, respectively. For heat load; This refers to heat loss in the heating network pipeline.

[0032] The natural gas system balance constraint in this embodiment is: ;in, For the amount of gas purchased; The electro-gas conversion coefficient; and These refer to the gas release from the gas storage device and the gas storage capacity, respectively. This refers to the gas consumption of a combined heat and power (CHP) unit. For other natural gas loads; For gas network losses; in an embodiment considering only gas turbine demand and gas storage tanks, Can be taken as zero or merged. .

[0033] The operating constraints of the energy storage device in this embodiment are: ; ; ;in, In energy storage state; This is the self-loss coefficient; and These are the charge / discharge efficiencies; and These are the maximum charging and discharging power, respectively. and These represent the upper and lower limits of the energy storage state, respectively.

[0034] The coupling relationship of the cogeneration equipment in this embodiment is as follows: ; ;in, and These refer to the power generation efficiency and heating efficiency of the combined heat and power (CHP) equipment, respectively. It represents the lower heating value of natural gas.

[0035] The coupling relationship of the electro-gas conversion equipment in this embodiment is as follows: ;in, The gas power of the electro-gas conversion equipment.

[0036] The coupling relationship of the electric boiler equipment in this embodiment is as follows: ;in, This refers to the heat output power of the electric boiler equipment.

[0037] The energy consumption relationship of the carbon capture equipment in this embodiment is as follows: ;in, Carbon capture amount; The power consumption coefficient per unit of capture volume.

[0038] The unified carbon emission flow model in this embodiment is: ; ; ; ; ; in, The generalized branch flux matrix represents the carbon flow direction in the overall energy mix, where... This represents the power flow balance matrix within subsystem x, including the unit injection from energy conversion equipment. This represents the carbon emission coupling matrix of the energy coupling device in different energy systems; x and y are system symbols, x,y∈{E,H,G}, where E represents the power system, H represents the thermal system, and G represents the natural gas system; For the generalized carbon potential matrix of the power system, For the generalized carbon potential matrix of the thermodynamic system, The generalized carbon potential matrix of the natural gas system; express The diagonal elements are the sum of the node-injected power; Represents the set of branches ji that are injected into node i; for Off-diagonal elements The off-diagonal elements are not 0; This indicates that the energy conversion device is located at node n of system x. x To node n of system y y The injected power is taken as a negative value. Taking an electric boiler as an example, when the power system injects power into the heating system... This indicates that the electric boiler is located at node n1 of the power system and is injecting power into node n2 of the thermal system. The n2*n1 position; A comprehensive energy generalized carbon potential matrix; The carbon potential matrix for injection includes the corresponding carbon potential injected by external units. This corresponds to the branch distribution matrix of the system; Inject the distribution matrix into the corresponding external units of the system; Let x be the number of nodes in system x; Let y be the number of nodes in system y; Power is injected into the branch flowing from node j to node i; The generator injection power at node i; In the matrix middle The off-diagonal elements at a given position are the negative values ​​of the power injected into node i by the corresponding branch j.

[0039] The power system node carbon potential calculation model in this embodiment is as follows: ;in, Represents the nodal carbon potential of node i; , These represent the injection power and carbon flux density of branch S, respectively. Represents the set of branches s connected to node i; and These represent the generator power and generator carbon emission intensity connected to the node, respectively.

[0040] In this embodiment, the node carbon potential of the thermal system is: ;in, Represents the nodal carbon potential of node i; The heat power flowing from node j to node i; and Let i be the set of upstream inflow nodes and the set of downstream outflow nodes, respectively. Let i be the heat source power at node i; This represents the set of nodes in a thermal system.

[0041] The nodal carbon potential model of the natural gas system in this embodiment is as follows: ;in, Let be the nodal carbon potential of node i; The flow rate of natural gas from node j to node i; Let i be the load flow at node i; Let i be the set of upstream inflow nodes and downstream outflow nodes, respectively. The equivalent carbon emission rate injected at node i by an external gas source (such as a natural gas well or a P2G injection point); This represents the set of nodes in the gas system.

[0042] The node carbon potential model for renewable energy power generation equipment in this embodiment is as follows: ;in, This represents the carbon potential of the renewable energy generation equipment connected to node i at time t.

[0043] The node carbon potential model for the combined heat and power unit in this embodiment is as follows: ; ; ; in, The nodal carbon potential at the gas network node; The flow rate of natural gas consumed by the CHP unit; and These represent the electrical and thermal power generated by CHP, respectively. The thermoelectric ratio of the CHP unit; For CHP heat production efficiency; The power generation efficiency of CHP. The Carnot factor characterizes the difference in quality between thermal and electrical energy.

[0044] The carbon potential model for the carbon capture device node in this embodiment is as follows: ;in, The equivalent carbon potential of CCS itself; For carbon capture efficiency; This refers to the proportion of flue gas diversion. This represents the amount of CO2 absorbed by the CCS. The electrical power consumed by the CCS; ; ; The original carbon emission factor of CHP without CCS; The equivalent factor for carbon emissions after treatment; This represents the electrical power output by CHP after connection to CCS.

[0045] The node carbon potential model for the electro-gas conversion equipment in this embodiment is as follows: ;in, Carbon potential for P2G access to gas grid nodes; The node carbon potential of power system node i connected via P2G; This refers to the power consumption of P2G devices; The gas power generated for P2G equipment.

[0046] The node carbon potential model for the electric boiler in this embodiment is as follows: ;in, The node carbon potential corresponding to the electric boiler thermal system; The electrical power consumed by the electric boiler; Let be the node carbon potential of EB access node i at time t; The thermal power produced by the electric boiler.

[0047] The node carbon potential model for energy storage devices (including electrical energy storage, thermal energy storage, and gas storage tanks) in this embodiment is as follows: ;in, The internal carbon potential of the energy storage connected to node i; and The energy storage capacity states at time t and time t-1 are respectively. This refers to the self-loss rate; and These are the charging and discharging power, respectively. The internal carbon potential for energy storage at time t-1; This represents the node carbon potential of the energy storage device connected to node i.

[0048] As one or more implementation methods, this embodiment constructs a daily-level physical optimization scheduling model and generates neural network training labels; specifically: The daily-level physical optimization scheduling model in this embodiment includes an economically optimal objective function and corresponding constraints, and generates corresponding labels after training. The objective function of the daily-level physical optimization scheduling model is: ;in, For time-of-use electricity pricing, For gas prices, For the power purchase capacity, To purchase gas volume, For equipment operation and maintenance costs, The penalty coefficient for wind and solar power curtailment. and This refers to the power of wind and solar power that has been curtailed.

[0049] like Figure 3 As shown, this embodiment constructs daily-level training samples and node features, specifically: One daily training sample corresponds to a complete 24-hour scheduling cycle. Input data includes wind power output, photovoltaic power output, electricity load, heat load, gas load, electricity price, gas price, carbon price, node type code, node normalized number, and node nominal status.

[0050] The input sample on day d is represented as ;in, Input the sample for day d; Let N be the node feature matrix for time period t; N is the number of nodes. Let be the dimension of the node features.

[0051] Node characteristics may include node type encoding, node normalized number, wind power output, photovoltaic power output, electrical load, heat load, gas load, electricity price, and gas price. By solving the above daily-level physical optimization model, a strictly optimal scheduling result for 24 hours is obtained, which is used as a label for supervised learning of the neural network.

[0052] In this embodiment, the daily optimal scheduling sample is generated by a rigorous physical optimization scheduling model. One training sample corresponds to a 24-hour scheduling cycle, and the daily label is represented as: ; ; in, Tag for daily training; Let t be the vector of scheduling variables for time period t; The operating cost for time period t.

[0053] As one or more implementation methods, this embodiment constructs a daily-level graph convolutional network oriented towards a 24-hour scheduling cycle, directly predicting the combined heat and power output, energy storage charging and discharging power, electricity-to-gas power, electric boiler power, purchased electricity power, purchased gas power, optimal scheduling cost, and calculating carbon flow results within 24 hours; specifically: In this embodiment, the daily graph convolutional network for a 24-hour scheduling cycle includes a topology graph containing integrated energy conversion branches and a corresponding adjacency matrix. The graph convolutional network model adopts a graph convolutional structure without attention weights and a loss function for model training.

[0054] The integrated energy diagram structure in this embodiment can be represented as follows: ;in, It is a set of nodes, including power grid nodes, heating network nodes, gas network nodes, and energy conversion equipment nodes; The set of edges includes power line edges, heating network pipeline edges, gas network pipeline edges, and energy conversion equipment coupling edges; It is an adjacency matrix.

[0055] The adjacency matrix A is used to describe the connection relationships between nodes, and is obtained after adding self-loops and normalization. ;in, It is the identity matrix; for The corresponding degree matrix; This is the normalized adjacency matrix after adding self-loops.

[0056] The daily-level graph convolutional network shares the same set of graph convolution parameters in each time period; for the t-th time period, the topology aggregation and feature transformation process is as follows: ;in, The node features after topological aggregation in time period t; The input sample is for day t.

[0057] The graph convolutional hidden layer is represented as follows: ;in, Represented as hidden layer nodes; The weight matrix of the graph convolutional layer; It is the bias vector; It is a vector of all ones, used to extend the bias to all nodes; It is a linear rectification activation function.

[0058] Graph pooling is ;in, The system-level graph representation for time period t; Let be the hidden layer vector of the i-th node in time period t.

[0059] The daily representation is obtained by concatenating the 24-hour system-level features. ;in, This indicates vector concatenation; This is a daily chart representing a 24-hour period.

[0060] The daily output layer is ;in, The graph convolutional network predicts the scheduling results for a 24-hour period. and These are the weight matrix and bias of the daily output layer, respectively.

[0061] This embodiment employs a graph convolutional structure without attention weights because the physical feasibility of integrated energy is primarily guaranteed by the physical loss function and the feasibility projection layer. While attention weights can improve the model's expressive power, they can easily increase the parameter size and overfitting risk in scenarios with limited training samples and fixed system topology. Using an attention-free graph convolutional structure allows for stable learning of topological adjacency relationships with fewer parameters, improving the stability of model training and prediction.

[0062] The loss function of physical constraints in this embodiment, i.e., the objective function, is: ;in, For data fitting loss; Loss due to physical constraints; This is the regularization loss; represents the physical loss weights in the k-th training iteration.

[0063] The data fitting loss is ;in, For training days; This is the network prediction result for day d; The label for day d generated for the rigorous physical optimization model; The scaling factor for the output variables; To prevent extremely small positive numbers with a denominator of zero.

[0064] Physical constraint loss is ;in, , and These are the residuals of the energy balance for electricity, heat, and gas, respectively. For equipment capacity constraint residuals; For energy storage state constraint residuals; Residuals are constrained for network security; This indicates that only the portion of the constraint that exceeds the boundary will be penalized.

[0065] Regularization loss is ;in, This is the weight decay coefficient; It is the Frobenius norm; It can be determined based on the number of training samples and the output dimension. ;in, Given a regular strength reference value, This is a daily-level output dimension.

[0066] To avoid directly assigning physical loss weights based on experience, this embodiment uses gradient equalization to adaptively determine the physical loss weights, i.e. ;in, For smoothing coefficients; For neural network parameters; and These represent the lower and upper bounds of the physical loss weights, respectively. The physical loss weight formula is used to keep the gradient magnitudes of the data fitting loss and the physical constraint loss relatively balanced during parameter updates. This avoids the physical loss being too large, making it difficult for the model to fit the optimal scheduling label, and also avoids the physical loss being too small, causing the network to output "physically infeasible".

[0067] As one or more implementation methods, this embodiment sets up a physical feasibility judgment and projection layer. This layer is set after the graph convolutional network output. When the network output does not meet the system's physical constraints, the feasible region is reconstructed based on the current scenario. After satisfying the feasibility constraints, the power flow and carbon flow calculation results are output. Specifically: After the neural network outputs, the original prediction result is first evaluated. Does it satisfy the physical constraints of the current scenario? If it does, then let... = If not, the feasible region will be reconstructed based on the current wind power, solar power, load, and price scenarios. ,Right now ;in, This represents the comprehensive energy feasible domain under the current scenario s; For the set of equipment capacity and ramp constraints; It is a set of constraints for energy storage state, charge / discharge power, and the first and last states of the cycle; It is a set of constraints for the balance of electrical, thermal, and gaseous energy; A constrained set of coupling equipment such as cogeneration, power-to-gas conversion, electric boilers, and carbon capture; It is a set of network security constraints for power grids, heating networks, and gas networks.

[0068] The optimization problem for solving the projection layer is as follows: ;in, The projected feasible scheduling result; The original prediction results of the graph convolutional network; Let k be the scheduling variable for time period t; The scale weight is the weight of the k-th type variable. This weight can prevent large-volume variables from occupying too large a proportion in the projected distance, and make variables of different dimensions comparable. The standard deviation is determined by the standard deviation of the labels of the k-th class of variables in the training set, i.e. ; The standard deviation of the output variable label of the k-th class in the training set is, i.e. , Let be the label value of class k in time period t on day d. This standard deviation is not calculated only for a single 24-hour period, but is statistically obtained over all training days and all 24-hour periods.

[0069] The projection layer is not a simple variable-wise clipping, but rather a joint correction of all coupled constraints within the feasible region of the current scenario. If the feasible region is not empty and the solver solves successfully, the projection result simultaneously satisfies the device boundary, energy storage boundary, electrothermal-gas balance, and network security constraints. For extreme scenarios where the feasible region is empty, slack variables can be introduced and solved, i.e. ;in, A large penalty coefficient; Let be the slack variable for the i-th constraint; Equality constraint; For inequality constraints; if all Then the projection result strictly satisfies the physical constraints; if there exists If the condition is not met, it indicates that there is a constraint conflict or infeasibility in the current extreme scenario, and the location and degree of violation need to be output.

[0070] This embodiment directly learns the mapping relationship from a day's wind, solar, and load price scenario to the day's optimal scheduling result through a daily graph convolutional network, avoiding insufficient temporal consistency caused by hourly independent predictions; through physical constraint loss and feasibility projection layer, the neural network output can meet the physical constraints of integrated energy (electricity, gas, and heat); through a data-driven method, the integrated energy carbon flow calculation result is calculated, ensuring that the carbon flow result meets the unified carbon emission flow model, thereby achieving the unification of rapid integrated energy carbon flow calculation and carbon emission responsibility tracking.

[0071] This embodiment is expected to further promote the application of integrated energy in the field of low-carbon energy, and provide strong support for achieving the green transformation of the energy system and carbon emission reduction goals.

[0072] Example 2 Embodiment 2 of the present invention introduces an integrated energy optimization scheduling and carbon flow calculation system for electricity, gas and heat.

[0073] like Figure 5 The illustrated integrated energy optimization scheduling and carbon flow calculation system for electricity, gas, and heat includes: The acquisition module is configured to acquire the physical flow model of integrated electric-gas-heat energy and the carbon emission flow model; The generation module is configured to construct a daily-level physical optimization scheduling model based on the acquired physical power flow model and generate daily-level optimal scheduling samples. The prediction module is configured to predict the coupled scheduling of integrated energy resources based on the generated daily optimal scheduling samples and combined with a daily graph convolutional network. The optimization module is configured to obtain the operational constraints of integrated energy based on the obtained coupled scheduling results, calculate the carbon flow of integrated energy in combination with the obtained carbon emission flow model, and complete the optimized scheduling and carbon flow calculation of integrated energy of electricity, gas and heat.

[0074] The detailed steps are the same as those provided in Example 1 for the integrated energy optimization scheduling and carbon flow calculation method of electricity-gas-heat, and will not be repeated here.

[0075] Example 3 Embodiment 3 of the present invention provides a computer-readable storage medium.

[0076] A computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity-gas-heat as described in Embodiment 1 of the present invention.

[0077] The detailed steps are the same as those provided in Example 1 for the integrated energy optimization scheduling and carbon flow calculation method of electricity-gas-heat, and will not be repeated here.

[0078] Example 4 Embodiment 4 of the present invention provides an electronic device.

[0079] An electronic device includes a memory, a processor, and a program stored in the memory and running on the processor. When the processor executes the program, it implements the steps in the integrated energy optimization scheduling and carbon flow calculation method for electricity, gas, and heat as described in Embodiment 1 of the present invention.

[0080] The detailed steps are the same as those provided in Example 1 for the integrated energy optimization scheduling and carbon flow calculation method of electricity-gas-heat, and will not be repeated here.

[0081] Example 5 Embodiment 5 of the present invention provides a computer program product.

[0082] A computer program product includes software code, wherein the program in the software code performs the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity-gas-heat as described in Embodiment 1 of the present invention.

[0083] The detailed steps are the same as those provided in Example 1 for the integrated energy optimization scheduling and carbon flow calculation method of electricity-gas-heat, and will not be repeated here.

[0084] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0085] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0086] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0087] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0088] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0089] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0090] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. An electric-gas-thermal integrated energy optimization scheduling and carbon flow calculation method, characterized in that, include: Physical flow model and carbon emission flow model for obtaining integrated electricity-gas-heat energy; Based on the obtained physical power flow model, a daily-level physical optimization scheduling model is constructed, and daily-level optimal scheduling samples are generated. Based on the generated daily optimal scheduling samples, combined with a daily graph convolutional network, the coupled scheduling of integrated energy is predicted. Based on the obtained coupled scheduling results, the operational constraints of integrated energy are obtained. Combined with the obtained carbon emission flow model, the carbon flow of integrated energy is calculated, and the optimal scheduling and carbon flow calculation of integrated energy (electricity, gas, and heat) are completed.

2. The method for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat as described in claim 1, characterized in that, The integrated electricity-gas-heat energy system includes an electric power system, a heating system, and a natural gas system. The electric power system includes wind turbines, photovoltaic generators, a power purchase interface with the upstream power grid, combined heat and power (CHP) units, electrical load nodes, and electrical energy storage equipment. The heating system includes heat sources, heat load nodes, heating network pipelines, electric boilers, and thermal energy storage equipment. The natural gas system includes gas sources, gas load nodes, gas network pipelines, electricity-to-gas conversion equipment, and gas storage equipment. The CHP unit connects the natural gas system, the electric power system, and the heating system. The electricity-to-gas conversion equipment connects the electric power system and the natural gas system. The electric boiler connects the electric power system and the heating system. Both the CHP unit and the electricity-to-gas conversion equipment are associated with carbon capture equipment.

3. The method for integrated energy optimization scheduling and carbon flow calculation based on electricity, gas, and heat as described in claim 1, characterized in that, The physical flow model of the integrated electricity-gas-heat energy system includes at least the power balance constraints of the power system, the power balance constraints of the thermal system, the balance constraints of the natural gas system, the operating constraints of the energy storage equipment, the coupling relationship of the cogeneration equipment, the coupling relationship of the power-to-gas equipment, the coupling relationship of the electric boiler equipment, and the energy consumption relationship of the carbon capture equipment.

4. The method for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat as described in claim 1, characterized in that, The constructed daily-level physical optimization scheduling model takes economic optimization as the objective function, that is, ;in, For time-of-use electricity pricing, For gas prices, For the power purchase capacity, To purchase gas volume, For equipment operation and maintenance costs, The penalty coefficient for wind and solar power curtailment. and These are the power curtailed from wind and the power curtailed from solar power, respectively.

5. The method for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat as described in claim 1, characterized in that, The daily-level graph convolutional network includes a topological graph of the integrated energy conversion branch and its corresponding adjacency matrix, as well as a graph convolutional network model; wherein, the graph convolutional network model adopts a graph convolutional structure without attention weights.

6. The method for integrated energy scheduling and carbon flow calculation based on electricity, gas, and heat as described in claim 1, characterized in that, The operational constraints of the integrated energy system are physical constraints, and the loss function of the physical constraints is: ;in, For data fitting loss; Loss due to physical constraints; This is the regularization loss; represents the physical loss weights in the k-th training iteration.

7. A comprehensive energy optimization scheduling and carbon flow calculation system integrating electricity, gas, and heat, characterized in that, include: The acquisition module is configured to acquire the physical flow model of integrated electric-gas-heat energy and the carbon emission flow model; The generation module is configured to construct a daily-level physical optimization scheduling model based on the acquired physical power flow model and generate daily-level optimal scheduling samples. The prediction module is configured to predict the coupled scheduling of integrated energy resources based on the generated daily optimal scheduling samples and combined with a daily graph convolutional network. The optimization module is configured to obtain the operational constraints of integrated energy based on the obtained coupled scheduling results, calculate the carbon flow of integrated energy in combination with the obtained carbon emission flow model, and complete the optimized scheduling and carbon flow calculation of integrated energy of electricity, gas and heat.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity, gas and heat as described in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the program, it implements the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity-gas-heat as described in any one of claims 1-6.

10. A computer program product, comprising software code, characterized in that, The program in the software code executes the steps of the integrated energy optimization scheduling and carbon flow calculation method for electricity, gas and heat as described in any one of claims 1-6.

Images