Urban multi-domain collaborative automatic modeling simulation method

Abstract

The invention discloses an urban multi-domain collaborative automatic modeling simulation method, and belongs to the technical field of model fusion. The problem that in the prior art, due to the fact that an urban multi-domain collaborative modeling simulation method is high in modeling complexity, the accuracy of an output result is poor is solved. According to the invention, research scene analysis is carried out; determining an interactive collaboration object, constructing a knowledge graph to perform polling retrieval, performing clustering and causal relationship inference on obtained interactive collaboration elements having a field interaction relationship with modeling field elements and attributes thereof, and identifying to obtain high-dimensional covariables and a relationship between the high-dimensional covariables; modeling domain elements needing to be researched are determined, related variables of the modeling domain elements are selected, and the relation between the modeling domain elements is obtained according to the Pearson's correlation coefficient between the related variables; and constructing a topological model to realize multi-domain collaborative modeling simulation. According to the method, the accuracy of a model result obtained by a cross-domain automatic modeling method is effectively improved, and the method can be applied to simulating the running state of urban multi-domain collaboration.

Description

Technical Field

[0001] The present invention relates to an automated modeling and simulation method, and particularly to an automated modeling and simulation method for urban multi-domain collaboration, belonging to the technical field of model fusion. Background Art

[0002] At present, with the rapid development of urbanization, cities have evolved into a highly complex complex. The urban multi-domain collaborative modeling and simulation technology has emerged in response to this trend. Its core purpose is to integrate various types of information in multiple fields and display the operating state of the urban complex system through simulation, so as to provide a scientific and reliable basis for urban planning, management, and decision-making. In the early stage, urban modeling work mainly focused on a single field, such as separately simulating traffic flow or only analyzing energy consumption. However, with the continuous increase in urban development requirements and the rapid progress of simulation technology, multi-domain collaborative modeling and simulation has gradually become a popular research direction. It breaks the traditional domain boundaries, realizes the deep integration and collaborative analysis of multi-domain data, and can comprehensively and dynamically present the true appearance of the urban system.

[0003] In the prior art, the advantages and disadvantages of the multi-domain collaborative modeling and simulation methods in some cities are as follows: (1) The mathematical modeling analysis method for multi-objective optimization. In the multi-domain collaborative modeling and simulation of cities, the mathematical modeling for multi-objective optimization is a highly valuable analysis method. First, the problem needs to be clearly defined, the boundaries of the urban multi-domain system should be clarified, the system inputs, outputs, and key variables should be determined, and multiple interrelated and possibly conflicting objectives should be identified. Then, decision variables are selected, and different domains have different variable options. Next, the objective function is constructed to quantify each objective with mathematical expressions, and at the same time, the constraint conditions are determined. In the solution stage, traditional algorithms or intelligent algorithms can be selected to obtain non-dominated solutions. Finally, by analyzing the Pareto front, performing sensitivity analysis and scenario analysis, the system performance and change trends can be deeply understood. However, the model construction and solution of the mathematical modeling analysis method for multi-objective optimization are relatively complex, requiring certain mathematical foundations and computing capabilities. For large-scale multi-objective problems, the computational workload is large, the solution time is long, and there may be subjectivity and uncertainty in determining the objective function and constraint conditions. (2) The hybrid modeling analysis method based on multiple rules. In the multi-domain collaborative modeling and simulation of cities, the hybrid modeling based on multiple rules is a comprehensive and in-depth analysis method. First, it is necessary to deeply understand the characteristics and operating mechanisms of each domain in the city, and transform the knowledge, experience, and scientific principles of different domains into corresponding rules. Subsequently, these different types of rules are organically integrated to construct a hybrid model. In the hybrid model, different rules interact with each other to jointly simulate the operating state of the urban multi-domain system. During the analysis process, by inputting different initial conditions and parameter settings, the output results of the model are observed to evaluate the impacts of different strategies and solutions on the urban multi-domain system, thereby providing a scientific and reasonable decision-making basis for urban planning and management. However, the types of rules in the hybrid modeling analysis method based on multiple rules are numerous, the technical complexity is high, the system integration and data docking of different domains have a high technical threshold, and the accuracy of the hybrid model is limited by data quality and the rule assumptions of the hybrid model. (3) The conventional system dynamics modeling analysis method. In the multi-domain collaborative modeling and simulation of cities, using the system dynamics modeling analysis is an effective means. First, it is necessary to sort out the complex causal relationships and feedback mechanisms in each domain of the city, determine the variables in the model, and then draw a flow diagram to clearly show the flow and feedback paths between variables and construct a system dynamics model. After the system dynamics model is constructed, by setting different scenarios, adjusting the initial values and parameters of the variables, observing the dynamic changes of the system over time, and analyzing the change trends of each variable, forward-looking and feasible decision-making suggestions can be provided for the multi-domain collaborative development of cities. However, the modeling complexity of the system dynamics model is high, the whole process relies on manual work, it is difficult to quickly construct a multi-system cross-domain integration model, and it is difficult to ensure the result accuracy with heuristic parameter settings.(4) Large model analysis. In the urban multi-domain collaborative modeling and simulation, using a large model for analysis can provide suggestions with greater depth and breadth for urban development. First, widely collect urban multi-domain data. Then, clean, label, and preprocess the multi-domain data to meet the input requirements of the large model. Use the massive data to train the large model, enabling the large model to learn the complex non-linear relationships between different fields. During actual analysis, input specific urban development scenario parameters. Finally, interpret and verify the results output by the large model, and provide comprehensive and accurate suggestions for urban planners and decision-makers in combination with the actual situation. However, in the large model analysis method, the input data requirements for the large model are high. Training the large model requires a large amount of computing resources and high costs, and the internal structure of the model is complex with poor interpretability.

[0004] In summary, an automated modeling and simulation method for urban multi-domain collaboration is needed. Summary of the Invention

[0005] A brief overview of the present invention is given below to provide a basic understanding of certain aspects of the present invention. It should be understood that this overview is not an exhaustive overview of the present invention. It is not intended to identify the key or important parts of the present invention, nor is it intended to limit the scope of the present invention. Its purpose is merely to present certain concepts in a simplified form as a prelude to the more detailed description to follow.

[0006] In view of this, to solve the problem of poor accuracy of the output results caused by the high modeling complexity in the traditional urban multi-domain collaborative modeling and simulation method in the prior art, the present invention provides an automated modeling and simulation method for urban multi-domain collaboration.

[0007] The technical solution is as follows: The automated modeling and simulation method for urban multi-domain collaboration includes the following steps: S1. Conduct research scenario analysis, that is, based on the retrieval-enhanced generation method of the graph structure, generate a knowledge graph for retrieval and importance calculation, and determine the key elements and their attributes of the research scenario. S2. According to the key elements and their attributes of the research scenario, determine the interactive collaboration objects, construct a knowledge graph for polling retrieval, cluster and infer the causal relationship of the interactive collaboration elements and their attributes that have a domain interaction relationship with the modeling domain elements, and identify the high-dimensional covariates and the relationships between the high-dimensional covariates. S3. Determine the modeling domain elements to be studied, select their relevant variables, and obtain the relationships between the modeling domain elements according to the Pearson correlation coefficients between the relevant variables. S4. According to the identified high-dimensional covariates and the relationships between the high-dimensional covariates, the modeling domain elements, and the relationships between the modeling domain elements, construct a topological model to achieve multi-domain collaborative modeling and simulation.

[0008] Furthermore, in S1, the following steps are included: S11. Determine the research scenario, use its relevant data as input data, and construct a knowledge graph based on the entities and the relationships between entities extracted from the input data; S12. Retrieve the knowledge graph and calculate the importance of the retrieved entities; S13. Determine the key elements, i.e., the necessary nodes, of the research scenario according to the importance of the entities and a preset importance criterion, and output the attributes of the necessary nodes; In S11, collect the data related to the research scenario in the public database, extract the entities related to the research scenario from it, establish the relationships between the entities, and integrate the relationships between the entities to construct a knowledge graph; In S12, use the key elements of the research scenario as a query to retrieve within the knowledge graph, obtain the relevant subgraph, determine the corresponding entities of the subgraph, i.e., the nodes, and calculate the degree centrality, betweenness centrality, and closeness centrality of the nodes; For a node i , its degree centrality is calculated by the formula: , where is the degree of node i , that is, the number of edges connected to node i , n is the total number of nodes in the graph; For a node i , its betweenness centrality is calculated by the formula: , where is the number of shortest paths from node s to node t , is the number of shortest paths from node s to node t and passing through node i ; For a node i , its closeness centrality is calculated by the formula: , where is the shortest path length from node i to node j ; In S13, the attributes of the necessary nodes are determined by the edges connected to the nodes and the relevant text descriptions.

[0009] Furthermore, in S2, the following steps are included: S21. Determine the interactive collaboration object, use its relevant data as input data, and construct a knowledge graph based on the interactive collaboration elements extracted from the input data and the relationships between the interactive collaboration elements; S22. Use the modeling domain elements of the interactive collaboration object as a query, and perform a polling search in the knowledge graph to obtain the interactive collaboration elements and their attributes that have a domain interaction relationship with the modeling domain elements; S23. Cluster the interactive collaboration elements and their attributes that have a domain interaction relationship with the modeling domain elements based on cosine similarity, so that the elements with similar relationships are clustered to form high-dimensional covariates; S24. For the interactive collaboration elements that have a domain interaction relationship with the modeling domain elements and whose attributes have a logical relationship, perform causal relationship inference based on a Bayesian network to determine the logical relationship between the elements, so that the elements with a causal logical relationship are clustered to form high-dimensional covariates; In the above S21, collect the data related to the interactive collaboration object in the public database, extract the interactive collaboration elements related to the interactive collaboration object from it, integrate the interactive collaboration elements and the relationships between the interactive collaboration elements, and construct a knowledge graph; In the above S23, for the attribute vectors of the interactive collaboration elements that have a domain interaction relationship with the modeling domain elements obtained by calculation, that is, the first attribute vector and the second attribute vector , perform numerical processing, and further calculate the similarity of the results after numerical processing to obtain the cosine similarity to complete the clustering. The calculation formula of the cosine similarity is expressed as , where is the element of the first attribute vector, is the element of the second attribute vector; In the above S24, in the Bayesian network, for the interactive collaboration elements that have a domain interaction relationship with the modeling domain elements and whose attributes have a logical relationship, that is, the first variable X and the second variable Y, and the known joint probability distribution of the two is P(X, Y), through the Bayesian result , judge whether there is a causal relationship between the first variable X and the second variable Y. If the Bayesian result P(X|Y) is different from the first variable P(X) and satisfies the causal inference condition, it is considered that the second variable Y has a causal impact on the first variable X.

[0010] Furthermore, in the above S3, determine the modeling domain elements to be studied, that is, the first modeling domain element and the second modeling domain element, select the relevant variables of the first modeling domain element and the second modeling domain element, that is, the first variable X and the second variable Y, and obtain the relationship between the modeling domain elements according to their Pearson correlation coefficient; Pearson correlation coefficient Expressed as: ; where is the Pearson correlation coefficient, is the number of data points, is the th observation of the first variable X, the th observation of the second variable Y, is the mean of the first variable X, is the mean of the second variable Y; The value range of the Pearson correlation coefficient is [-1, 1]. When , it indicates that the two variables are completely positively correlated. When , it indicates that the two variables are completely negatively correlated. When , it indicates that there is no linear correlation between the two variables.

[0011] Furthermore, in step S4, in the process of constructing a topological model based on the modeling domain elements and the relationships between them, for the internal domain, using the principles of topology, the modeling domain elements are taken as nodes and the relationships between the modeling domain elements are taken as edges to construct an internal domain topological model. For cross-domain research scenarios, high-dimensional covariates are taken as nodes and the multiple relationships associated with the high-dimensional covariates are taken as edges to connect the networks of multiple domains to form a cross-domain complex topological structure, and finally a topological model is obtained , , where, for the topological model, is the set of nodes, is the set of edges. For the internal domain topological model, is the set of modeling domain elements, is the set of element relationships. For the cross-domain complex topological structure, is the set of high-dimensional covariates, is the set of multiple relationships between high-dimensional covariates.

[0012] The beneficial effects of the present invention are as follows: the present invention is not implemented by a single model, but a set of comprehensive methods formed by means of scene analysis, establishment of high-dimensional covariates, etc., and finally outputs a set of network topology simulation models, which can be applied to the field of collaborative simulation modeling of urban multi-domain systems; when analyzing key elements in the scene, the present invention no longer relies on tedious manual operations, but is implemented through programs and algorithms, and humans only participate in correcting the output results. The whole process reduces the cost of manual participation and modeling deviations caused by subjective consciousness, and solves the problems of complex fusion model construction and high degree of manual dependence; when determining the research object, the present invention confirms the specific elements included in the modeling research object through scene analysis, limits the boundaries of the research, and prevents excessive conjectures and assumptions from affecting the accuracy of the model. The problem of difficulty in collaborative integration of systems in different fields is solved; the present invention sets up high-dimensional covariates as the main mode of cross-domain collaboration, which can not only comprehensively analyze the various modes of cross-domain collaboration in multiple fields, but also clearly see the specific elements and the degree of mutual influence that occur in cross-domain collaborative feedback, thereby improving the interpretability of the model; the present invention provides a full set of analytical processes: a method for solving a scenario-limited model and determining the elements included in the model, high-dimensional covariate identification to confirm the channels for collaboration in different fields, and the joint establishment of multi-field model evolution models to achieve fusion deduction; the present invention can realize automated modeling, including automated identification of model elements, high-dimensional covariates, and logical interactions (qualitative and quantitative) between elements; the present invention can perform cross-domain collaborative feedback through covariates, that is, connect models in different fields through multi-dimensional associations, and uniformly transmit the mutual influence between field models through covariates. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 Schematic diagram of the process of automated modeling and simulation method for urban multi-domain collaboration. DETAILED DESCRIPTION

[0014] In order to make the technical solutions and advantages of the embodiments of the present invention more clearly understood, the exemplary embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than an exhaustive list of all the embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0015] refer to Figure 1Detailed description of this embodiment, an automated modeling and simulation method for urban multi-domain collaboration, specifically includes the following steps: S1. Conduct research scenario analysis, that is, based on the retrieval-enhanced generation method of graph structure, generate a knowledge graph for retrieval and importance calculation, and determine the key elements and their attributes of the research scenario; S2. According to the key elements and their attributes of the research scenario, determine the interactive collaboration objects, construct a knowledge graph for polling retrieval, cluster and infer the causal relationship of the interactive collaboration elements and their attributes that have a domain interaction relationship with the modeling domain elements, and identify the high-dimensional covariates and the relationships between high-dimensional covariates; S3. Determine the modeling domain elements to be studied, select their relevant variables, and obtain the relationships between the modeling domain elements according to the Pearson correlation coefficients between the relevant variables; S4. According to the identified high-dimensional covariates and the relationships between high-dimensional covariates, the modeling domain elements and the relationships between the modeling domain elements, construct a topological model to achieve multi-domain collaborative modeling and simulation.

[0016] Furthermore, in the above S1, it includes the following steps: S11. Determine the research scenario, use its relevant data as input data, and construct a knowledge graph based on the entities and the relationships between entities extracted from the input data; S12. Retrieve the knowledge graph and calculate the importance of the retrieved entities; S13. According to the importance of the entities and the preset importance criteria, determine the key elements of the research scenario, that is, the necessary nodes, and output the attributes of the necessary nodes; In the above S11, collect the data related to the research scenario in the public database, extract the entities related to the research scenario from it, establish the relationships between entities, integrate the entities and the relationships between entities, and construct a knowledge graph; In the above S12, use the key elements of the research scenario as a query, retrieve within the knowledge graph, obtain the relevant subgraph, determine the subgraph corresponding entities, that is, nodes, and calculate the degree centrality, betweenness centrality and closeness centrality of the nodes; For a node i , its degree centrality is calculated by the formula: , where is the degree of node i , that is, the number of edges connected to node i , n is the total number of nodes in the graph; For a node i , its betweenness centrality is calculated by the formula: , where For Node s To Node t The number of shortest paths, For slave nodes s To Node t And through the node i The number of shortest paths; For a node i , its proximity to centrality The calculation formula is expressed as: ,in, For Node i To the node j The shortest path length; In S13, the attributes of the necessary nodes are determined by the edges connected to the nodes and the related text descriptions.

[0017] Specifically, when the research scenario is determined, the urban field that needs to be analyzed is also confirmed. With the papers, research models, policy documents, and reports in this field as input, the retrieval enhancement generation method based on the graph structure is used to perform a global query on the input data and identify the associated words closely related to the research content as the necessary elements for the scenario operation based on the standards of importance, feasibility, security, and privacy, and simultaneously output the necessary attributes of the elements; The process of determining the research scenario involves the graph-based retrieval enhancement generation method, related word identification and element determination. The specific principles are as follows: (1) The principle of the graph-based retrieval enhancement generation method is to construct a knowledge graph, taking various entities (such as concepts, terms, objects, etc.) in the input data (i.e., papers, research models, policy documents, reports, etc. in public databases) as nodes, and the semantic relationships between entities (such as causal relationships, subordinate relationships, association relationships, etc.) as edges; through the above graph structure, the intrinsic connections between data can be captured more comprehensively, compared with the traditional retrieval method based on text matching. (1) For the search method, the knowledge graph can better understand the semantics and improve the accuracy and relevance of the search; (2) for the identification of related words and the determination of elements, the principle is to measure the importance of nodes (i.e. entities) in the retrieved related subgraphs based on indicators such as the degree of the node (the number of edges connected to the node) and the centrality of the node in the graph (such as betweenness centrality, closeness centrality, etc.). Nodes with higher degrees and stronger centrality are often more critical in the entire knowledge system and are more likely to be necessary elements when the scene is running. For each determined element, its attributes are determined by the type of edges connected to the node and the relevant text description.

[0018] The importance measurement of the node takes into account three indicators: degree centrality, betweenness centrality, and closeness centrality. The specific measurement principle is as follows: (1) Degree centrality reflects the activity level of the node in the local network. The higher the degree, the more extensive the direct connections between the node and other nodes, indicating its potential importance within the local scope. (2) Betweenness centrality measures the bridging role of the node in the entire network. The higher the betweenness centrality of a node, the higher the frequency of its appearance in the shortest paths connecting other nodes, which has an important impact on the connectivity and information transmission of the network. (3) Closeness centrality reflects the average distance between the node and all other nodes. The higher the closeness centrality, the shorter the average distance from the node to other nodes, indicating its advantage in aspects such as information dissemination and also reflecting its importance in the network; In this embodiment, the research scenario is the bus priority development strategy in the urban intelligent transportation system. The specific process of scenario analysis is as follows: In S11, a series of papers, policy documents, and reports on urban transportation, bus priority policies, intelligent transportation technologies, etc. are collected from public databases. For example, a paper mentions that "the intelligent bus dispatching system optimizes the departure frequency by real-time monitoring of the bus vehicle positions and combining passenger flow data". Entities such as "intelligent bus dispatching system", "bus vehicle position monitoring", "passenger flow data", and "departure frequency optimization" are extracted from it, and the relationships between the entities are established. For example, the "intelligent bus dispatching system" utilizes "bus vehicle position monitoring" and "passenger flow data" to achieve "departure frequency optimization". These entities and relationships are constructed into a knowledge graph; In S12, using "key elements of the bus priority development strategy" as the query, a search is conducted in the knowledge graph to find relevant subgraphs. Suppose there are "bus lane nodes", "intelligent bus dispatching system nodes", "bus subsidy policy nodes", etc. in this subgraph, and the degree centrality, betweenness centrality, and closeness centrality of the above nodes are calculated; (1) Degree centrality calculation: Suppose the "bus lane node" is connected to three edges of "traffic congestion mitigation node", "bus operation efficiency improvement node", and "bus priority right-of-way guarantee node", and the total number of nodes n in the graph is 10. Then the degree of the "bus lane node" is 3, and its degree centrality ; (2) Betweenness centrality calculation: Assume that there are 5 shortest paths from the "Traffic congestion mitigation node" to the "Urban sustainable development node", among which 3 pass through the "Bus lane node". And there are 4 shortest paths from the "Improved bus operation efficiency node" to the "Improved citizen travel satisfaction node", among which 2 pass through the "Bus lane node". Also assume there are path situations between other node pairs. By calculating the ratio of the number of shortest paths passing through the "Bus lane node" between all node pairs to the total number of shortest paths and summing them up, the betweenness centrality of the "Bus lane node" is obtained. , After calculation, ; (3) Closeness centrality calculation: Assume that by calculating the length of each shortest path in the knowledge graph and accumulating, the sum of the lengths of the shortest paths from the "Bus lane node" to the other 9 nodes is 15, then the closeness centrality of the "Bus lane node" is obtained. , ; According to the weighted average of degree centrality, betweenness centrality, and closeness centrality, comprehensively calculate the importance score of each node. Assume that the weights of degree centrality, betweenness centrality, and closeness centrality are 0.3, 0.4, and 0.3 respectively, then the importance score of the "Bus lane node" is obtained. , ; In S13, by comparing the importance scores of each node, it is determined that the "Bus lane node", the "Intelligent bus scheduling system node", the "Bus subsidy policy", etc. are essential elements for the scenario operation. Taking the "Bus lane node" in the essential elements as an example, its attributes are determined by the connected edges and relevant text descriptions. For example, the edge from the "Bus lane node" to the "Traffic congestion mitigation node" indicates its attribute of alleviating traffic congestion, and its construction standard attributes (such as width, setting location, etc.) can be known from relevant public databases.

[0019] Further, in S2, the following steps are included: S21. Determine the interaction and collaboration object, use its relevant data as input data, and construct a knowledge graph based on the interaction and collaboration elements extracted from the input data and the relationships between the interaction and collaboration elements. S22. Use the modeling domain elements of the interaction and collaboration object as a query, and poll and retrieve in the knowledge graph to obtain the interaction and collaboration elements that have a domain interaction relationship with the modeling domain elements and their attributes. S23. Perform clustering based on cosine similarity on the interaction and collaboration elements and their attributes that have a domain interaction relationship with the modeling domain elements, so that the elements with similar relationships are clustered to form high-dimensional covariates. S24. For the interaction and collaboration elements that have a logical relationship between attributes and have a domain interaction relationship with the modeling domain elements, perform causal relationship inference based on the Bayesian network to determine the logical relationship between the elements, cluster the elements with causal logical relationships, and form high-dimensional covariates; In S21, collect data related to the interaction and collaboration objects in the public database, extract the interaction and collaboration elements related to the interaction and collaboration objects from it, integrate the relationship between the interaction and collaboration elements and the interaction and collaboration elements, and construct a knowledge graph; In S23, for the attribute vectors of the interaction and collaboration elements that have a domain interaction relationship with the modeling domain elements obtained by calculation, that is, the first attribute vector and the second attribute vector , perform numerical processing, and further calculate the similarity of the results after numerical processing to obtain the cosine similarity to complete clustering. The calculation formula of the cosine similarity is expressed as , where is the element of the first attribute vector, is the element of the second attribute vector. The element vector is composed of various attribute values of the element. By calculating the cosine similarity, the similarity degree of two elements in the attribute space can be measured. The higher the similarity, the more similar the two elements are and the more likely they are to be clustered into the same class; In S24, in the Bayesian network, for the interaction and collaboration elements that have a logical relationship between attributes and have a domain interaction relationship with the modeling domain elements, that is, the first variable X and the second variable Y, the joint probability distribution of the two is known as P(X, Y). Through the Bayesian result , judge whether there is a causal relationship between the first variable X and the second variable Y. If the Bayesian result P(X|Y) has a significant difference from the first variable P(X) and satisfies the causal inference conditions (such as the causal Markov condition, etc.), it is considered that the second variable Y has a causal impact on the first variable X.

[0020] Specifically, in the process of identifying high-dimensional covariates, specifically use papers, related models, etc. involving domain interaction and collaboration in the public database as input, use the retrieval-enhanced generation method based on the graph structure to poll all the elements included in the modeling domain obtained through scene parsing, identify the elements required for domain interaction and refine them to the corresponding attributes; based on the element attributes, cluster similar or logically related elements to form a high-dimensional covariate; The recognition of high-dimensional covariates mainly involves two methods, namely, the retrieval-enhanced generation method based on graph structure and the formation of high-dimensional covariates by factor clustering. Among them, the principle of the retrieval-enhanced generation method based on graph structure in step S2 is basically the same as that in step S1, but the polling of the knowledge graph is involved in the retrieval process. The process principle is to first preprocess the input data, extract the elements and relationships therein, and construct a knowledge graph. Then, for each element in the modeling field determined by scenario parsing, perform a polling search in the knowledge graph. By traversing the nodes and edges, find other elements that have domain interaction relationships with this element, and extract their relevant attributes. For the formation of high-dimensional covariates by factor clustering, the principle is based on the element attributes obtained by retrieval. Using a clustering algorithm, elements with similar attributes or logical relationships between attributes (such as causal relationships, complementary relationships, etc.) are grouped into one category, and each category of element set forms a high-dimensional covariate. The purpose is to integrate the scattered and complex elements to more effectively analyze the collaborative relationships between fields. In this embodiment, it is assumed that the interactive collaborative objects under study are the urban transportation and energy fields. In S21, collect materials such as papers, models, and policy documents on urban transportation and energy. For example, a paper mentions that "the popularization of electric vehicles depends on the improvement of charging pile infrastructure, and at the same time, the use of electric vehicles will affect the distribution of urban traffic flow". Extract interactive collaborative elements such as "electric vehicles", "charging pile infrastructure", and "urban traffic flow distribution" from this sentence, and establish the relationships between the interactive collaborative elements. For example, there is a dependency relationship between "electric vehicles" and "charging pile infrastructure", and an impact relationship between "electric vehicles" and "urban traffic flow distribution", and construct a knowledge graph. In S22, assume that the modeling field elements determined include "transportation tools", "energy supply facilities", "traffic flow indicators", etc. Taking "transportation tools" as an example, perform a polling search in the knowledge graph to find elements that have domain interaction relationships with "transportation tools", such as "electric vehicles", whose attributes include driving range, charging time, energy consumption type, etc., "fuel vehicles", whose attributes include fuel type, fuel consumption per 100 kilometers, etc., and "hybrid vehicles", whose attributes include the attributes of "electric vehicles" and "fuel vehicles". Therefore, the attribute categories of transportation tools are driving range, charging time, energy consumption type, fuel type, fuel consumption per 100 kilometers, and fuel tank capacity. In S23, calculate the attribute vectors of "electric vehicles" and "hybrid vehicles". Assume that the attribute vector of "electric vehicles" is which represents a driving range of 400 kilometers, a charging time of 1 hour, an energy consumption type of electricity, and no fuel-related attributes (missing attributes are filled with zeros). The attribute vector of "hybrid vehicles" is , which represents a cruising range of 500 kilometers, a refueling or charging time of 0.5 hours, a hybrid energy consumption type, a gasoline fuel type, a fuel consumption of 5 liters per 100 kilometers, and a fuel tank capacity of 40 liters. Numerically process the attribute vector (such as encoding the energy consumption type, with electricity being 1, oil being 2, and hybrid being 3), and calculate the cosine similarity of the numerically processed attribute vector; Cosine similarity of the attribute vectors of "electric vehicles" and "hybrid vehicles" after numerical processing It is expressed as:

[0021]

[0022]

[0023] Among them, is an element of "electric vehicles", is an element of "hybrid vehicles"; Due to the high similarity, "electric vehicles" and "hybrid vehicles" can be clustered into the same category as a high-dimensional covariate regarding "new types of transportation tools"; In the aforementioned S24, use a Bayesian network to analyze the relationship between "charging pile infrastructure" and "electric vehicle ownership". Assume that the joint probability distribution is obtained by collecting data , , , ; According to Bayes' formula, calculate the Bayesian result (In actual calculations, the probability value should be between 0 and 1. Here, assume the data is only for illustrating the calculation process) Bayesian result and the probability of a perfect charging pile infrastructure There is a significant difference, indicating a causal relationship between the two. "Charging pile infrastructure" and "electric vehicle ownership" can be clustered together to form a high-dimensional covariate regarding "electric vehicle energy supply and development".

[0024] Furthermore, in the aforementioned S3, determine the modeling domain elements to be studied, namely the first modeling domain element and the second modeling domain element. Select the relevant variables of the first modeling domain element and the second modeling domain element, namely the first variable X and the second variable Y. Based on their Pearson correlation coefficient, obtain the relationship between the modeling domain elements; Pearson correlation coefficient It is expressed as: ; Among them, is the Pearson correlation coefficient, is the number of data points, is the th observation of the first variable X, the th observation of the second variable Y, is the mean of the first variable X, is the mean of the second variable Y; The value range of the Pearson correlation coefficient is [-1, 1]. When it is, it indicates that the two variables are completely positively correlated. When it is, it indicates that the two variables are completely negatively correlated. When it is, it indicates that there is no linear correlation between the two variables.

[0025] Specifically, through scenario analysis and high-dimensional covariate identification, the elements with logical causal relationships can be qualitatively judged. In the process of element relationship identification, by using the correlation analysis method in statistics, the correlation coefficients between different elements can be quantitatively calculated to obtain the linear relationship between the elements; Correlation analysis is a statistical method used to measure the strength and direction of the linear relationship between variables. By collecting a large amount of data related to each element and calculating the correlation coefficient, the degree of association between the elements can be accurately represented; In this embodiment, assume that we study the relationship between urban traffic flow and air quality. Select the daily traffic flow data (taking the number of vehicles passing through the selected area as an index, unit: vehicle) of a certain area in the city over a period of time (such as one month, a total of 30 days) as the first variable X, and the daily air quality index (AQI) as the second variable Y. According to the data related to each element in Table 1, the Pearson correlation coefficient ;

[0026] Table 1 Pearson correlation coefficient The calculation process is expressed as:

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] It can be seen from this that there is a certain positive correlation between traffic flow and air quality index, but the degree of correlation is not very strong. This means that as traffic flow increases, the air quality index has a certain upward trend, but there are also other factors affecting air quality.

[0033] Further, in step S4, in the process of constructing a topological model based on the domain elements of the model and the relationships between them, for the internal domain, using the principles of topology, the domain elements of the model are taken as nodes, and the relationships between the domain elements of the model are taken as edges to construct an internal domain topological model. For the research scenarios across domains, high-dimensional covariates are taken as nodes, and the multiple relationships (similarity relationships and causal logical relationships) associated with the high-dimensional covariates are taken as edges to connect the networks of multiple domains to form a complex topological structure across domains, and finally a topological model is obtained. , , where, for the topological model, is the set of nodes (vertices), is the set of edges. For the internal domain topological model, is the set of domain elements of the model, is the set of element relationships. For the complex topological structure across domains, is the set of high-dimensional covariates, is the set of multiple relationships between high-dimensional covariates.

[0034] Specifically, in step S4, mainly a topological graph is established, and the finally output topological model is also composed of nodes and edges. In this embodiment, taking the fields of urban traffic, energy, and environment as an example, the identified high-dimensional covariates are: "number of electric vehicles in use" (related to the fields of traffic and energy), "charging pile coverage rate" (energy field), "air quality index" (environment field), "proportion of public transportation trips" (traffic field); The relationships between the covariates include: there is a positive correlation between the "number of electric vehicles in use" and the "charging pile coverage rate" (the higher the charging pile coverage rate, the more conducive to the possession and use of electric vehicles); there is a negative correlation between the "number of electric vehicles in use" and the "air quality index" (when the usage of electric vehicles increases, tail gas emissions can be reduced and air quality can be improved); there is a certain substitution relationship between the "proportion of public transportation trips" and the "number of electric vehicles in use" (when the proportion of public transportation trips is high, the demand for electric vehicles may decrease relatively).

[0035] Construct a complex topological structure across domains :

[0036]

[0037] In addition to the above steps, sensitivity analysis, correctness verification, etc. can be performed on the topological model to further verify the performance of the topological model.

[0038] Although the present invention has been described based on a limited number of embodiments, those skilled in the art in this technical field will understand that other embodiments can be conceived within the scope of the present invention thus described. In addition, it should be noted that the language used in this specification is mainly selected for the purpose of readability and teaching, rather than for the purpose of explaining or limiting the subject matter of the present invention. Therefore, many modifications and variations are obvious to those of ordinary skill in the art in this technical field without departing from the scope and spirit of the appended claims. For the scope of the present invention, the disclosure of the present invention is illustrative rather than restrictive, and the scope of the present invention is defined by the appended claims.

Claims

1. An automated modeling and simulation method for urban multi-domain collaboration, characterized in that: The following steps are involved: S1. Conduct research scenario analysis, i.e., generate knowledge graphs for retrieval and importance calculation based on graph structure, and determine the key elements and attributes of the research scenario; S2. According to the key elements and attributes of the research scenario, determine the interactive and collaborative objects, build a knowledge graph for polling retrieval, cluster and infer the causal relationship of the interactive and collaborative elements and their attributes that have domain interaction relationships with the modeling domain elements, and identify high-dimensional covariates and the relationship between high-dimensional covariates; S3. Determine the modeling domain elements that need to be studied, select their related variables, and obtain the relationship between the modeling domain elements based on the Pearson correlation coefficient between the related variables; S4. Based on the identified high-dimensional covariates and the relationships between the high-dimensional covariates, the modeling domain elements and the relationships between the modeling domain elements, a topological model is constructed to realize multi-domain collaborative modeling and simulation.

2. The automated modeling and simulation method for urban multi-domain collaboration according to claim 1 is characterized in that: The S1 comprises the following steps: S11. Determine the research scenario, use its relevant data as input data, and build a knowledge graph based on the entities and relationships between entities extracted from the input data; S12. Search the knowledge graph and calculate the importance of the retrieved entities; S13. According to the importance of the entity and the pre-set importance standard, determine the key elements of the research scenario, i.e., the necessary nodes, and output the attributes of the necessary nodes; In S11, data related to the research scenario is collected from a public database, entities related to the research scenario are extracted, relationships between entities are established, relationships between entities are integrated, and a knowledge graph is constructed; In S12, the key elements of the research scenario are used as queries to search in the knowledge graph to obtain relevant subgraphs, determine the entities corresponding to the subgraphs, i.e., nodes, and calculate the degree centrality, betweenness centrality, and closeness centrality of the nodes; For a node i , its degree centrality The calculation formula is expressed as: ,in, For Node i degree, that is, the degree of the node i The number of connected edges, n is the total number of nodes in the graph; For a node i , its betweenness centrality The calculation formula is expressed as: ,in, For Node s To Node t The number of shortest paths, For slave nodes s To Node t And through the node i The number of shortest paths; For a node i , its proximity to centrality The calculation formula is expressed as: ,in, For Node i To the node j The shortest path length; In S13, the attributes of the necessary nodes are determined by the edges connected to the nodes and the related text descriptions.

3. The automated modeling and simulation method for urban multi-domain collaboration according to claim 2 is characterized in that: The S2 comprises the following steps: S21. Determine the interactive collaboration object, use its related data as input data, and construct a knowledge graph based on the interactive collaboration elements extracted from the input data and the relationships between the interactive collaboration elements; S22. Using the modeling domain elements of the interactive collaborative object as a query, searching in the knowledge graph, and obtaining the interactive collaborative elements and their attributes that have a domain interaction relationship with the modeling domain elements; S23. Clustering the interactive collaborative elements and their attributes that have domain interactive relationships with the modeling domain elements based on cosine similarity, so that the elements with similar relationships are clustered to form high-dimensional covariates; S24. For interactive collaborative elements with logical relationships between attributes and domain interaction relationships between modeling domain elements, causal relationship inference is performed based on the Bayesian network to determine the logical relationship between elements, cluster the elements with causal logical relationships, and form high-dimensional covariates; In S21, data related to the interactive collaborative object is collected from a public database, interactive collaborative elements related to the interactive collaborative object are extracted, and the interactive collaborative elements and the relationships between the interactive collaborative elements are integrated to construct a knowledge graph; In S23, the calculated attribute vector of the interactive collaborative element that has a domain interaction relationship with the modeling domain element, that is, the first attribute vector and the second attribute vector , perform numerical processing, and further calculate the similarity of the results after numerical processing to obtain the cosine similarity to complete the clustering. The calculation formula is expressed as ,in, is the element of the first attribute vector, is the element of the second attribute vector; In S24, in the Bayesian network, for the interactive collaborative elements with logical relationships between attributes and domain interaction relationships between modeling domain elements, that is, the first variable X and the second variable Y, the joint probability distribution of the two is known to be P(X, Y). , determine whether there is a causal relationship between the first variable X and the second variable Y. If the Bayesian result P(X|Y) is different from the first variable P(X) and meets the causal inference conditions, then it is considered that the second variable Y has a causal influence on the first variable X.

4. The automated modeling and simulation method for urban multi-domain collaboration according to claim 3 is characterized in that: In S3, the modeling domain elements to be studied, i.e., the first modeling domain element and the second modeling domain element, are determined, and related variables of the first modeling domain element and the second modeling domain element, i.e., the first variable X and the second variable Y, are selected, and the relationship between the modeling domain elements is obtained according to the Pearson correlation coefficient between the two; Pearson correlation coefficient It is expressed as: ； in, is the Pearson correlation coefficient, is the number of data points, is the first variable X Observations, The second variable Y Observations, is the mean of the first variable X, is the mean of the second variable Y; The value range of Pearson correlation coefficient is [-1, 1]. When , it means that the two variables are completely positively correlated. When , it means that the two variables are completely negatively correlated. , it means there is no linear correlation between the two variables.

5. The automated modeling and simulation method for urban multi-domain collaboration according to claim 4 is characterized in that: In S4, in the process of constructing a topological model based on the modeling domain elements and the relationships between them, for the internal domain, the topological principle is applied to the modeling domain elements as nodes, the relationships between the modeling domain elements as edges, and the internal domain topological model is constructed. For cross-domain research scenarios, high-dimensional covariates are used as nodes, and the multiple relationships associated with high-dimensional covariates are used as edges. The networks of multiple domains are connected to form a complex cross-domain topological structure, and finally a topological model is obtained. , , where, for the topological model, is a collection of nodes, is a collection of edges, for the internal topological model of the domain, is a collection of modeling domain elements. It is a collection of feature relationships, targeting complex topological structures across domains. is a set of high-dimensional covariates, is a collection of multiple relationships between high-dimensional covariates.

Images