A comprehensive energy resilience assessment method and system based on multi-source factor coupling
By building a multi-attribute coupled physical model and graph convolution network spatiotemporal graph model, dynamically adjusting edge weights and node characteristics, the problem that the coupling relationship and dynamic change characteristics of multi-energy systems in traditional evaluation methods is difficult to reflect, and more accurate and practical assessment of urban integrated energy systems, especially in disaster situations.
Patent Information
- Application Number
- CN202411205144.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-30
AI Technical Summary
The traditional urban comprehensive energy system evaluation method lacks an effective multi-energy system coupling model, making it difficult to dynamically adjust model parameters, resulting in limited accuracy and practicality of the evaluation results, especially in disaster situations, it is difficult to fully reflect the coupling relationship between multi-energy systems. and its dynamic changes.
Using a comprehensive energy toughness evaluation method based on multi-source factor coupling, a data-driven graph convolution network spatiotemporal graph model is established by constructing a multi-attribute coupled physical model and combining geo-correction functions and symbolic filtering technology, and dynamically adjusting edge weights in disaster situations and node characteristics, calculate the disaster resilience and recovery capabilities of urban energy systems.
It significantly improves the accuracy and practicality of the assessment, provides a scientific basis for the optimization of urban energy systems, can effectively simulate the mutual impact between different energy systems, and provides dynamic adjustments and efficient evaluation in disaster situations.
Smart Images

Figure CN118735135B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of urban integrated energy systems, and in particular to an integrated energy resilience assessment method and system based on the coupling of multiple source factors. Background Art
[0002] With the acceleration of urbanization, the complexity and interconnectivity of urban integrated energy systems are increasing. In this context, the resilience of urban integrated energy systems in the face of natural and man-made disasters is particularly important. Traditional energy system assessment methods mainly rely on single physical models or data-driven models, which often cannot fully reflect the coupling relationship between multiple energy systems and their dynamic change characteristics.
[0003] In existing technologies, power systems, natural gas systems, and thermal systems are modeled and analyzed independently, lacking an effective coupling model to simulate the mutual impact between different energy systems. In addition, traditional assessment methods are difficult to dynamically adjust model parameters in disaster scenarios, which limits the accuracy and practicality of assessment results.
[0004] At present, there is an urgent need for an evaluation method that can comprehensively consider the coupling relationship and dynamic behavior of multi-energy systems. Summary of the invention
[0005] The present application aims to solve one of the technical problems in the related art at least to some extent.
[0006] To this end, the first objective of this application is to propose a comprehensive energy resilience assessment method based on the coupling of multi-source factors.
[0007] The second objective of this application is to propose a comprehensive energy resilience assessment system based on the coupling of multiple source factors.
[0008] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a comprehensive energy resilience assessment method based on the coupling of multiple source factors, including: constructing a multi-attribute coupling physical model of the urban integrated energy system, and using a geographic correction function to correct the coupling effect of the model under different geographic coordinates during the construction; obtaining the spatiotemporal data of the urban integrated energy system, and preprocessing the spatiotemporal data to obtain preprocessed spatiotemporal data; based on the preprocessed spatiotemporal data, using a graph convolutional network to construct a spatiotemporal graph model; embedding the output of the multi-attribute coupling physical model as an additional feature of the node into the spatiotemporal graph model to obtain a spatiotemporal graph model that integrates physical information; obtaining disaster data, and updating the spatiotemporal graph model that integrates physical information based on the disaster data to obtain a dynamic assessment model; determining the dynamic change characteristics of the node status based on the dynamic assessment model, and calculating the resilience index of the urban integrated energy system based on the dynamic change characteristics.
[0009] The comprehensive energy resilience assessment method based on the coupling of multiple source factors in the embodiment of the present application constructs a multi-attribute coupled physical model of electricity, natural gas and thermal systems, combines geographic correction functions and symbolic filtering technology, establishes a data-driven graph convolutional network spatiotemporal graph model, dynamically adjusts edge weights and node characteristics under disaster scenarios, and comprehensively evaluates the disaster resistance and recovery capabilities of urban energy systems through indicators such as recovery time and system failure degree, thereby improving the accuracy and practicality of the assessment and providing a scientific basis for the optimization of urban energy systems.
[0010] Optionally, in one embodiment of the present application, constructing a multi-attribute coupling physical model of an urban integrated energy system includes:
[0011] Use the node admittance matrix method to construct the power system model;
[0012] Construct a natural gas system model based on gas flow equations;
[0013] The thermal system model is constructed using the heat conduction equation;
[0014] Based on the power system model, the natural gas system model and the thermal system model, the power system coupling model, the natural gas system coupling model and the thermal system coupling model are constructed respectively, and the geographic correction function is used to correct the coupling effect of the coupling model under different geographic coordinates during the construction.
[0015] Optionally, in one embodiment of the present application, the power system model is expressed as:
[0016]
[0017] in, Inject power vector into the node, is the node voltage vector, is the system admittance matrix, represents the Hadamard product, denotes conjugation;
[0018] The natural gas system model is expressed as:
[0019]
[0020] in, is the gas flow rate, is the flow coefficient, , are the inlet pressure and outlet pressure respectively;
[0021] The thermal system model is expressed as:
[0022]
[0023] in, is the heat flow, is the heat transfer coefficient, is the heat transfer area, , are the inlet temperature and outlet temperature respectively.
[0024] Optionally, in one embodiment of the present application, the geographic correction function is in a linear form or a nonlinear form. The form of the geographic correction function is set according to the state and historical data of the urban integrated energy system. The linear form of the geographic correction function is expressed as:
[0025]
[0026] The nonlinear form of the geographic correction function is expressed as:
[0027]
[0028] in, For geographical location, is the height, is the geographical correction function of the power system coupling model, is the geographical correction function of the natural gas system coupling model, is the geographic correction function of the thermal system coupling model, , , , , , , , , , , , , , , , , , All are coefficients.
[0029] Optionally, in one embodiment of the present application, the power system coupling model is expressed as:
[0030]
[0031] in, , , is the coupling parameter, is the coupling function of the natural gas system to the power system, is the coupling function of the thermal system to the power system;
[0032] The coupled model of the natural gas system is expressed as:
[0033]
[0034] in, , , is the coupling parameter, is the coupling function of the thermal system to the natural gas system, is the coupling function of the power system to the natural gas system, V is the voltage of the power system, Indicates the power of the power system;
[0035] The coupled model of the thermal system is expressed as:
[0036]
[0037] in, , , is the coupling parameter, is the coupling function of the natural gas system to the thermal system, is the coupling function of the power system to the thermal system.
[0038] Optionally, in one embodiment of the present application, the spatiotemporal data is data recorded by region and time point, the spatiotemporal data includes energy consumption data, meteorological data and disaster data, and the energy consumption data includes electricity consumption data, natural gas consumption data and heat consumption data;
[0039] Preprocess the spatiotemporal data to obtain preprocessed spatiotemporal data, including:
[0040] The symbol filtering technology is used to preprocess the spatiotemporal data to obtain preprocessed spatiotemporal data, wherein the preprocessing process includes symbolization and noise reduction, and the symbolization process includes: dividing the continuous data into a number of intervals, and representing each interval with a symbol;
[0041] Based on the preprocessed spatiotemporal data, a spatiotemporal graph model is constructed using a graph convolutional network, including:
[0042] Constructing a spatiotemporal graph based on preprocessed spatiotemporal data ,in, , Represents a collection of nodes. represents the edge set, the node feature vector is X, the edge weight matrix is A, and the elements of the edge weight matrix A are Representation Node and The mutual influence strength between them, each node represents the state of a specific area at a specific time point, and the node feature vector , For power consumption, For natural gas consumption, For heat consumption, is the temperature, For humidity, is the wind speed, For wind direction, is the precipitation, is a state of disaster;
[0043] Based on the spatiotemporal graph, a spatiotemporal graph model is constructed using a graph convolutional network, where the calculation formula of the graph convolutional network is:
[0044]
[0045] Among them, the initial input of the graph convolutional network , the output of the graph convolutional network is the node representation matrix after multi-layer graph convolution, For the The node representation of the layer, For the The weight matrix of the layer, is the edge weight matrix after symbol filtering.
[0046] Optionally, in one embodiment of the present application, the output of the physical model is used as an additional feature of the node, and the expanded node feature vector is expressed as:
[0047]
[0048] in, is the expanded node feature vector, is the node feature vector, Inject power vector into the node, is the gas flow rate, is the heat flow;
[0049] The calculation formula of graph convolution after introducing physical information is:
[0050]
[0051] in, For the The physical information weight matrix of the layer.
[0052] Optionally, in one embodiment of the present application, the spatiotemporal graph model integrating physical information is updated based on disaster data to obtain a dynamic assessment model, including:
[0053] Update the edge weight matrix and the expanded node feature vector based on the disaster data to obtain a new edge weight matrix and node feature vector;
[0054] Based on the new edge weight matrix and node feature vector, the graph convolutional network after introducing physical information is updated, and the updated spatiotemporal graph model integrating physical information is determined, and the updated spatiotemporal graph model integrating physical information is used as a dynamic evaluation model;
[0055] Determine the dynamic change characteristics of node status based on the dynamic evaluation model, including:
[0056] Through the forward propagation of the dynamic evaluation model, the node representation is calculated at different time steps to evaluate the dynamic change characteristics of the node state;
[0057] Among them, the new edge weight matrix is expressed as:
[0058]
[0059] in, is the disaster impact matrix, The elements represent the degree of disaster impact in a specific area and time point. represents the Hadamard product;
[0060] The new node feature vector is expressed as:
[0061]
[0062] The calculation formula of the updated graph convolutional network is:
[0063]
[0064] Initial input of the updated graph convolutional network , during forward propagation, the node at time step t+1 is represented as , by comparing the node representations at time step t and time step t+1 and Evaluate the dynamic changes of node status.
[0065] Optionally, in one embodiment of the present application, the resilience index of the urban integrated energy system includes recovery time and system failure degree, and the recovery time and system failure degree are negatively correlated with the resilience index;
[0066] The recovery time is calculated as:
[0067]
[0068] in, The assessment period after the disaster. is the number of nodes, is the representation of the i-th node at the t-th time;
[0069] The calculation formula of system failure degree is:
[0070]
[0071] in, is the representation of the i-th node in the normal state.
[0072] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a comprehensive energy resilience assessment system based on multi-source factor coupling, including:
[0073] The physical model building module is used to build a multi-attribute coupled physical model of the urban integrated energy system, and uses a geographic correction function to correct the coupling effect of the model under different geographic coordinates during the construction;
[0074] A data processing module is used to obtain the spatiotemporal data of the urban integrated energy system and preprocess the spatiotemporal data to obtain preprocessed spatiotemporal data;
[0075] The spatiotemporal graph model construction module is used to construct a spatiotemporal graph model using a graph convolutional network based on the preprocessed spatiotemporal data;
[0076] The model fusion module is used to embed the output of the multi-attribute coupling physical model into the space-time graph model as an additional feature of the node to obtain the space-time graph model that integrates physical information;
[0077] The model updating module is used to obtain disaster data and update the spatiotemporal graph model integrating physical information based on the disaster data to obtain a dynamic assessment model;
[0078] The evaluation module is used to determine the dynamic change characteristics of the node status based on the dynamic evaluation model, and calculate the resilience index of the urban integrated energy system based on the dynamic change characteristics.
[0079] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0080] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0081] Figure 1 A schematic diagram of a flow chart of a comprehensive energy resilience assessment method based on multi-source factor coupling provided in Example 1 of the present application;
[0082] Figure 2 A schematic diagram of the structure of a comprehensive energy resilience assessment system based on the coupling of multiple source factors provided in an embodiment of the present application. DETAILED DESCRIPTION
[0083] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0084] The following describes the comprehensive energy resilience assessment method and system based on multi-source factor coupling according to an embodiment of the present application with reference to the accompanying drawings.
[0085] Figure 1 A flow chart of a comprehensive energy resilience assessment method based on the coupling of multiple source factors provided in Example 1 of the present application.
[0086] like Figure 1 As shown in the figure, the comprehensive energy resilience assessment method based on multi-source factor coupling includes the following steps:
[0087] Step 101, constructing a multi-attribute coupling physical model of an urban integrated energy system, and using a geographic correction function to correct the coupling effect of the model under different geographic coordinates during the construction;
[0088] Step 102, acquiring the spatiotemporal data of the urban integrated energy system, and preprocessing the spatiotemporal data to obtain preprocessed spatiotemporal data;
[0089] Step 103, constructing a spatiotemporal graph model using a graph convolutional network based on the preprocessed spatiotemporal data;
[0090] Step 104, embedding the output of the multi-attribute coupling physical model into the space-time graph model as an additional feature of the node to obtain a space-time graph model integrating physical information;
[0091] Step 105, obtaining disaster data, and updating the spatiotemporal graph model integrating physical information based on the disaster data to obtain a dynamic assessment model;
[0092] Step 106, determining the dynamic change characteristics of the node status based on the dynamic evaluation model, and calculating the resilience index of the urban integrated energy system based on the dynamic change characteristics.
[0093] The comprehensive energy resilience assessment method based on the coupling of multiple source factors in the embodiment of the present application constructs a multi-attribute coupled physical model of electricity, natural gas and thermal systems, combines geographic correction functions and symbolic filtering technology, establishes a data-driven graph convolutional network spatiotemporal graph model, dynamically adjusts edge weights and node characteristics under disaster scenarios, and comprehensively evaluates the disaster resistance and recovery capabilities of urban energy systems through indicators such as recovery time and system failure degree, thereby improving the accuracy and practicality of the assessment and providing a scientific basis for the optimization of urban energy systems.
[0094] Optionally, in one embodiment of the present application, constructing a multi-attribute coupling physical model of an urban integrated energy system includes:
[0095] Use the node admittance matrix method to construct the power system model;
[0096] Construct a natural gas system model based on gas flow equations;
[0097] The thermal system model is constructed using the heat conduction equation;
[0098] Based on the power system model, the natural gas system model and the thermal system model, the power system coupling model, the natural gas system coupling model and the thermal system coupling model are constructed respectively, and the geographic correction function is used to correct the coupling effect of the coupling model under different geographic coordinates during the construction.
[0099] Optionally, in one embodiment of the present application, the power system model is expressed as:
[0100]
[0101] in, Inject power vector into the node, is the node voltage vector, is the system admittance matrix, represents the Hadamard product, denotes conjugation;
[0102] The natural gas system model is expressed as:
[0103]
[0104] in, is the gas flow rate, is the flow coefficient, , are the inlet pressure and outlet pressure respectively;
[0105] The thermal system model is expressed as:
[0106]
[0107] in, is the heat flow, is the heat transfer coefficient, is the heat transfer area, , are the inlet temperature and outlet temperature respectively.
[0108] Optionally, in one embodiment of the present application, the geographic correction function , and Used to correct the coupling effects of various systems in different geographic coordinates. These correction functions can be linear or nonlinear, and fit according to actual conditions and historical data, such as:
[0109] Linear correction function:
[0110]
[0111] Nonlinear correction function:
[0112]
[0113] in, is the geographic location (such as longitude and latitude), is the height, is the geographical correction function of the power system coupling model, is the geographical correction function of the natural gas system coupling model, is the geographic correction function of the thermal system coupling model, , , , , , , , , , , , , , , , , , All are coefficients.
[0114] Optionally, in one embodiment of the present application, the power system Natural gas system 、Thermal system and its own variables Assume that the influence of geographic coordinates is corrected by the function After correction, the power system coupling model is as follows:
[0115]
[0116] in, , , is the coupling parameter, is the coupling function of the natural gas system to the power system, is the coupling function of the thermal system to the power system;
[0117] Natural gas system Power system 、Thermal system and its own variables The influence of geographic coordinates is corrected by the function After correction, the natural gas system coupling model is as follows:
[0118]
[0119] in, , , is the coupling parameter, is the coupling function of the thermal system to the natural gas system, is the coupling function of the power system to the natural gas system, V is the voltage of the power system, Indicates the power of the power system;
[0120] Power system output power It is determined by voltage V and other factors. Voltage V is also a coupling function of the natural gas system. Input, and There is an indirect connection between them through the physical parameters and coupling relationships of the power system;
[0121] In the power system model, P represents the power output of the entire power system; in the following node feature vector, P is the node injection power vector, which represents the power injected or consumed at each node in the power system.
[0122] Thermal system Power system , Natural gas system and its own variables The influence of geographic coordinates is corrected by the function After correction, the thermal system coupling model is as follows:
[0123]
[0124] in, , , is the coupling parameter, is the coupling function of the natural gas system to the thermal system, is the coupling function of the power system to the thermal system.
[0125] Optionally, in one embodiment of the present application, spatiotemporal data of the urban energy system is collected, including energy consumption data, meteorological data, disaster data, etc. Symbolic filtering technology is used to pre-process the data to reduce noise and highlight key features.
[0126] The collected data include energy consumption data, meteorological data and disaster data. Energy consumption data include electricity consumption data (recorded by region and time point), natural gas consumption data (recorded by region and time point) and heat consumption data (recorded by region and time point); meteorological data include temperature, humidity, wind speed, wind direction, precipitation (recorded by region and time point); disaster data include the occurrence time and affected area of natural disasters (such as earthquakes, floods, typhoons) and man-made disasters (such as fires, explosions).
[0127] Preprocess the data using symbol filtering techniques to reduce noise and highlight key features. Symbol filtering techniques convert continuous data into a sequence of symbols, thereby reducing random noise in the data and highlighting key features.
[0128] The specific steps include:
[0129] Symbolization: Divide continuous data into several intervals and represent each interval with a symbol. For example, temperature data can be divided into three symbols: "high", "medium", and "low";
[0130] Noise reduction: By symbolizing the data, short-term random fluctuations can be eliminated and long-term trends and important features can be retained.
[0131] Optionally, in one embodiment of the present application, constructing a spatiotemporal graph model includes:
[0132] Use graph convolutional networks (GCNs) to build spatiotemporal graph models. Define graph ,in Represents a collection of nodes. represents the edge set, the node feature vector is X, and the edge weight matrix is A.
[0133] The specific process of building a spatiotemporal graph model includes:
[0134] (1) Node definition
[0135] Each node Represents the state of a specific area at a specific point in time, and the node feature vector Contains the following information:
[0136]
[0137] in, For power consumption, For natural gas consumption, For heat consumption, is the temperature, For humidity, is the wind speed, For wind direction, is the precipitation, It is a disaster state.
[0138] (2) Edge definition
[0139] Edges represent the mutual influence between different nodes, and the elements of the edge weight matrix A are Representation Node and After the introduction of the symbol filtering technology, the edge weight matrix after symbol filtering is .
[0140] (3) Basic Graph Convolutional Network
[0141] The basic formula of graph convolutional network is:
[0142]
[0143] in, is the edge weight matrix after symbol filtering, For the The node representation of the layer, For the The weight matrix of the layer, is the activation function (such as ReLU);
[0144] (4) Sign filtering of edge weight matrix
[0145] Edge weight matrix after symbol filtering It can be calculated by the following steps:
[0146] Symbolization of edge weight matrix A: Symbolize the elements in edge weight matrix A to obtain symbolic matrix .
[0147] Symbol filtering operation: For the symbol matrix Perform filtering operation to obtain the matrix after symbol filtering .
[0148]
[0149] (5) Determine the graph convolution calculation formula
[0150] Considering the edge weight matrix after sign filtering , the calculation formula of the graph convolutional network is as follows:
[0151]
[0152] Initial input: node feature matrix
[0153] Output: Node representation matrix after multi-layer graph convolution .
[0154] Optionally, in one embodiment of the present application, the output of the physical model is embedded into the spatiotemporal graph model as part of the node features, so as to better capture the dynamic behavior and coupling relationship of the physical system.
[0155] The original node feature vector X contains the spatiotemporal data of the urban energy system, which converts the output of the physical model (electric power , Gas flow , heat flow ) as an additional feature, the expanded node feature vector is:
[0156]
[0157] in, is the expanded node feature vector, is the node feature vector, Inject power vector into the node, is the gas flow rate, is the heat flow;
[0158] In graph convolution calculations, the weight of physical information is considered to better integrate physical models with data-driven models.
[0159] The calculation formula of graph convolution after introducing physical information is:
[0160]
[0161] in, is the edge weight matrix after symbol filtering, For the The node representation of the layer, For the The weight matrix of the layer, is the expanded node feature vector, For the The physical information weight matrix of the layer, is the activation function (such as ReLU).
[0162] Optionally, in one embodiment of the present application, the spatiotemporal graph model integrating physical information is updated based on disaster data to obtain a dynamic assessment model, including:
[0163] In disaster scenarios, the edge weight matrix and node features of the spatiotemporal graph model need to be dynamically updated to reflect the impact of the disaster on the system;
[0164] Specifically, the edge weight matrix is dynamically updated, including:
[0165] In a disaster scenario, the edge weight matrix A needs to be adjusted according to the disaster impact matrix D. The elements of the disaster impact matrix D represent the degree of disaster impact in a specific area and time point.
[0166] Updated edge weight matrix The calculation formula is:
[0167]
[0168] in, represents the Hadamard product, i.e., element-wise multiplication. Through this formula, the edge weights between areas that are more affected by the disaster will be weakened or completely eliminated.
[0169] Update node features, including:
[0170] Node Features It also needs to be updated according to the disaster situation to reflect the impact of the disaster on the node status. Can be dynamically adjusted to include hazard information, for example by extending the feature vector To reflect the impact of disasters:
[0171]
[0172] Where D represents the disaster impact degree of the node.
[0173] Updated dynamic evaluation model, including:
[0174] In disaster scenarios, a graph convolutional network (GCN) is used for dynamic evaluation. The node representation is calculated through forward propagation, and the node representation at different time steps t is evaluated to obtain dynamic change characteristics.
[0175] (1) Dynamic Evaluation Formula of Graph Convolutional Network
[0176] In the disaster scenario, the edge weight matrix after considering the disaster impact and the expanded node feature vector , the calculation formula of the graph convolutional network is:
[0177]
[0178] in, is the edge weight matrix after considering the impact of disasters, For the The node representation of the layer, For the The weight matrix of the layer, is the expanded node feature vector (including disaster impact information), For the The physical information weight matrix of the layer, is the activation function (such as ReLU).
[0179] (2) Dynamic change feature evaluation
[0180] Through the forward propagation of the graph convolutional network, the node representation can be calculated at different time steps t , thereby evaluating the dynamic change characteristics of the node status.
[0181] The specific steps are as follows:
[0182] Initialization node representation: ;
[0183] Forward propagation: For each time step t, compute the node representation
[0184]
[0185] Dynamic change characteristics: By comparing the node representations at different time steps t and t+1 and , evaluate the dynamic change characteristics of node status.
[0186] Optionally, in one embodiment of the present application, the resilience indicators of the urban energy system are calculated through the node representation H, mainly including the recovery time (Recovery Time, RT) and the system failure degree (System Failure Degree, SFD), etc., to evaluate the system's disaster resistance and recovery capabilities under disaster scenarios.
[0187] Recovery time refers to the time required for the city's energy system to return to normal operation after a disaster occurs. The specific calculation formula is as follows:
[0188]
[0189] in, The assessment period after the disaster. is the number of nodes, It is the representation of the i-th node at the t-th time, usually taking some quantitative representation of the node state, such as node voltage, power load, etc.
[0190] System failure degree refers to the degree of failure of the system after a disaster occurs, that is, the performance degradation of the system. It can be defined as the difference between the overall performance of the system after a disaster occurs and the normal state. The specific calculation formula is as follows:
[0191]
[0192] in, is the representation of the i-th node in the normal state.
[0193] Based on the calculation results of the recovery time (RT), we can evaluate the time required for the system to recover from a failed state to a normal state after a disaster. The shorter the RT, the stronger the system's recovery capability and the higher its disaster resilience. By analyzing the recovery time of different nodes, we can identify the weak links in the system that recover slowly and take targeted improvement measures.
[0194] The above two resilience indicators can be used to comprehensively evaluate the disaster resistance and recovery capabilities of urban energy systems in disaster situations. Specifically,
[0195] The calculation results of the system failure degree (SFD) can be used to evaluate the failure degree of the system after a disaster. The smaller the SFD, the less the system performance decreases and the stronger the disaster resistance. By comparing the SFD at different time steps, the dynamic process of system performance recovery can be understood and the key recovery stage of the system after a disaster can be identified.
[0196] The comprehensive energy resilience assessment method based on multi-source factor coupling in the embodiment of the present application effectively simulates the mutual influence between different energy systems by constructing a multi-attribute coupled physical model of electricity, natural gas and thermal systems, overcoming the limitation of independent modeling of a single system in traditional methods; using symbolic filtering technology and geographic correction functions, combined with graph convolutional networks (GCN), dynamic adjustment and efficient assessment in disaster scenarios are achieved, so that the model can update edge weights and node features in real time according to disasters, significantly improving the accuracy and practicality of the assessment; embedding the output of the physical model into the spatiotemporal graph model, fully combining the advantages of data-driven and physical information, improving the ability to capture the dynamic behavior and coupling relationship of the system, and achieving more accurate assessment; by calculating resilience indicators such as recovery time (RT) and system failure degree (SFD), the disaster resistance and recovery capabilities of the urban energy system in disaster scenarios are comprehensively evaluated, providing detailed dynamic change characteristics, and providing a scientific basis for system optimization and decision-making. This embodiment is applicable to a variety of disaster scenarios and complex urban energy systems, has good practicality and scalability, and can provide strong technical support for urban energy management and disaster response.
[0197] The comprehensive energy resilience assessment method based on the coupling of multiple source factors of the present application is described in detail below through specific embodiments.
[0198] In this embodiment, the comprehensive energy resilience assessment method based on multi-source factor coupling specifically includes:
[0199] Step 1: Select a specific time point and area
[0200] In the implementation case, it is assumed that there are three regions: Region 1, Region 2, and Region 3.
[0201] Step 2: Prepare the data table
[0202] Prepare energy consumption data table, including electricity consumption, natural gas consumption and heat consumption. The data table is as follows:
[0203]
[0204] Step 3: Calculate node feature vector
[0205] Extract the energy consumption data of the three regions at the time point 00:00 from the data table and obtain the node feature vector:
[0206] Region 1: [100,50,80], Region 2: [120,55,85], Region 3: [90,45,70]
[0207] Step 4: Build a spatiotemporal graph model
[0208] In the spatiotemporal graph model, each region is considered as a node, and the connections between nodes represent energy transmission paths. For example, the transmission paths of electricity, natural gas, and heat between different regions can constitute the edges of the spatiotemporal graph.
[0209] Step 5: Embed physics information into the data-driven model
[0210] Assume the following output of the physical model:
[0211] Electric power: [130,140,110], Natural gas flow: [60,65,50], Heat flow: [95,100,85]
[0212] The outputs of these physical models are added to the node features to obtain the expanded node feature vector:
[0213] Region 1: [100,50,80,130,60,95], Region 2: [120,55,85,140,65,100], Region 3: [90,45,70,110,50,85]
[0214] Step 6: Dynamic Evaluation Methods
[0215] In this step, the system resilience assessment will be conducted. Two indicators are selected: recovery time (RT) and system failure degree (SFD).
[0216] 1) Recovery time (RT) calculation
[0217] For each region, the average value of its energy consumption is calculated, that is, the average consumption of each energy source.
[0218] Region 1:
[0219]
[0220] Region 2:
[0221] Region 3:
[0222] 2) System failure degree (SFD) calculation
[0223] For each area, the average of the absolute values of the differences between each energy consumption and the energy consumption in the normal state is calculated.
[0224] Region 1:
[0225]
[0226] Region 2:
[0227]
[0228] Region 3:
[0229]
[0230] In order to implement the above embodiments, the present application also proposes a comprehensive energy resilience assessment system based on the coupling of multiple source factors.
[0231] Figure 2 A schematic diagram of the structure of a comprehensive energy resilience assessment system based on the coupling of multiple source factors provided in an embodiment of the present application.
[0232] like Figure 2 As shown in the figure, the comprehensive energy resilience assessment system based on multi-source factor coupling includes:
[0233] The physical model building module is used to build a multi-attribute coupled physical model of the urban integrated energy system, and uses a geographic correction function to correct the coupling effect of the model under different geographic coordinates during the construction;
[0234] A data processing module is used to obtain the spatiotemporal data of the urban integrated energy system and preprocess the spatiotemporal data to obtain preprocessed spatiotemporal data;
[0235] The spatiotemporal graph model construction module is used to construct a spatiotemporal graph model using a graph convolutional network based on the preprocessed spatiotemporal data;
[0236] The model fusion module is used to embed the output of the multi-attribute coupling physical model into the space-time graph model as an additional feature of the node to obtain the space-time graph model that integrates physical information;
[0237] The model updating module is used to obtain disaster data and update the spatiotemporal graph model integrating physical information based on the disaster data to obtain a dynamic assessment model;
[0238] The evaluation module is used to determine the dynamic change characteristics of the node status based on the dynamic evaluation model, and calculate the resilience index of the urban integrated energy system based on the dynamic change characteristics.
[0239] It should be noted that the aforementioned explanation of the embodiment of the comprehensive energy resilience assessment method based on the coupling of multiple source factors is also applicable to the comprehensive energy resilience assessment system based on the coupling of multiple source factors of this embodiment, and will not be repeated here.
[0240] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.
[0241] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0242] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0243] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
[0244] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0245] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0246] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0247] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A comprehensive energy resilience assessment method based on the coupling of multiple source factors, It is characterized in that include: Construct a multi-attribute coupled physical model of the urban integrated energy system, and use a geographic correction function to correct the coupling effect of the model under different geographic coordinates during construction; Acquiring spatiotemporal data of an urban integrated energy system, and preprocessing the spatiotemporal data to obtain preprocessed spatiotemporal data; Based on the preprocessed spatiotemporal data, a spatiotemporal graph model is constructed using a graph convolutional network; The output of the multi-attribute coupling physical model is embedded into the space-time graph model as an additional feature of the node to obtain a space-time graph model integrating physical information; Obtaining disaster data, and updating the spatiotemporal graph model integrating physical information based on the disaster data to obtain a dynamic assessment model; Determining the dynamic change characteristics of the node status based on the dynamic evaluation model, and calculating the resilience index of the urban integrated energy system based on the dynamic change characteristics; The multi-attribute coupling physical model of the urban integrated energy system is constructed, including: Use the node admittance matrix method to construct the power system model; Construct a natural gas system model based on gas flow equations; The thermal system model is constructed using the heat conduction equation; Based on the power system model, natural gas system model and thermal system model, a power system coupling model, a natural gas system coupling model and a thermal system coupling model are respectively constructed, and a geographic correction function is used to correct the coupling effect of the coupling model at different geographic coordinates during the construction; The power system model is expressed as: in, Inject power vector into the node, is the node voltage vector, is the system admittance matrix, represents the Hadamard product, denotes conjugation; The natural gas system model is expressed as: in, is the gas flow rate, is the flow coefficient, , are the inlet pressure and outlet pressure respectively; The thermal system model is expressed as: in, is the heat flow, is the heat transfer coefficient, is the heat transfer area, , are the inlet temperature and outlet temperature respectively; The geographic correction function is in a linear form or a nonlinear form. The form of the geographic correction function is set according to the state and historical data of the urban integrated energy system. The linear form of the geographic correction function is expressed as: The nonlinear form of the geographic correction function is expressed as: in, For geographical location, is the height, is the geographical correction function of the power system coupling model, is the geographical correction function of the natural gas system coupling model, is the geographic correction function of the thermal system coupling model, , , , , , , , , , , , , , , , , , All are coefficients; The power system coupling model is expressed as: in, , , is the coupling parameter, is the coupling function of the natural gas system to the power system, is the coupling function of the thermal system to the power system; The natural gas system coupling model is expressed as: in, , , is the coupling parameter, is the coupling function of the thermal system to the natural gas system, is the coupling function of the power system to the natural gas system, V is the voltage of the power system, Indicates the power of the power system; The thermal system coupling model is expressed as: in, , , is the coupling parameter, is the coupling function of the natural gas system to the thermal system, is the coupling function of the power system to the thermal system; The spatiotemporal data are data recorded by region and time point, the spatiotemporal data are energy consumption data, meteorological data and disaster data, and the energy consumption data are electricity consumption data, natural gas consumption data and heat consumption data; The preprocessing of the spatiotemporal data to obtain the preprocessed spatiotemporal data includes: Preprocessing the spatiotemporal data by using a symbol filtering technique to obtain preprocessed spatiotemporal data, wherein the preprocessing process includes symbolization and noise reduction, and the symbolization process includes: dividing continuous data into a plurality of intervals, and representing each interval with a symbol; Based on the preprocessed spatiotemporal data, a spatiotemporal graph model is constructed using a graph convolutional network, including: Construct a spatiotemporal graph based on the preprocessed spatiotemporal data ,in, , Represents a collection of nodes. represents the edge set, the node feature vector is X, the edge weight matrix is A, and the elements of the edge weight matrix A are Representation Node and The mutual influence strength between them, each node represents the state of a specific area at a specific time point, and the node feature vector , For power consumption, For natural gas consumption, For heat consumption, is the temperature, For humidity, is the wind speed, For wind direction, is the precipitation, is a state of disaster; Based on the spatiotemporal graph, a spatiotemporal graph model is constructed using a graph convolutional network, wherein the calculation formula of the graph convolutional network is: Among them, the initial input of the graph convolutional network , the output of the graph convolutional network is the node representation matrix after multi-layer graph convolution, For the The node representation of the layer, For the The weight matrix of the layer, is the edge weight matrix after symbol filtering; The output of the physical model is used as an additional feature of the node, and the expanded node feature vector is expressed as: in, is the expanded node feature vector, is the node feature vector, Inject power vector into the node, is the gas flow rate, is the heat flow; The calculation formula of graph convolution after introducing physical information is: in, For the The physical information weight matrix of the layer.
2. The comprehensive energy resilience assessment method based on multi-source factor coupling as claimed in claim 1, It is characterized in that The spatiotemporal graph model integrating physical information is updated based on the disaster data to obtain a dynamic assessment model, including: Based on the disaster data, an edge weight matrix and an expanded node feature vector are updated to obtain a new edge weight matrix and a node feature vector; Based on the new edge weight matrix and node feature vector, the graph convolutional network after introducing the physical information is updated, and an updated spatiotemporal graph model integrating the physical information is determined, and the updated spatiotemporal graph model integrating the physical information is used as the dynamic evaluation model; Determining the dynamic change characteristics of the node state based on the dynamic evaluation model includes: By forward propagation of the dynamic evaluation model, node representations are calculated at different time steps to evaluate the dynamic change characteristics of node states; Among them, the new edge weight matrix is expressed as: in, is the disaster impact matrix, The elements represent the degree of disaster impact in a specific area and time point. represents the Hadamard product; The new node feature vector is expressed as: The calculation formula of the updated graph convolutional network is: Initial input to the updated graph convolutional network , during forward propagation, the node at time step t+1 is represented as , by comparing the node representations at time step t and time step t+1 and Evaluate the dynamic changes of node status.
3. The comprehensive energy resilience assessment method based on multi-source factor coupling as claimed in claim 2, It is characterized in that The resilience index of the urban integrated energy system is the recovery time and the system failure degree, and the recovery time and the system failure degree are negatively correlated with the resilience index; The calculation formula of the recovery time is: in, The assessment period after the disaster. is the number of nodes, is the representation of the i-th node at the t-th time; The calculation formula of the system failure degree is: in, is the representation of the i-th node in the normal state.
4. A comprehensive energy resilience assessment system based on multi-source factor coupling, It is characterized in that The system implements the comprehensive energy resilience assessment method according to claim 1, and the system comprises: The physical model building module is used to build a multi-attribute coupled physical model of the urban integrated energy system, and uses a geographic correction function to correct the coupling effect of the model under different geographic coordinates during the construction; A data processing module is used to obtain the spatiotemporal data of the urban integrated energy system and preprocess the spatiotemporal data to obtain preprocessed spatiotemporal data; A spatiotemporal graph model construction module, used to construct a spatiotemporal graph model using a graph convolutional network based on the preprocessed spatiotemporal data; A model fusion module, used to embed the output of the multi-attribute coupling physical model into the space-time graph model as an additional feature of the node, so as to obtain a space-time graph model integrating physical information; A model updating module, used for acquiring disaster data, and updating the spatiotemporal graph model integrating physical information based on the disaster data to obtain a dynamic assessment model; An evaluation module is used to determine the dynamic change characteristics of the node status based on the dynamic evaluation model, and calculate the resilience index of the urban integrated energy system based on the dynamic change characteristics.
