Conduction logic and cascade model analysis method in energy power field

Abstract

The invention discloses a conduction logic and cascade model analysis method in the field of energy power, and the method comprises the steps: data collection and preprocessing: collecting global energy trade data, including energy inlet and outlet quantity, price and the like, and carrying out the preprocessing to construct a global energy cascade conduction network; and model establishment: based on the collected data, an under-load cascade fault model considering overload limitation is constructed, and the model can simulate a cascade fault process in an energy trade network. And scene setting and simulation: setting different scenes, simulating node load upper limit and lower limit fluctuation, and performing simulation analysis by using the model. And result analysis and report: according to the simulation result, analyzing the influence of the international energy power pattern on the energy safety of China, and generating an analysis report to provide reference for policy formulators. According to the invention, through simulation under different scenes, the influence of the international energy power pattern on the energy power development of China can be found.

Description

Technical Field

[0001] The present invention relates to the field of energy and electricity, and in particular to a conduction logic and cascade model analysis method for analyzing the impact of an international energy and electricity structure on domestic energy and electricity development. Background Art

[0002] In recent years, my country is vigorously promoting electrification, actively promoting the construction of a new generation of power systems with smart grids as the core, and promoting the construction of a clean, low-carbon, safe and efficient energy system. On the electricity consumption side, electric energy is widely substituted for terminal fossil energy such as coal and oil, thereby significantly increasing the proportion of electric energy in terminal energy consumption; on the supply side, renewable energy is introduced on a large scale to replace traditional fossil energy. With the transformation of the international energy structure and economic development mode, my country's energy conservation, emission reduction and new energy industry development will present a new trend of green environmental protection, not only in the industrial field, but also in the fields of consumption and trade. my country will gradually form an energy structure with "electricity as the center and renewable energy as the main energy application method". The profound adjustment of the international energy structure has brought new opportunities for China's international energy trade and cooperation, but also created new challenges.

[0003] In order to explore the relationship between the international energy and power structure and my country's domestic energy and power development, many experts and scholars at home and abroad have conducted relevant research. However, theoretical research has lagged far behind practical requirements. This is mainly reflected in the following aspects: there is no clear relationship between changes in the international energy and power structure and the domestic energy and power development situation. There are both connections and differences between the two. It is necessary to conduct a systematic study on the internal mechanism of how the external energy and power structure affects the domestic energy and power development situation.

[0004] In the context of economic globalization, energy supply and demand between countries are closely linked. In complex networks, the failure of regional factors can lead to global disasters, which is called cascading failure. The cascading failure process has been revealed in networks such as physical infrastructure networks, economic and financial networks, and Internet networks. Taking the economic and trade network as an example, the cascading failure process can be described as: once a certain product in a country has a supply and demand problem, then due to global economic and trade relations, it will trigger a global cascading failure in the trade network. However, the cascading failure method is still rarely used in the study of energy trade systems, and cascading failure analysis rarely considers both overload and underload states at the same time.

[0005] In summary, the current energy and power sector needs a more complete analytical method to study the impact of the international energy and power structure on domestic energy and power development based on conduction logic and cascade models. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a conduction logic and cascade model analysis method in the field of energy and electricity. The method establishes a global energy cascade conduction network and combines it with an underload cascade failure model that considers overload limitations to evaluate the impact of the international energy and electricity structure on my country's energy security.

[0007] The technical solution of the present invention is as follows: A conduction logic and cascade model analysis method in the field of energy and power, comprising the following steps:

[0008] (1) Data collection and preprocessing: Collect global energy trade data, including energy import and export volumes and prices, and perform preprocessing to build a global energy cascade transmission network;

[0009] (2) Model building: Based on the collected data, an underload cascading failure model considering overload limits is constructed. This model can simulate the cascading failure process in the energy trading network.

[0010] (3) Scenario setting and simulation: Set different scenarios, simulate the fluctuation of the upper and lower limits of node load, and use the model for simulation analysis;

[0011] (4) Result analysis and reporting: Based on the simulation results, analyze the impact of the international energy and power structure on my country's energy security and generate an analysis report to provide reference for policymakers.

[0012] Preferably, in step (1), the data collection includes multiple text file formats such as PDF, TXT, and WORD. When processing different types of text content, it is necessary to first determine the text type, and then use the corresponding program to process the file and extract the text information in the file; if multiple files are input at the same time, it is necessary to process the content of each file separately, and merge all the text contents as input for the next step of processing; after extracting the text content, merge all the text contents and proceed to the next step of processing; by processing multiple types of text files, useful information can be effectively extracted and support can be provided for subsequent analysis and decision-making.

[0013] More specifically, for PDF files, you can use a PDF parser to extract text information; for TXT files, you can directly read the text content; for WORD files, you can use the API that comes with Microsoft Office to extract text information.

[0014] Specifically, in step (2), the modeling method of the underload cascading failure model considering the overload limit is as follows:

[0015] Step 1: The basic components of the network model, which are also global invariants, are listed below:

[0016] Initial load at point i

[0017] Formula (1)

[0018] Export in formula i represents the total export volume of node i to all its export nodes;

[0019] Upper load limit at point i and lower limit

[0020] , Formula (2)

[0021] Formula (3)

[0022] α represents a fixed upper limit parameter, which is a floating point data greater than or equal to 1; β represents a fixed lower limit parameter, which is a floating point data greater than 0 and less than 1.

[0023] represents the upper limit of the load at point i, passing through the initial load at the initial point i It can be calculated with the upper limit parameter α.

[0024] represents the lower limit of the load at point i, passing through the initial load at the initial point i With lower limit parameters β and β and import volume The minimum value of the calculated result is obtained.

[0025] Step 2: First-level allocation rules for network models

[0026] Distribution rules

[0027] In order to better describe the distribution rules and specific connection relationships of nodes in the network, the following distinctions are made between different types of nodes:

[0028] n——the country of the initial faulty node;

[0029] t – country from which n is imported;

[0030] k – countries that export to t, but not including n.

[0031] Primary Allocation Process

[0032] Once the initial node n fails, the export quota from node n will be proportionally distributed to other exporting countries k that have trade relations with the target importing country t. This process is defined as primary distribution in the present invention. The load change brought by the primary distribution process to this part of the edges in the network can be expressed as follows:

[0033] + Formula (4)

[0034] In the formula, EW represents the load of an edge in the network, and the subscript indicates that this load represents the export amount from one node to another node. represents the initial edge load from node k to t; represents the initial edge load from node n to t, represents the sum of the initial edge loads from all nodes j to t.

[0035] In order to ensure that the first-level allocation process does not bring too much export pressure to the country represented by node k, the load upper limit of the initial edge from node k to t is defined here , which can be expressed as follows:

[0036] Formula (5)

[0037] The edge load EW of the first-level allocation is determined by whether it is greater than or equal to the edge load upper limit EC, and the edge load is updated accordingly. It can be expressed as the following formula:

[0038] Formula (6)

[0039] This formula shows that this model takes overload limits into consideration. The partial edge overloads caused by the above-mentioned primary allocation process need to be within a limited range. However, if the load exceeds the upper limit, the model will not fail. Instead, the edge loads that are overloaded in the model will be calculated based on the upper limit.

[0040] Step 3: Secondary allocation rules for network models

[0041] Distribution rules

[0042] After the first-level allocation, some of the edges from node k to node t are overloaded, while others are not. The following distinction is made between the descriptions of the nodes corresponding to different edges:

[0043] l——In formula 6 , the original k nodes that are overloaded in the first-level allocation;

[0044] m——In formula 6 ; The original k nodes that are not overloaded in the first-level allocation;

[0045] Secondary Allocation Process

[0046] After the first-level allocation, the edge load corresponding to the overloaded l node is equal to its upper limit, and the part exceeding the upper limit needs to be further allocated to the edge corresponding to the m node that is not overloaded in the first-level allocation. The present invention defines this process as second-level allocation.

[0047] The edge load and upper limit corresponding to the l-node and m-node before the secondary allocation can be expressed as follows:

[0048] Formula (7)

[0049] The sum of the parts exceeding the upper limit that need to be allocated to the edges corresponding to the m nodes in the secondary allocation can be expressed as follows:

[0050] Formula (8)

[0051] The load change brought by the secondary allocation process to the edge from the exit of node m to node t in the network and the load after the load upper limit of the edge is finally updated can be expressed as follows:

[0052] Formula (9)

[0053] This formula shows that the update process of the secondary allocation is similar to that of the primary allocation, and the load of the edge where overload occurs is calculated based on the upper limit.

[0054] Model parameter settlement

[0055] Finally, statistics are taken for all nodes in the network, and the final load of the nodes after two-level allocation is calculated, which can be expressed as follows:

[0056] Formula (10)

[0057] Model Performance Evaluation

[0058] For the network built in S3, the underload cascading failure model considering overload limits has the following four important variable parameters:

[0059] The initial fault node represents the node country where the fault first occurred in the world energy trade network set in the model simulation;

[0060] Initial load of the node , see formula (1) for details, which represents the total export volume of a country. When the network load exceeds a certain amount, the higher the load, the more likely the network will fail;

[0061] Node Cap Parameters in , see formula (2) for details. It is generally believed that the network with a higher node upper limit can better cope with attacks;

[0062] Node lower limit Parameters in , see formula (3) for details. It is generally believed that the network with a higher node lower limit is more fragile and prone to cascading failures due to underload.

[0063] Corresponding to the different parameters to be changed in the simulation experiment, the present invention proposes the following two different evaluation scenarios:

[0064] Upper limit increase scenario: Ensure that the parameter in ≥1, simulating the scenario where each national node in the network actively adjusts and recovers.

[0065] Load increase upper limit decrease scenario: Ensure that the parameter in ≤1, and after calculating , make the initial load of the node greater than the original value, simulating the scenario where each national node in the network falls into panic and does not actively adjust.

[0066] There are the following performance evaluation criteria:

[0067] The robustness of the network can be expressed by the following formula:

[0068] Formula (12)

[0069] TN represents the total number of nodes in the network. In the formula, represents the upper limit parameter in formula (2) when, the number of effective nodes in the network after cascading failure simulation.

[0070] The determination method of the effective node NE is - if a node, after the settlement of formula (10), the node load is greater than or equal to the node load lower limit at formula (3), then there is no underload, and such a node is identified as an effective node, as follows:

[0071] Formula (13)

[0072] The resilience of the network is generally expressed in the following formula:

[0073] In the upper limit increase scenario, it体现为 the recovery ability of the network to actively adjust and can be expressed by the following formula:

[0074] a>1 Formula (14)

[0075] In the load increase upper limit decrease scenario, it体现为 the resistance of the network to panic and can be expressed by the following formula:

[0076] 0<b<1 Formula (15)

[0077] Note: In the translation, "体现为" is directly translated as "体现为" as there is no more context to accurately translate it into a more appropriate English expression. It might need to be adjusted according to the specific meaning in the relevant technical context.Beneficial effects of the present invention: (1) The method of the present invention considers the cross-regional transmission mechanism from the conflict outbreak node to the occurrence of the joint failure, from the local to the surrounding, from the single to the multiple cross-regional transmission mechanism, from the offline to the online cross-field transmission mechanism, establishes an underload cascading failure model with overload limitation, studies the transmission mechanism between the international energy and power structure and the domestic energy and power development, and reveals the causal relationship between the international energy and power structure and the domestic energy and power development.

[0078] (2) On the one hand, the present invention uses multi-text information recognition technology to provide a basis for indicator selection and quantification. On the other hand, it improves and develops based on the existing cascading failure model, connecting the conduction mechanism-model construction-cascading impact together, providing technical support for constructing the impact of geopolitical relations on the domestic energy and power structure, and also providing a basis for simulating future evolution scenario analysis models.

[0079] (3) The method for constructing a knowledge graph of the impact of energy structure based on text mining in the present invention extracts the impact of foreign language feature words ("international energy electrification status", "global electricity consumption", "global renewable energy power generation installed capacity") from relevant literature on the international energy and power structure through model training, matches the relationship between foreign language features and Chinese features, and obtains the cascading impact of some international feature factors (such as "international energy electrification status") on domestic feature factors (such as "energy consumption structure"). Based on Chinese reports, databases, etc., starting from Chinese features, the impact on the development of the domestic energy and power structure is predicted. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 This is a diagram showing the transmission logic of the international energy and power structure studied in this invention to the development of my country's domestic energy and power;

[0081] Figure 2 A flow chart of a method for constructing a knowledge graph of energy landscape impact based on text mining in the present invention;

[0082] Figure 3 This is a flowchart of multi-type text processing in the present invention;

[0083] Figure 4 It is a basic flow chart of the simulation of the present invention;

[0084] Figure 5 A distribution map of global energy structure contribution obtained by preprocessing data before simulation of the present invention;

[0085] Figure 6 A line graph showing the impact of the upper limit increase scenario and the load increase upper limit reduction scenario on the global energy trade network shown in the simulation of the present invention;

[0086] Figure 7Quadrant diagrams of network robustness and resilience in the upper limit increase scenario and load increase-upper limit reduction scenario shown in the simulation of the present invention; DETAILED DESCRIPTION

[0088] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and simulation schemes.

[0089] The basic process of the present invention is as follows Figure 4 As shown, the simulation is performed according to this process:

[0090] First, obtain the trade data of international energy and other related fields, and complete the network modeling of the cascade model. After that, before verifying the network in the present invention, the contribution of Russia, Ukraine and other countries to the global coal, oil and natural gas trade can be analyzed by analyzing the contribution of the global energy structure. Figure 5 The global energy structure contribution distribution map is shown.

[0091] The detailed simulation results of the data are as follows:

[0092] ULIS, the upper limit increase scenario:

[0093] In group a, the lower limit of nodes was kept unchanged at 0.4, and the upper limit of nodes was increased from 1.0 to 1.5;

[0094] In group b, the upper limit of nodes is kept unchanged at 1.3, and the lower limit of nodes is increased from 0.1 to 0.6;

[0095] LIULDS, load increase upper limit reduction scenario:

[0096] In the experiment of group c, the lower limit of nodes was kept unchanged at 0.4, the upper limit of nodes was reduced from 1.0 to 0.5, and the load was increased from 1.0 to 1.5;

[0097] In group d, the node upper limit is kept at 0.7 and the load is kept at 1.3, while the node lower limit is increased from 0.1 to 0.6.

[0098] Through simulations under different scenarios, we can find out the impact of the international energy and power pattern on the development of my country's energy and power. After the lower limit of the node load is determined, it can be found that as the upper limit of the node load increases, the number of retained countries is also increasing. Once the upper limit is raised to 120% of the current node load, all four energy networks will maintain 80% of the countries. The integrated energy network and the oil trade network are the most robust, and they can maintain more than 80% of the countries even if other countries do not have additional production capacity. In the load increase-upper limit reduction scenario, due to concerns about energy shortages, the amplification effect of energy demand and storage. As Figure 5Shown are the impacts on the global energy trading network based on a cap increase scenario and a load increase-cap reduction scenario.

[0099] The model provides a method to evaluate the robustness and resilience of the network to the international energy and power structure under the upper limit increase scenario and the load increase-upper limit decrease scenario. Under the upper limit increase scenario and the load increase-upper limit decrease scenario, the node load upper limit is set to 0.3 and the node load lower limit is set to 0.4, and the impact of the international energy and power structure on my country's energy and power development is evaluated. By observing the bubble size of the network, it can be found that in the energy network, the energy and oil integrated network has a strong robustness to the international energy and power structure ( Figure 6 ).

[0100] Under the scenario of load increase and upper limit reduction considering global panic, the model found that a large proportion of national nodes were destroyed or damaged, of which 72 national nodes in the energy network failed. Most of the failed nodes are small countries in Africa and Central Asia, such as Afghanistan, Pakistan, Algeria, Libya, Oman, Tunisia, etc. 1,644 edges were destroyed in the energy network, and the edges of energy trade flows with large reductions include Canada to the United States, Russia, Indonesia and Saudi Arabia to China, Australia to Japan, etc. The edges with large increases include Canada to Japan, Nigeria to the United States, Canada and Nigeria to China, and the United States to Turkey. Through model simulation, we can find the importance of the international energy and power pattern on the development of my country's energy and power. In the energy network, the natural gas and coal trade networks are more significantly affected by international relations than the oil network. The more diverse the energy types, the more likely we can survive in a dangerous world.

Claims

1. A conduction logic and cascade model analysis method in the field of energy and power, characterized in that: The following steps are involved: (1) Data collection and preprocessing: Collect global energy trade data, including energy import and export volume, price, etc., and preprocess them to build a global energy cascade transmission network; (2) Model establishment: Based on the collected data, construct an underload cascading failure model considering overload limits. This model can simulate the cascading failure process in the energy trade network; (3) Scenario setting and simulation: Set different scenarios, simulate the fluctuations of the upper and lower limits of node load, and use the model for simulation analysis; (4) Result analysis and reporting: Based on the simulation results, analyze the impact of the international energy and power structure on my country's energy security, and generate an analysis report to provide reference for policymakers.

2. The method according to claim 1, characterized in that: In step (1), data collection includes multiple text file formats including PDF, TXT, and WORD. When processing different types of text content, it is necessary to first determine the text type, and then use the corresponding program to process the file and extract the text information in the file; if multiple files are input at the same time, it is necessary to process the content of each file separately and merge all the text contents as input for the next step of processing; after extracting the text content, merge all the text contents and proceed to the next step of processing; by processing multiple types of text files, useful information can be effectively extracted and provide support for subsequent analysis and decision-making.

3. The method according to claim 2, characterized in that For PDF files, you can use a PDF parser to extract text information; for TXT files, you can directly read the text content; for WORD files, you can use the Microsoft Office built-in API to extract text information.

4. The method according to claim 1, characterized in that: In step (2), the modeling method of the underload cascading failure model considering overload limit is as follows: Step 1: The basic components of the network model, which are also global invariants, are listed below: Initial load at point i Formula (1) Export in formula i represents the total export volume of node i to all its export nodes; Upper load limit at point i and lower limit , Formula (2) Formula(3) α represents a fixed upper limit parameter, which is a floating point data greater than or equal to 1; β represents a fixed lower limit parameter, which is a floating point data greater than 0 and less than 1; represents the upper limit of the load at point i, passing through the initial load at the initial point i It can be calculated with the upper limit parameter α; represents the lower limit of the load at point i, passing through the initial load at the initial point i With lower limit parameters β and β and import volume The minimum value of the calculated result is obtained; Step 2: First-level allocation rules for network models Distribution rules In order to better describe the distribution rules and specific connection relationships of nodes in the network, the following distinctions are made between different types of nodes: n——the country of the initial faulty node; t – country from which n is imported; k – countries that export to t, but not including n; Primary Allocation Process Once the initial node n fails, the export quota from node n will be proportionally distributed to other exporting countries k that have trade relations with the target importing country t. This process is defined as primary distribution in the present invention. The load change brought by the primary distribution process to this part of the edges in the network can be expressed as follows: + Formula(4) In the formula, EW represents the load of an edge in the network, and the subscript indicates that this load represents the export amount from one node to another node; represents the initial edge load from node k to t; represents the initial edge load from node n to t, represents the sum of the initial edge loads from all nodes j to t; In order to ensure that the first-level allocation process does not bring too much export pressure to the country represented by node k, the load upper limit of the initial edge from node k to t is defined here , which can be expressed as follows: Formula (5) The edge load EW of the first-level allocation is determined by whether it is greater than or equal to the edge load upper limit EC, and the edge load is updated accordingly. It can be expressed as the following formula: Formula (6) This formula shows that this model takes overload limits into account. The overload of some edges caused by the above-mentioned primary allocation process needs to be within a limited range. However, if the load exceeds the upper limit, the model will not fail. Instead, the load of the overloaded edge in the model will be calculated based on the upper limit. Step 3: Secondary allocation rules for network models Distribution rules After the first-level allocation, some of the edges from node k to node t are overloaded, while others are not. The following distinction is made between the descriptions of the nodes corresponding to different edges: l——In formula 6 , the original k nodes that are overloaded in the first-level allocation; m——In formula 6 ; The original k nodes that are not overloaded in the first-level allocation; Secondary Allocation Process After the first-level allocation, the edge load corresponding to the overloaded l-node is equal to its upper limit, and the part exceeding the upper limit needs to be further allocated to the edge corresponding to the m-node that is not overloaded in the first-level allocation. The present invention defines this process as the second-level allocation; The edge load and upper limit corresponding to the l-node and m-node before the secondary allocation can be expressed as follows: Formula(7) The sum of the parts exceeding the upper limit that need to be allocated to the edges corresponding to the m nodes in the secondary allocation can be expressed as follows: Formula(8) The load change brought by the secondary allocation process to the edge from the exit of node m to node t in the network and the load after the load upper limit of the edge is finally updated can be expressed as follows: Formula(9) This formula shows that the updating process of the secondary allocation is similar to that of the primary allocation, and the load of the edge where overload occurs is calculated with the upper limit; Model parameter settlement Finally, statistics are taken for all nodes in the network, and the final load of the nodes after two-level allocation is calculated, which can be expressed as follows: Official(10).

5. The method according to claim 1, characterized in that For the network built in S3, the underload cascading failure model considering overload limits has the following four important variable parameters: The initial fault node represents the node country where the fault first occurred in the world energy trade network set in the model simulation; Initial load of the node , see formula (1) for details, which represents the total export volume of a country. When the network load exceeds a certain amount, the higher the load, the more likely the network will fail; Node Cap Parameters in , see formula (2) for details. It is generally believed that the network with a higher node upper limit can better cope with attacks; Node lower limit Parameters in , see formula (3) for details. It is generally believed that the network with a higher node lower limit is more fragile and prone to cascading failures due to underload.

6. The method according to claim 1, characterized in that There are two different evaluation scenarios corresponding to the different parameters to be changed in the simulation experiment: Upper limit increase scenario: guarantee the node upper limit Parameters in ≥1, simulating a scenario where each country node in the network actively adjusts and recovers; Scenario of load increase and upper limit reduction: Ensure the node upper limit Parameters in ≤1, after calculating After that, the initial load of the node is greater than the original value, simulating the situation that the nodes of various countries in the network fall into panic and do not actively adjust.

7. The method according to claim 1, characterized in that The performance evaluation criteria are as follows: The robustness of the network can be expressed as follows: Formula (12) TN represents the total number of nodes in the network; in the formula, represents the upper limit parameter in formula (2) When , the number of valid nodes in the network after the cascading failure simulation; The method for determining an effective node NE is that if a node load is greater than or equal to the lower limit of the node load in formula (3) after the settlement of formula (10), there is no underload, and such a node is considered to be an effective node, as follows: Official(13) The elasticity of the network is generally expressed as follows: In the scenario of increasing the upper limit, the recovery capacity of the network is actively adjusted, which can be expressed as follows: a>1 Formula(14) In the scenario where the load increase limit is reduced, the network's resistance to panic can be expressed as follows: 0 < b < 1, Equation (15).

Images