Comprehensive energy system polymorphic simulation method, system, equipment and medium

Abstract

This invention belongs to the field of integrated energy system simulation technology, specifically involving a multi-state simulation method, system, equipment, and medium for integrated energy systems. Addressing the shortcomings of existing integrated energy system simulation methods in terms of model accuracy and computational efficiency, this invention adopts the following technical solution: a multi-state simulation method for integrated energy systems, comprising: S1, constructing a multi-state model of the integrated energy system based on the physical characteristics and flow patterns of various forms of energy within the integrated energy system; S2, constructing a multi-state model library for the integrated energy system based on the multi-state model; and S3, performing simulation calculations based on the multi-state model library and combined with multi-scenario topology analysis. The beneficial effects of this invention's multi-state simulation method for integrated energy systems are: based on a mechanism- and data-driven modeling method, it improves the accuracy of simulation by dynamically modeling the integrated energy system.

Description

Technical Field

[0001] This invention belongs to the field of integrated energy system simulation technology, specifically involving multi-state simulation methods, systems, equipment, and media for integrated energy systems. Background Technology

[0002] As energy systems become increasingly complex, integrated energy systems are becoming crucial for promoting efficient energy use and sustainable development. Integrated energy systems can integrate multiple energy forms and energy conversion methods to achieve efficient utilization of energy resources and meet energy demands in different application scenarios. Therefore, the intelligent, efficient, and reliable management and operation of integrated energy systems are of paramount importance.

[0003] Simulation modeling of integrated energy systems is a highly effective method. However, current simulation techniques have many limitations in handling multi-energy flow coupling, interdisciplinary topologies, and dynamic multi-application scenarios, especially in balancing model accuracy and computational efficiency. Accurate modeling and efficient simulation of integrated energy systems across multiple application scenarios remain challenging.

[0004] Furthermore, existing integrated energy system simulation technologies generally focus on data-driven or physics-based simulation methods, rarely considering a true integration of "mechanism-data dual-drive." The novel "mechanism-data dual-drive" approach combines the real-time feedback capabilities of data-driven simulations with the accuracy of physics-based simulations to improve the accuracy and efficiency of simulations. Meanwhile, traditional energy system simulations are often limited to a single domain, failing to effectively reveal the complex interactions arising from multi-energy interconnections; they also lack comprehensive and efficient solutions for data integration across different domains, the collaboration and interaction of various equipment models, and interface communication between different programming languages.

[0005] Therefore, simulation technology based on integrated energy systems must overcome several challenges, including but not limited to interdisciplinary data integration, equipment model polymorphism, and multilingual interaction. For example, there are differences in data formats from different fields, data quality standardization, difficulty in data acquisition, and difficulty in managing large-scale data; there are numerous types of equipment involved, model changes caused by uncertain operating conditions, and complex coupling relationships between equipment; communication between equipment models using different programming languages ​​requires overcoming language conversion errors, data transmission efficiency issues, and interface stability. Summary of the Invention

[0006] This invention addresses the shortcomings of existing integrated energy system simulation methods in terms of model accuracy and computational efficiency. It proposes a multi-application-oriented multi-state simulation method for integrated energy systems, employing a mechanism-data dual-driven approach to improve simulation accuracy and efficiency, and supporting the collaborative operation of various energy devices in different application scenarios. This invention also proposes an integrated energy system multi-state simulation system, equipment, and medium based on this multi-state simulation method.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a multi-state simulation method for integrated energy systems, wherein the multi-state simulation method for integrated energy systems includes:

[0008] S1. Based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system, construct a multi-state model of the integrated energy system;

[0009] S2. Based on the multi-state model of the integrated energy system, construct a multi-state model library for the integrated energy system;

[0010] S3. Based on the polymorphic model library, combined with multi-scenario topology analysis, simulation calculation is performed.

[0011] Step S1 includes:

[0012] Step S11: For key equipment and energy supply pipelines in the integrated energy system, based on their energy transport, conversion, and storage mechanisms, a generalized impedance model is constructed. The generalized impedance model is expressed mathematically as follows:

[0013] Z i =f(P i Q i ,T i ,...) (1)

[0014] In the formula, Z i Let P be the impedance of the i-th device or pipe. i Q i 、T i These are key parameters related to active power, reactive power, and temperature.

[0015] Step S12: Analogous to the resistor-capacitor circuit in electricity, construct a dynamic transmission model for key equipment and power supply pipelines. The dynamic transmission model is expressed by the following mathematical formula:

[0016] F j =g(Z1,Z2,...,Z) n ,P j Q j (2)

[0017] In the formula, F j Let Z1 to Z be the flow rate of the j-th energy source. nFor the impedance of relevant key equipment or pipelines, P j Q j These are the relevant key parameters.

[0018] Step S13: Based on the dynamic transmission model, establish steady-state and dynamic mechanistic models for each energy flow. Combining historical and real-time data, construct a multi-state model of the integrated energy system driven by both mechanism and data. Step S13 includes:

[0019] Step S131: Establish steady-state and dynamic mechanism models for each energy flow, including:

[0020] Step S1311: For power flow, establish a steady-state model based on power grid flow analysis and a multi-state model of the integrated energy system based on transient stability analysis. The power flow equations in the steady-state model are expressed as follows:

[0021]

[0022] In the formula, P ij V represents the electrical power flow from node i to node j; i and V j θ represents the voltage at nodes i and j; i and θ j θ represents the voltage phase angle between nodes i and j. ij =θ i -θ j ;X ij This represents the reactance between nodes i and j;

[0023] The integrated energy system polymorphism model of electrical energy flow is represented as follows:

[0024]

[0025] V j =I j ·R j (5)

[0026] In the formula, I i V is the current in the i-th branch; j R is the voltage across the j-th element; j It is the resistance of the j-th element;

[0027] Step S1312: For thermal energy flow, establish a steady-state thermal network model based on the first and second laws of thermodynamics, and a multi-state model of the integrated energy system considering thermal inertia. The thermal network equations in the steady-state model are expressed as:

[0028]

[0029] Q ijT represents the thermal energy flow (heat) from node i to node j; i and T j R represents the temperature of nodes i and j; ij This represents the thermal resistance between nodes i and j;

[0030] The polymorphic model of the integrated energy system based on thermal energy flow is as follows:

[0031]

[0032] Q in =η heat ·P heat (8)

[0033]

[0034] In the formula, Q is the rate of change of thermal energy over time. in Q is the heat provided by the heat source; out It is the heat flowing to the heat load;

[0035] Q loss It is the heat lost by the system; η heat It is the conversion efficiency of the heat source; P heat It is the power of the heat source; T source and T sink These are the temperatures of the heat source and the heat load, respectively; R is the thermal resistance, which describes the resistance to heat flow.

[0036] Step S1313: For gas energy flow, establish steady-state and integrated energy system multistate models based on fluid mechanics and pipeline transmission. The gas energy flow equation in the steady-state model is expressed as:

[0037]

[0038] G ij Represents the gas energy flow rate from node i to node j; K represents the pipe constant; p i and p j The gas pressure at nodes i and j is represented by ρ; the gas density by Z; the gas compressibility factor by T; and the gas temperature by T.

[0039] The comprehensive energy system polymorphism model for gas energy flow is as follows:

[0040]

[0041] m in =ρ·Q gas,in (12)

[0042] m out =ρ·Q gas,out (13)

[0043] In the formula, It is the rate of change of natural gas quality over time; m in and m out These are the masses of natural gas entering and leaving the pipeline, respectively; m loss ρ is the mass of natural gas lost by the system; Q is the density of natural gas; gas,in and Q gas,out These are the gas volume flow rates entering and leaving the pipeline, respectively;

[0044] The electrothermal conversion equation is expressed as:

[0045] Q = η eh ·P (14)

[0046] In the formula, Q represents thermal power; η eh represents the electrothermal conversion efficiency; P represents the electrical power.

[0047] Step S13 also includes:

[0048] The training process of the mechanistic model includes:

[0049] Suppose the integrated energy system has historical data from n nodes. Let the historical data from these n nodes be used as input, denoted as x = [x1, x2, ..., xn]. n Substitute the parameters into the mechanism model to calculate the key node parameters and obtain the calculated value y. c The historical measurement data of key nodes are denoted as y. r The deviation function f(θ) between the two is constructed as follows:

[0050]

[0051] In the formula, θ represents the model parameters to be verified and optimized in the integrated energy mechanism model, θ=[θ1,θ2,...,θ m ]; n represents node data in the integrated energy system; and These are the historical measurement value and the calculated value of the i-th node, respectively;

[0052] During training, the parameter θ is optimized by minimizing the bias function.

[0053] Step S13 also includes:

[0054] Step S134: Based on the mechanism-data fusion model, construct a multi-energy flow equivalent model, including: coupling models of different energy forms to describe their conversion and interaction, introducing energy conversion coefficient and energy loss coefficient parameters to describe the conversion efficiency and loss between different energy forms, and optimizing the multi-energy flow equivalent model through optimization algorithms;

[0055] Step S135: Based on the multi-energy flow equivalent model, introduce time variables and dynamic control equations to construct a multi-state model of the integrated energy system. The multi-state model of the integrated energy system describes the dynamic behavior and response characteristics of the system at different time scales.

[0056] Step S136: Validate and evaluate the multi-state model of the integrated energy system using real-time data to verify the accuracy and reliability of the multi-state model of the integrated energy system.

[0057] Step S2 includes:

[0058] Step S21: Standardize the polymorphic model of the integrated energy system, as follows:

[0059] Standardize(Model i ) = Model std (17)

[0060] In the formula, Model i Represents the original integrated energy system polymorphic model; Model std It is a standardized model;

[0061] Step S22: Develop a multi-language interactive platform interface and use an application programming interface to realize the data interaction of the integrated energy system multi-state model between different programming languages;

[0062] Step S23: Based on the above steps, further construct a polymorphic model library:

[0063] Step S231: Classify and abstract the standardized polymorphic models of different integrated energy systems;

[0064] Step S232: Determine the input / output interfaces and parameters of the integrated energy system multi-state model, and further define the required input / output interfaces and parameter formats for each integrated energy system multi-state model;

[0065] Step S233: Use an open-source language to write model library code to ensure that the integrated energy system polymorphic model can be called and integrated by multiple programming languages.

[0066] Step S3 includes:

[0067] Step S31: Construct the topology based on the energy flow graph. The energy flow graph structure is represented as follows:

[0068] A·x=b (18)

[0069] In the formula, A is the coefficient matrix based on the energy flow diagram; x is the vector of unknown variables; b is the vector of known conditions; A, x, and b are determined through topological analysis, reflecting the structure and energy flow distribution of the integrated energy system.

[0070] Step S32: Use a parallel computing solver (Solver) parallel The decomposed simulation subtasks are processed to generate the final simulation result set, Results, as follows:

[0071] Solver parallel (Subtasks)→Results (19)

[0072] Step S33: Based on steps S31 and S32, perform co-simulation solution;

[0073] Step S331: Divide the simulation boundary according to the specific application scenario and determine an appropriate simulation step size to ensure the accuracy and efficiency of the simulation;

[0074] Step S332: Construct a collaborative simulation solution method library for multiple application scenarios to provide customized simulation solutions for different scenarios;

[0075] Step S333: Perform collaborative simulation solution using a distributed average consensus algorithm, as shown below:

[0076]

[0077] In the formula, This represents the value of node i at the k-th iteration. w is the set of neighbors of node i, representing all nodes connected to node i; ij It is the weight between node i and node j;

[0078] Furthermore, by continuously updating its own value and exchanging information with neighboring nodes, global information consistency is ultimately achieved. The information exchange between nodes is handled by a message passing protocol, represented as:

[0079]

[0080] In the formula, the function f(·) represents the information aggregation method; the function g(·) represents the information processing function.

[0081] A multi-state simulation system for integrated energy systems, employing the aforementioned multi-state simulation method for integrated energy systems, comprises:

[0082] The integrated energy system multi-state model construction module is used to construct a multi-state model of the integrated energy system based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system.

[0083] The multi-state model library construction module is used to build a multi-state model library for integrated energy systems.

[0084] The analytical solution module is used to perform simulation solutions based on the multi-state model library and combined with multi-scenario topology analysis.

[0085] A computer device includes a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, it implements the aforementioned integrated energy system multi-state simulation method.

[0086] A computer-readable storage medium having a computer program stored thereon, which, when executed, implements the aforementioned multi-state simulation method for integrated energy systems.

[0087] The beneficial effects of the multi-state simulation method for integrated energy systems of this invention are as follows: Based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system, dynamic modeling of the integrated energy system improves the accuracy of the simulation; it also addresses the challenge of language interaction. Furthermore, while ensuring accuracy, it improves the efficiency of simulation calculations to adapt to the complex computational needs under multi-energy flow coupling and multiple application scenarios; and it further enhances the accuracy of the simulation by leveraging key parameter identification technology. Attached Figure Description

[0088] Figure 1 This is a flowchart of the integrated energy system multi-state simulation method according to an embodiment of the present invention. Detailed Implementation

[0089] The technical solutions of the embodiments of the present invention will be explained and described below. However, the following embodiments are only preferred embodiments of the present invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of the present invention.

[0090] See Figure 1 The integrated energy system multi-state simulation method of the present invention includes:

[0091] S1. Based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system, construct a multi-state model of the integrated energy system;

[0092] S2. Based on the multi-state model of the integrated energy system, construct a multi-state model library for the integrated energy system and develop a simulation platform with multi-language interaction capabilities;

[0093] S3. Based on the polymorphic model library, combined with multi-scenario topology analysis, simulation calculation is performed.

[0094] Step S1 includes:

[0095] Step S11: For key equipment and energy supply pipelines in the integrated energy system, based on their energy transport, conversion, and storage mechanisms, a generalized impedance model is constructed. The generalized impedance model is expressed mathematically as follows:

[0096] Z i =f(P i Q i ,T i ,...) (1)

[0097] In the formula, Z i Let P be the impedance of the i-th device or pipe. i Q i 、T i These are key parameters related to active power, reactive power, and temperature.

[0098] Step S12: Analogous to the RC circuit in electricity, construct a dynamic transmission model for key equipment and power supply pipelines, where the multi-energy flow equivalent model is expressed mathematically as follows:

[0099] F j =g(Z1,Z2,...,Z) n ,P j Q j (2)

[0100] In the formula, F j Let Z1 to Z be the flow rate of the j-th energy source. n For the impedance of relevant key equipment or pipelines, P j Q j These are the relevant key parameters.

[0101] Step S13: Based on the dynamic transport model, establish steady-state and dynamic mechanistic models for each energy flow. Combining historical and real-time data, construct a multi-state model of the integrated energy system driven by both mechanism and data. The dynamic transport model is a general expression of the mechanistic model, which is further elaborated according to the specific energy flow.

[0102] Step S13 includes:

[0103] Step S131: Establish steady-state and dynamic mechanism models for each energy flow, including:

[0104] Step S1311: For power flow, establish a steady-state model based on power grid flow analysis and a multi-state model of the integrated energy system based on transient stability analysis. The power flow equations in the steady-state model are expressed as follows:

[0105]

[0106] In the formula, P ij V represents the electrical power flow from node i to node j; i and V j θ represents the voltage at nodes i and j; i and θ j θ represents the voltage phase angle between nodes i and j. ij =θ i -θ j ;X ij This represents the reactance between nodes i and j;

[0107] The integrated energy system polymorphism model of electrical energy flow is represented as follows:

[0108]

[0109] V j =I j ·R j (5)

[0110] In the formula, I i V is the current in the i-th branch; j R is the voltage across the j-th element; j It is the resistance of the j-th element;

[0111] Step S1312: For thermal energy flow, establish a steady-state thermal network model based on the first and second laws of thermodynamics, and a multi-state model of the integrated energy system considering thermal inertia. The thermal network equations in the steady-state model are expressed as:

[0112]

[0113] Q ij T represents the thermal energy flow (heat) from node i to node j; i and T j R represents the temperature of nodes i and j; ij This represents the thermal resistance between nodes i and j;

[0114] The polymorphic model of the integrated energy system based on thermal energy flow is as follows:

[0115]

[0116] Q in =η heat ·P heat (8)

[0117]

[0118] In the formula, Q is the rate of change of thermal energy over time. in Q is the heat provided by the heat source; out It is the heat flowing to the heat load;

[0119] Q loss It is the heat lost by the system; η heat It is the conversion efficiency of the heat source; P heat It is the power of the heat source; T source and T sink These are the temperatures of the heat source and the heat load, respectively; R is the thermal resistance, which describes the resistance to heat flow.

[0120] Step S1313: For gas energy flow, establish steady-state and integrated energy system multistate models based on fluid mechanics and pipeline transmission. The gas energy flow equation in the steady-state model is expressed as:

[0121]

[0122] G ij Represents the gas energy flow rate from node i to node j; K represents the pipe constant; p i and p j The gas pressure at nodes i and j is represented by ρ; the gas density by Z; the gas compressibility factor by T; and the gas temperature by T.

[0123] The comprehensive energy system polymorphism model for gas energy flow is as follows:

[0124]

[0125] m in =ρ·Q gas,in (12)

[0126] m out =ρ·Q gas,out (13)

[0127] In the formula, It is the rate of change of natural gas quality over time; m in and m out These are the masses of natural gas entering and leaving the pipeline, respectively; m loss ρ is the mass of natural gas lost by the system; Q is the density of natural gas; gas,in and Q gas,out These are the gas volume flow rates entering and leaving the pipeline, respectively;

[0128] The electrothermal conversion equation is expressed as:

[0129] Q = η eh ·P (14)

[0130] In the formula, Q represents thermal power; ηeh represents the electrothermal conversion efficiency; P represents the electrical power.

[0131] Step S13 also includes:

[0132] Step S132: Collect historical and real-time data, including parameters such as flow rate, pressure, temperature, and concentration of various energy forms. Preprocess the data, including data cleaning, outlier handling, and data imputation. Extract and select features from the data to obtain the features required for modeling.

[0133] Step S133: Combine the mechanism model with the data-driven model, train the mechanism model, and use historical data to perform parameter verification and optimization of the mechanism model to obtain the mechanism-data fusion model. In the fusion model, the mechanism model provides constraints of physical laws, and the data-driven model provides an approximation of the actual system behavior.

[0134] The training process of the mechanistic model includes:

[0135] Suppose the integrated energy system has historical data from n nodes. Let the historical data from these n nodes be used as input, denoted as x = [x1, x2, ..., xn]. n Substitute the parameters into the mechanism model to calculate the key node parameters and obtain the calculated value y. c The historical measurement data of key nodes are denoted as y. r The deviation function f(θ) between the two is constructed as follows:

[0136]

[0137] In the formula, θ represents the model parameters to be verified and optimized in the integrated energy mechanism model, θ=[θ1,θ2,...,θ m ]; n represents node data in the integrated energy system; and These are the historical measurement value and the calculated value of the i-th node, respectively;

[0138] During training, the parameter θ is optimized by minimizing the bias function, such as by minimizing the error between the calculated value y from the mechanistic model and the true value (e.g., mean squared error, cross entropy). This can be achieved using optimization algorithms such as backpropagation and gradient descent.

[0139] Step S134: Based on the mechanism-data fusion model, construct a multi-energy flow equivalent model, including: coupling models of different energy forms to describe their conversion and interaction, introducing energy conversion coefficient and energy loss coefficient parameters to describe the conversion efficiency and loss between different energy forms, and optimizing the multi-energy flow equivalent model through optimization algorithms;

[0140] Step S135: Based on the multi-energy flow equivalent model, introduce time variables and dynamic control equations to construct a multi-state model of the integrated energy system. The multi-state model of the integrated energy system describes the dynamic behavior and response characteristics of the system at different time scales.

[0141] Step S136: Validate and evaluate the multi-state model of the integrated energy system using real-time data to verify the accuracy and reliability of the multi-state model of the integrated energy system.

[0142] Step S2 includes:

[0143] Step S21: Standardize the polymorphic model of the integrated energy system, as follows:

[0144] Standardize(Model i ) = Model std (17)

[0145] In the formula, Model i Represents the original integrated energy system polymorphic model; Model std It is a standardized model;

[0146] Standardization includes data standardization, feature standardization, and model parameter standardization.

[0147] Data standardization includes:

[0148] Data preprocessing includes: missing value handling: filling missing values ​​using the mean, median, mode, interpolation, or machine learning algorithms (such as K-nearest neighbors, decision trees, etc.); outlier handling: identifying and handling outliers based on business logic or statistical methods (such as IQR rules, Z-score, etc.).

[0149] Data normalization: Transforming data to the same scale. Common methods include max-min normalization (scaling data to the [0,1] interval) and Z-score normalization (converting data to a standard normal distribution).

[0150] Feature standardization includes:

[0151] Feature selection: Select features that have a significant impact on model performance based on methods such as correlation analysis, information gain, and mutual information.

[0152] Feature transformation: Transform the original features, such as logarithmic transformation, polynomial transformation, Box-Cox transformation, etc., to meet the model requirements.

[0153] Feature scaling: such as scaling features to unit variance (e.g., Z-score standardization), or scaling to a specific range (e.g., [0,1]).

[0154] Model parameter standardization includes:

[0155] Parameter initialization: Use a unified parameter initialization method, such as random initialization, He initialization, Xavier initialization, etc.

[0156] Hyperparameter settings: For hyperparameters such as learning rate, number of iterations, and batch size, ensure that the same settings are used during training and validation.

[0157] Optimization algorithm: Select a suitable optimization algorithm (such as gradient descent, Adam, RMSprop, etc.) and ensure the consistency of algorithm parameters (such as learning rate decay, momentum, etc.).

[0158] Step S22: Develop a multi-language interactive platform interface and use an application programming interface to realize the data interaction of the integrated energy system multi-state model between different programming languages;

[0159] Step S23: Based on the above steps, further construct a polymorphic model library:

[0160] Step S231: Classify and abstract the standardized polymorphic models of different integrated energy systems;

[0161] Step S232: Determine the input / output interfaces and parameters of the integrated energy system multi-state model, and further define the required input / output interfaces and parameter formats for each integrated energy system multi-state model;

[0162] Step S233: Use an open-source language to write model library code to ensure that the integrated energy system polymorphic model can be called and integrated by multiple programming languages.

[0163] Step S3 includes:

[0164] Step S31: Construct the topology based on the energy flow graph. The energy flow graph structure is represented as follows:

[0165] A·x=b (18)

[0166] In the formula, A is the coefficient matrix based on the energy flow diagram; x is the vector of unknown variables; b is the vector of known conditions; A, x, and b are determined through topological analysis, reflecting the structure and energy flow distribution of the integrated energy system.

[0167] Step S32: Use a parallel computing solver (Solver) parallel The decomposed simulation subtasks are processed to generate the final simulation result set, Results, as follows:

[0168] Solver parallel (Subtasks)→Results (19)

[0169] Step S33: Based on steps S31 and S32, perform co-simulation solution;

[0170] Step S331: Divide the simulation boundary according to the specific application scenario and determine an appropriate simulation step size to ensure the accuracy and efficiency of the simulation;

[0171] Step S332: Construct a collaborative simulation solution method library for multiple application scenarios to provide customized simulation solutions for different scenarios;

[0172] Step S333: Perform collaborative simulation solution using a distributed average consensus algorithm, as shown below:

[0173]

[0174] In the formula, This represents the value of node i at the k-th iteration. w is the set of neighbors of node i, representing all nodes connected to node i; ij It is the weight between node i and node j;

[0175] Furthermore, by continuously updating its own value and exchanging information with neighboring nodes, global information consistency is ultimately achieved. The information exchange between nodes is handled by a message passing protocol, represented as:

[0176]

[0177] In the formula, the function f(·) represents the information aggregation method; the function g(·) represents the information processing function.

[0178] The multi-state simulation method for integrated energy systems in this embodiment mainly includes the following steps in constructing the multi-state model of the integrated energy system: First, based on the physical characteristics and flow laws of various forms of energy in the integrated energy system, establish steady-state and dynamic mechanistic models of each energy flow. Then, combine the mechanistic model with the data-driven model, use historical data to perform parameter verification and optimization of the mechanistic model, train the mechanistic model, and obtain the mechanistic-data fusion model. Furthermore, based on the mechanistic-data fusion model, construct the multi-energy flow equivalent model. Finally, based on the multi-energy flow equivalent model, introduce time variables and dynamic control equations to construct the multi-state model of the integrated energy system. The multi-state model of the integrated energy system describes the dynamic behavior and response characteristics of the system at different time scales.

[0179] Theoretical Application Examples

[0180] Consider a regional multi-energy system that incorporates various energy forms, including electricity, heat, and natural gas. The system comprises gas turbines, generators, photovoltaic (PV) power generation, wind power generation (WT), energy storage systems (ES), and control systems. Energy efficiency is achieved through an energy hub. The simulation process for this multi-energy system includes:

[0181] Step S1: Model the energy hub. Although the multi-energy system includes various energy forms such as electricity, heat, and natural gas, the energy forms are abstracted into two basic needs: electricity and heat.

[0182] Step S131: Construct steady-state and dynamic mechanism models, which include:

[0183] Step S1311: For power flow, establish a steady-state model based on power grid flow analysis and a multi-state model of the integrated energy system based on transient stability analysis. The power flow equations in the steady-state model are expressed as follows:

[0184]

[0185] In the formula, P ij V represents the electrical power flow from node i to node j; i and V j θ represents the voltage at nodes i and j; i and θ j θ represents the voltage phase angle between nodes i and j. ij =θ i -θ j ;X ij This represents the reactance between nodes i and j;

[0186] The integrated energy system polymorphism model of electrical energy flow is represented as follows:

[0187]

[0188] V j =I j ·R j (5)

[0189] In the formula, I i V is the current in the i-th branch; j R is the voltage across the j-th element; j It is the resistance of the j-th element;

[0190] Step S1312: For thermal energy flow, establish a steady-state thermal network model based on the first and second laws of thermodynamics, and a multi-state model of the integrated energy system considering thermal inertia. The thermal network equations in the steady-state model are expressed as:

[0191]

[0192] Q ij T represents the thermal energy flow (heat) from node i to node j; i and T j R represents the temperature of nodes i and j; ij This represents the thermal resistance between nodes i and j;

[0193] The polymorphic model of the integrated energy system based on thermal energy flow is as follows:

[0194]

[0195] Q in =η heat ·P heat (8)

[0196]

[0197] In the formula, Q is the rate of change of thermal energy over time. in Q is the heat provided by the heat source; out It is the heat flowing to the heat load;

[0198] Q loss It is the heat lost by the system; η heat It is the conversion efficiency of the heat source; P heat It is the power of the heat source; T source and T sink These are the temperatures of the heat source and the heat load, respectively; R is the thermal resistance, which describes the resistance to heat flow.

[0199] Step S1313: For gas energy flow, establish steady-state and integrated energy system multistate models based on fluid mechanics and pipeline transmission. The gas energy flow equation in the steady-state model is expressed as:

[0200]

[0201] G ij Represents the gas energy flow rate from node i to node j; K represents the pipe constant; p i and p j The gas pressure at nodes i and j is represented by ρ; the gas density by Z; the gas compressibility factor by T; and the gas temperature by T.

[0202] The comprehensive energy system polymorphism model for gas energy flow is as follows:

[0203]

[0204] m in =ρ·Q gas,in (12)

[0205] m out =ρ·Q gas,out (13)

[0206] In the formula, It is the rate of change of natural gas quality over time; m in and m out These are the masses of natural gas entering and leaving the pipeline, respectively; m loss ρ is the mass of natural gas lost by the system; Q is the density of natural gas; gas,in and Q gas,out These are the gas volume flow rates entering and leaving the pipeline, respectively;

[0207] The electrothermal conversion equation is expressed as:

[0208] Q = η eh ·P (14)

[0209] In the formula, Q represents thermal power; η eh represents the electrothermal conversion efficiency; P represents the electrical power.

[0210] More specifically, the dynamic equation for the speed of the gas turbine is expressed as:

[0211]

[0212] In the formula, N turbine (t) represents the gas turbine speed (rpm) at time t; T mech It is the mechanical time constant; f fuel It is the conversion function from fuel flow rate to mechanical power; m fuel (t) represents the natural gas fuel flow rate (kg / s) at time t; f loss It is the mechanical power loss function; T exhaust (t) represents the gas turbine exhaust temperature (K) at time t.

[0213] The power output equation of a gas turbine is expressed as:

[0214] P turbine,elec (t)=η turbine ·f mech-elec (N turbine (t)) (23)

[0215] In the formula, P turbine,elec (t) represents the power output (kW) of the gas turbine at time t; η turbine The efficiency of a gas turbine; f mech-elec It is a conversion function from mechanical power to electrical output.

[0216] The state equation of an energy storage system (using battery energy storage) is expressed as:

[0217]

[0218] In the formula, E ES (t) is the energy level (kWh) of the energy storage system at time t; η ES,ch and η ES,dis These represent the charging and discharging efficiencies of the energy storage system, respectively; P ES,ch (t) represents the charging power (kW) of the energy storage system at time t; P ES,dis (t) represents the discharge power (kW) of the energy storage system at time t.

[0219] The power output equation for a CHP (combined heat and power) system is expressed as follows:

[0220] P CHP,elec (t)=η CHP ·F gas,CHP (t) (25)

[0221] In the formula, P CHP,elec (t) represents the power output of the CHP system at time t; η CHP Indicates the power conversion efficiency of the combined heating and power supply system; F gas,CHP (t) is the natural gas flow rate (m³) supplied to the CHP system. 3 / h).

[0222] The thermal output equation of the CHP system is expressed as:

[0223] Q CHP,heat (t)=η heat ·F gas,CHP (t) (26)

[0224] In the formula, Q CHP,heat (t) represents the thermal output of the CHP system at time t; η heat This indicates the thermal conversion efficiency of CHP.

[0225] The boiler's heat output equation is expressed as:

[0226]

[0227] In the formula, Q boiler (t) represents the boiler thermal output at time t; η boiler Indicates the boiler's heat conversion efficiency; F gas,boiler (t) represents the flow rate (m³) of natural gas supplied to the boiler. 3 / h).

[0228] The energy demand balance equation is expressed as:

[0229]

[0230] Q heat (t)=Q CHP,heat (t)+Q boiler (t) (29)

[0231] In the formula, P elec (t) represents the region's electricity demand (kW) at time t; Q heat (t) represents the region's heat demand (kW) at time t; P grid (t) represents the electricity obtained from the power grid; P loss (t) represents the losses during power transmission and conversion; P PV (t) represents the photovoltaic power generation (kW) at time t; P WT (t) represents the wind power generation (kW) at time t; P ES,ch (t) and P ES,dis (t) represents the charging and discharging power (kW) of the energy storage system at time t, respectively.

[0232] Step S132: Using historical meteorological data, equipment operation data, load demand data, etc., predict and verify the demand for energy such as electricity and heat, and then optimize and verify the parameters of the dynamic mechanism model of the energy hub.

[0233] 1) Data source collection

[0234] Meteorological data: Collect historical and real-time meteorological data, including temperature, humidity, wind speed, wind direction, sunshine duration, etc., to predict the output of renewable energy sources such as photovoltaic power generation and wind power generation.

[0235] Equipment status data: Collect operating status data, such as temperature, pressure, speed, current, and voltage, from various energy equipment (e.g., gas turbines, generators, photovoltaic panels, wind turbines, etc.) to monitor the health status of the equipment and predict maintenance needs.

[0236] Load data: Collect load data for energy sources such as electricity, heat, and natural gas, including historical load, real-time load, and forecast load, to assess energy demand and optimize energy allocation.

[0237] 2) Data preprocessing

[0238] Data cleaning: Remove duplicate, erroneous, or abnormal data to ensure data accuracy and reliability.

[0239] Data transformation: Converting data into a standardized format and units for data analysis and model building.

[0240] Data imputation: For missing data, interpolation, regression and other methods are used to fill in the gaps.

[0241] Step S133: Combine the mechanism model with the data-driven model, use historical data to perform parameter verification and optimization of the mechanism model, train the mechanism model, and obtain the mechanism-data fusion model. The training process of the mechanism model includes:

[0242] Suppose the integrated energy system has historical data from n nodes. Let the historical data from these n nodes be used as input, denoted as x = [x1, x2, ..., xn]. n Substitute the parameters into the mechanism model to calculate the key node parameters and obtain the calculated value y. c The historical measurement data of key nodes are denoted as y. r The deviation function f(θ) between the two is constructed as follows:

[0243]

[0244] In the formula, θ represents the model parameters to be verified and optimized in the integrated energy mechanism model, θ=[θ1,θ2,...,θ m ]; n represents node data in the integrated energy system; and These are the historical measurement value and the calculated value of the i-th node, respectively;

[0245] During training, the parameter θ is optimized by minimizing the bias function. This can be achieved using optimization algorithms such as backpropagation and gradient descent.

[0246] Ideally, minf(θ) should be satisfied as 0. In this example, the error range is controlled within a given minimum value ε according to engineering requirements, so that minf(θ) ≤ ε, thus transforming the resistance characteristic identification and calibration problem into a nonlinear function optimization problem.

[0247] Furthermore, with the goal of minimizing the deviation function, the Levenberg-Marquarelt (LM) algorithm is used to solve for the mechanistic model parameters θ. The i-th update iteration is represented as:

[0248] θ i+1 =θ i +Δθ i (30)

[0249]

[0250] In the formula, Δθ is the update iteration variable; J is the Jacobian matrix of f(θ); I is the identity matrix; ξ i Ι represents the damping term, ξ i It is a positive parameter to prevent Δθ approaching singularity iToo large.

[0251] The process of identifying the parameter θ of the mechanistic model using a data-driven model is as follows:

[0252] (1) Randomly assign an initial value θ0 and an allowable error ε based on the reference value of the θ engineering requirements;

[0253] (2) Calculate the state parameter y based on the mechanism model c (θ);

[0254] (3) Calculate the objective function f(θ);

[0255] (4) Calculate the Jacobian matrix J and perform iterative updates;

[0256] (5) If f(θ)≤ε, stop and output the result; otherwise, return to step (2).

[0257] Step S134: Based on the mechanism-data fusion model, construct a multi-energy flow equivalent model, including: coupling models of different energy forms to describe their conversion and interaction, introducing energy conversion coefficient and energy loss coefficient parameters to describe the conversion efficiency and loss between different energy forms, and optimizing the multi-energy flow equivalent model through optimization algorithms;

[0258] Step S135: Based on the multi-energy flow equivalent model, introduce time variables and dynamic control equations to construct a multi-state model of the integrated energy system. The multi-state model of the integrated energy system describes the dynamic behavior and response characteristics of the system at different time scales.

[0259] Step S136: Validate and evaluate the multi-state model of the integrated energy system using real-time data to verify the accuracy and reliability of the multi-state model of the integrated energy system.

[0260] Step S2 includes:

[0261] Step S21: Standardize the polymorphic model of the integrated energy system, as follows:

[0262] Standardize(Model i ) = Model std (17)

[0263] In the formula, Model i Represents the original integrated energy system polymorphic model; Model std It is a standardized model;

[0264] Step S22: Develop a multi-language interactive platform interface and use an application programming interface to realize the data interaction of the integrated energy system multi-state model between different programming languages;

[0265] Step S23: Based on the above steps, further construct a polymorphic model library:

[0266] Step S231: Classify and abstract the standardized polymorphic models of different integrated energy systems to more effectively manage and organize a large number of integrated energy system models; classify solar photovoltaic, wind turbines, energy storage devices, etc., according to their functional characteristics, and establish a hierarchical model library based on this; taking the photovoltaic model as an example, the inputs include light intensity Φ and ambient temperature T. ambient Photovoltaic area, etc. A PV Photovoltaic conversion efficiency η PV The output includes voltage, current, etc.

[0267] Step S232: Determine the input / output interfaces and parameters of the integrated energy system multi-state model, and further define the required input / output interfaces and parameter formats for each integrated energy system multi-state model;

[0268] Step S233: Use an open-source language to write model library code to ensure that the integrated energy system polymorphic model can be called and integrated by multiple programming languages.

[0269] Step S3 includes:

[0270] Step S31: Construct the topology based on the energy flow graph. The energy flow graph structure is represented as follows:

[0271] A·x=b (18)

[0272] In the formula, A is the coefficient matrix based on the energy flow diagram; x is the unknown variable vector; b is the known condition vector; A, x, and b are determined through topological analysis, reflecting the structure and energy flow distribution of the integrated energy system; the electricity generated by CHP, GT, PV, WT, and ES flows to the electrical load and ES (during charging); the heat flows from GT and CHP to the heat load.

[0273] Step S32: Use a parallel computing solver (Solver) parallel The decomposed simulation subtasks are processed to generate the final simulation result set, Results, as follows:

[0274] Solver parallel (Subtasks)→Results (19)

[0275] Each subtask can be simulated independently, and the output simulations of CHP, GT, PV, and WT can be processed in parallel as subtasks.

[0276] Step S33: Based on steps S31 and S32, perform co-simulation solution, including:

[0277] Step S331: Define the simulation boundary according to the specific application scenario and determine an appropriate simulation step size to ensure the accuracy and efficiency of the simulation; determine the simulation range based on system components and energy flow diagrams, considering the simulation of the power component or the joint simulation of power and heat. Determine the simulation step size based on the system dynamic characteristics and simulation accuracy requirements, selecting a simulation iteration every 15 minutes;

[0278] Step S332: Construct a collaborative simulation solution method library for multiple application scenarios to provide customized simulation solutions for different scenarios;

[0279] Step S333: Perform collaborative simulation solution using a distributed average consensus algorithm, as shown below:

[0280]

[0281] In the formula, This represents the value of node i at the k-th iteration. w is the set of neighbors of node i, representing all nodes connected to node i; ij It is the weight between node i and node j;

[0282] Furthermore, by continuously updating its own value and exchanging information with neighboring nodes, global information consistency is ultimately achieved. The information exchange between nodes is handled by a message passing protocol, represented as:

[0283]

[0284] In the formula, the function f(·) represents the information aggregation method, and the weighted average function is selected; the function g(·) represents the information processing function.

[0285] Each component, such as CHP, GT, PV, and WT, is considered a node, and the components that have energy flow with it constitute its neighbor set.

[0286] Heuristic algorithms can include particle swarm optimization, ant colony optimization, and simulated return algorithms.

[0287] Based on the dynamic mechanism model of the energy hub, optimized control strategies can be formulated. For example, based on predicted electricity and heat demand, the natural gas supply to the CHP system and boiler, as well as the electricity obtained from the grid, can be adjusted to achieve efficient energy utilization and stable system operation. The detailed steps of the optimized control strategy based on the energy hub are as follows.

[0288] Based on the dynamic mechanism model established in the preceding steps, electricity and heat demand are predicted using historical data D. hist and prediction model M pred Make a prediction:

[0289] P elec Q heat =M pred (D hist ,t) (32)

[0290] Furthermore, an optimized control strategy is formulated, with the optimization objective being to minimize cost. The objective function is expressed as:

[0291]

[0292] In the formula, C nat It's the price of natural gas, C grid It refers to the time-of-use pricing of the power grid, and penalty terms include penalties for violating demand, system constraints, etc.

[0293] Based on the balance constraints in the energy hub modeling process, the objective function is solved by combining optimization algorithms such as predictive control, dynamic programming, and genetic algorithms to obtain the optimal decision variable values.

[0294] Furthermore, based on the output of the optimization algorithm, the natural gas supply to the gas turbine and CHP unit, as well as the electricity obtained from the grid, are adjusted. Iteration and optimization are performed based on real-time data and system operation to achieve efficient energy utilization and stable system operation.

[0295] The beneficial effects of the multi-state simulation method for integrated energy systems in this invention are as follows: Key parameters are identified using genetic algorithms, and power flow analysis models are solved using forward and backward substitution methods, thereby achieving more accurate modeling of the dynamic behavior of the integrated energy system; Through refined calculation methods and the combination of the operating characteristics of each energy supply device and data-driven key parameter identification, the simulation of system dynamic behavior becomes more realistic and credible; A method for constructing a multi-state model library is proposed, standardizing the original equipment models into a unified format to ensure seamless collaboration and effective integration of these models; A multi-language interactive platform interface is developed, enabling cross-programming language exchange of model data through application programming interfaces, achieving high flexibility and interoperability for collaborative work of models from different domains; Finally, based on multi-scenario topology analysis technology, a parallel computing solver is used to process the decomposed simulation subtasks, achieving collaborative simulation solution; Parallel computing offers high computational efficiency and speed.

[0296] This invention also provides a multi-state simulation system for integrated energy systems, employing the aforementioned multi-state simulation method for integrated energy systems. The multi-state simulation system for integrated energy systems includes:

[0297] The integrated energy system multi-state model construction module is used to construct a multi-state model of the integrated energy system based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system.

[0298] The multi-state model library construction module is used to build a multi-state model library for integrated energy systems.

[0299] The simulation platform development module is used to develop a simulation platform with multilingual interactive capabilities for developing multi-state models of integrated energy systems.

[0300] The analytical solution module is used to perform simulation solutions based on the multi-state model library and combined with multi-scenario topology analysis.

[0301] This invention also provides a computer device, including a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, it implements the aforementioned integrated energy system multi-state simulation method.

[0302] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed, implements the aforementioned multi-state simulation method for integrated energy systems.

[0303] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes, but is not limited to, the content described in the above specific embodiments. Any modifications that do not depart from the functional and structural principles of the present invention will be included within the scope of the claims.

Claims

1. A multi-state simulation method for integrated energy systems, characterized by: include: S1. Based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system, construct a multi-state model of the integrated energy system; S2. Based on the multi-state model of the integrated energy system, construct a multi-state model library for the integrated energy system; S3. Based on the multi-state model library, combined with multi-scenario topology analysis, simulation calculation is performed; Step S1 includes: Step S11: For key equipment and energy supply pipelines in the integrated energy system, a generalized impedance model is constructed based on their energy transport, conversion, and storage mechanisms. The generalized impedance model for a single energy flow is expressed mathematically as follows: (1) In the formula, For the first i The impedance of a device or pipeline , , For relevant key parameters; Step S12: Based on the generalized impedance model, construct a dynamic transmission model for key equipment and power supply pipelines. The dynamic transmission model is expressed mathematically as follows: (2) In the formula, Let the flow rate of the j-th energy source be... to For the impedance of relevant key equipment or pipelines, , For relevant key parameters; Step S13: Based on the dynamic transmission model, establish steady-state and dynamic mechanism models of each energy flow, and combine historical data and real-time data to construct a multi-state model of the integrated energy system driven by both mechanism and data. Step S13 includes: Step S134: Based on the mechanism-data fusion model, construct a multi-energy flow equivalent model, including: coupling models of different energy forms to describe their conversion and interaction, introducing energy conversion coefficient and energy loss coefficient parameters to describe the conversion efficiency and loss between different energy forms, and optimizing the multi-energy flow equivalent model through optimization algorithms; Step S135: Based on the multi-energy flow equivalent model, introduce time variables and dynamic control equations to construct a multi-state model of the integrated energy system. The multi-state model of the integrated energy system describes the dynamic behavior and response characteristics of the system at different time scales. Step S136: Validate and evaluate the multi-state model of the integrated energy system using real-time data to verify the accuracy and reliability of the multi-state model of the integrated energy system.

2. The multi-state simulation method for integrated energy systems according to claim 1, characterized in that: Step S13 also includes: Step S131: Establish steady-state and dynamic mechanism models for each energy flow; Step S132: Collect historical and real-time data, preprocess the data, extract and select features from the data to obtain a data-driven model; Step S133: Combine the mechanism model with the data-driven model, use historical data to perform parameter verification and optimization of the mechanism model, train the mechanism model, and obtain the mechanism-data fusion model. The training process of the mechanism model includes: Assuming the integrated energy system has Historical data of each node, The historical data of each node is used as input, denoted as... Substitute the parameters into the mechanism model to calculate the key node parameters and obtain the calculated values. Historical measurement data of key nodes are recorded as follows: Construct the deviation function between the two. as follows: (15) (16) In the formula, This represents the model parameters to be verified and optimized in the integrated energy mechanism model. ; This represents node data in an integrated energy system; and These are the historical measurement value and the calculated value of the i-th node, respectively; During training, the parameters are optimized by minimizing the bias function. .

3. The multi-state simulation method for integrated energy systems according to claim 1 or 2, characterized in that: Step S2 includes: Step S21: Standardize the multi-state model of the integrated energy system; Step S22: Develop a multi-language interactive platform interface and use application programming interfaces to realize dynamic model data interaction between different programming languages; Step S23: Based on steps S21 and S22, further construct the polymorphic model library: Step S231: Classify and abstract the standardized polymorphic models of different integrated energy systems; Step S232: Determine the input / output interfaces and parameters of the integrated energy system multi-state model, and further define the required input / output interfaces and parameter formats for each integrated energy system multi-state model; Step S233: Write model library code using an open-source language.

4. The multi-state simulation method for integrated energy systems according to claim 1 or 2, characterized in that: Step S3 includes: Step S31: Construct the topology based on the energy flow graph. The energy flow graph structure is represented as follows: (18) In the formula, It is a coefficient matrix based on the energy flow diagram; It is a vector of unknown variables; It is a vector with known conditions; , and The topological analysis revealed the structure and energy flow distribution of the integrated energy system. Step S32: Use a parallel computing solver For the decomposed simulation subtasks The data is processed to produce the final set of simulation results. , means as follows: (19) Step S33: Based on steps S31 and S32, perform co-simulation solution, including: Step S331: Divide the simulation boundary according to the specific application scenario and determine an appropriate simulation step size to ensure the accuracy and efficiency of the simulation; Step S332: Construct a collaborative simulation solution method library for multiple application scenarios to provide customized simulation solutions for different scenarios; Step S333: Perform collaborative simulation solution using a distributed average consensus algorithm, as shown below: (20) In the formula, Representative node i The value at the k-th iteration; It is a node i The set of neighbors, representing the set of nodes. i All connected nodes; It is a node i and nodes j Weights between them; Furthermore, by continuously updating its own value and exchanging information with neighboring nodes, global information consistency is ultimately achieved. The information exchange between nodes is handled by a message passing protocol, represented as: (21) In the formula, the function Indicates the method of information aggregation; function This represents an information processing function.

5. A multi-state simulation system for integrated energy systems, employing the multi-state simulation method for integrated energy systems as described in any one of claims 1 to 4, characterized in that: The integrated energy system multi-state simulation system includes: The integrated energy system multi-state model construction module is used to construct a multi-state model of the integrated energy system based on the physical characteristics and flow patterns of various forms of energy in the integrated energy system. The multi-state model library construction module is used to build a multi-state model library for integrated energy systems. The analytical solution module is used to perform simulation solutions based on the multi-state model library and combined with multi-scenario topology analysis.

6. A computer device comprising a processor and a storage medium, wherein the storage medium stores a computer program, characterized in that: When the computer program is executed by the processor, it implements the integrated energy system multi-state simulation method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: It stores a computer program, which, when executed, implements the integrated energy system multi-state simulation method as described in any one of claims 1 to 4.

Images