Data partition management planning method and system for AI smart city

Abstract

The invention discloses a data partition management planning method and system for an AI smart city, and relates to the technical field of data planning management, and the method comprises the following steps: data classification: classifying initial data of the smart city according to different types and purposes, including traffic data, environmental data, energy data and social governance data; and data partitioning: associating the data based on the data type, respectively calculating the geographical partitioning association degree, the time partitioning association degree and the subject partitioning association degree of multiple sub-data in the same type of data, and performing partitioning management based on the association degree among the data during data partitioning management planning. Through machine learning and an artificial intelligence algorithm, deep analysis and simulation prediction can be performed on city data, scientific decision support is provided for city managers, and city planning and management are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of data planning management, and specifically to a data partition management planning method and system for an AI smart city. Background Art

[0002] By using technologies such as artificial intelligence, Internet of Things, big data, and digital twin, through the construction of a full-element, all-round, and all-time automatic perception of events, the realization of full business, full coverage, and instant discovery. Automatically identify various urban problems, risks, and events such as random stacking of urban garbage, road waterlogging, and unlicensed business operations. With the help of the widely distributed video surveillance resources in the city, build an "AI perception center", construct a smart city brain, comprehensively grasp the urban operation situation, display the urban operation signs in one map, realize visual analysis and decision-making, dispatching and command, etc., and improve the urban public safety management and urban precise governance capabilities.

[0003] Patent Publication No. CN118153925B discloses a digital twin-based smart city data management method and system, which relates to the technical field of digital twin. The method includes: obtaining the urban basic information of the command and dispatch area, and establishing a digital twin model of the command and dispatch area; obtaining the safety event information of the command and dispatch area, and obtaining the basic diffusion model of the safety event according to the static basic information and the safety event information; on the basis of the basic diffusion model, superimposing the influence of the dynamic basic information to obtain the simulation diffusion model of the safety event; according to the simulation diffusion model, display the influence range of the safety event on the digital twin model. It can effectively identify and simulate safety events in urban areas, and visually and dynamically display the influence of safety events based on the simulation results and real situations, which helps residents and managers in the city quickly and accurately understand the development of safety events, and then take reasonable countermeasures.

[0004] Using digital twin technology, construct a digital twin urban technology system to create a virtual mapping of urban physical entities in a digital way, so as to simulate, predict, interact with, and control the whole life cycle process of urban physical entities, in order to realize the multi-dimensional mapping and connection of urban physical space and social space in the digital space. As the "new direction" of the future urban form evolution and the "new foundation" of urban digital transformation.

[0005] For the existing digital twin-based smart city data management method and system, there are still the following deficiencies in the existing smart city data management:

[0006] There are deficiencies in the planning of data partition management, and there are barriers between the data of different partitions, that is, each partition only manages the data of its own partition;

[0007] In a smart city, there is a lack of prediction for the data changes corresponding to facility construction. It can only monitor and manage the facilities in the city in real time, and cannot determine the changes in urban data corresponding to the investment in facility monitoring. Summary of the Invention

[0008] One of the purposes of the present invention is to provide a method and system for data partition management and planning in an AI smart city, which is used for partitioning and managing urban data, establishing the correlation of data in each partition, and breaking down the barriers between data partitions.

[0009] To achieve the above purpose, the present invention is realized through the following technical solutions: A method for data partition management and planning in an AI smart city includes the following steps:

[0010] Data classification: Classify the initial data of the smart city according to different types and uses, including traffic data, environmental data, energy data, and social governance data;

[0011] Data partitioning: Based on the data type, associate the data, calculate the geographical partition correlation degree, time partition correlation degree, and theme partition correlation degree of multiple sub-data in the same type of data respectively. When conducting data partition management and planning, perform partition management based on the correlation degree between the data;

[0012] Data collection and perception: Real-time collect various data of the city through Internet of Things devices, and use digital twin technology to informatize and digitalize the physical entities of the city;

[0013] Data processing and integration: Use data processing and integration technology to clean, transform, and integrate data from different sources and types to ensure data consistency and availability;

[0014] Data analysis and simulation: Apply machine learning and artificial intelligence algorithms to predict the types of changing data corresponding to the construction of urban planning facilities, predict the change amount of the types of changing data corresponding to urban facilities based on the urban facility planning data, and determine the final variables in combination with the geographical partition correlation degree, time partition correlation degree, and theme partition correlation degree;

[0015] Real-time monitoring and feedback: Use holographic display technology to monitor the operation status of the city in real time and provide real-time feedback and countermeasure suggestions.

[0016] In one or more embodiments of the present invention, analyze the data source, determine the format and content of the initial data, mark the initial data, obtain the format and characteristics of the initial data, classify the initial data based on the format, and store the initial data in the storage system respectively based on the initial data type;

[0017] The initial data format is divided into structured, semi-structured, and unstructured;

[0018] Retrieve the initial data formats corresponding to traffic data, environmental data, energy data, and social governance data, and determine the initial data types.

[0019] In one or more embodiments of the present invention, the correlation calculation includes geographical partition correlation calculation, time partition correlation calculation, and theme partition correlation calculation. The calculation steps are as follows:

[0020] Geographical partition correlation calculation:

[0021] Obtain sub-data and perform sub-data cleaning to ensure the integrity and consistency of the sub-data;

[0022] Add geographical location information to each sub-data point and calculate the geographical distance between sub-data points;

[0023] Apply a clustering algorithm to divide geographically close data points into the same partition;

[0024] According to the clustering result, calculate the average distance of sub-data points within each partition as the geographical partition correlation;

[0025] Time partition correlation calculation:

[0026] Obtain the cleaned sub-data;

[0027] Add timestamp information to each sub-data point and calculate the time interval between sub-data points;

[0028] Apply a time series analysis method to divide data points that are close in time into the same partition;

[0029] According to the time series analysis result, calculate the average time interval of data points within each partition as the time partition correlation;

[0030] Theme partition correlation calculation:

[0031] Obtain the cleaned sub-data;

[0032] Analyze the sub-data text, extract theme keywords, and calculate the theme similarity between sub-data points;

[0033] Apply a clustering algorithm to divide sub-data points with similar themes into the same partition;

[0034] According to the clustering result, calculate the average similarity or other statistics of data points within each partition as the theme partition correlation.

[0035] In one or more embodiments of the present invention, the partition correlation calculation formula is as follows:

[0036] The formula for calculating the geographical distance d is:

[0037]

[0038] Among them, d is the geographical distance between two points, r is the radius of the earth, and are the latitudes of the two points is the difference in latitudes between the two points, and Δλ is the difference in longitudes between the two points;

[0039] The calculation formula for the time interval Δt is:

[0040] Δt = t 2 - t 1 ;

[0041] Among them, t1 and t2 are the timestamps of two data points;

[0042] The topic similarity coaine s The calculation formula is:

[0043]

[0044] Among them, A and B are two vectors, A·B represents the dot product of vector A and vector B, and ||A|| and ||B|| represent the magnitudes of vector A and vector B respectively.

[0045] In one or more embodiments of the present invention, when performing data partition management planning, partition management is performed by comprehensively considering the geographical partition correlation, time partition correlation, and topic partition correlation:

[0046] Weight assignment: According to actual needs, weights are assigned to the geographical, time, and topic partition correlations, and the calculation formula for the comprehensive correlation k is as follows:

[0047] k = w g ·d + w t ·Δt + w s ·coaine s ;

[0048] Comprehensive scoring: Calculate the comprehensive score of each sub - data point in the three dimensions;

[0049] Final partition: According to the comprehensive score, the sub - data points are divided into the most suitable partitions.

[0050] In one or more embodiments of the present invention, according to the data types related to urban planning facilities and historical change data, a prediction model is established to predict the future change data types and change amounts:

[0051] The calculation formula for the change amount y is as follows:

[0052] y = β 0 + β 1 x 1 + β2 x 2 +…+β n x n ;

[0053] where βn is the regression coefficient and xn is the influencing factor;

[0054] The predicted value Y of time t t The calculation formula is as follows:

[0055] Y t = c + φ 1 Y t-1 + φ 2 Y t-2 +…+ φ p Y t-p + θ 1 ∈ t-1 + θ 2 ∈ t-2 +…θ q ∈ t-q + ∈ t ;

[0056] where φ and θ are model parameters, ∈ is the error term, and c is the constant term.

[0057] In one or more embodiments of the present invention, the final variable FV of each sub-data point is calculated by combining the comprehensive correlation degree and the output of the prediction model i :

[0058] FV i = α·k + β·y;

[0059] where α and β are weight coefficients.

[0060] In one or more embodiments of the present invention, based on the analysis results and the prediction model, real-time feedback and countermeasure suggestions are provided:

[0061] Automated response: For preset events, automatically execute the preset emergency plan;

[0062] Decision support: Provide data-driven decision support for urban managers;

[0063] Intelligent early warning: For potential risks and problems, give early warnings and provide suggestions on countermeasures.

[0064] This application also provides a data partition management system for the above data analysis management planning method. The partition management system includes:

[0065] A data classification module for classifying the data of the smart city according to different types and uses, including traffic data, environmental data, energy data, and social governance data;

[0066] The partition strategy determination module selects at least one of the strategies of geographical partition, time partition, and theme partition to manage data partitioning according to the characteristics of the data;

[0067] The data collection and storage module uses Internet of Things devices and sensors to collect various data of the city in real time and stores the data in cloud storage or a distributed database;

[0068] The data processing and integration module uses ETL tools to clean, transform, and integrate the data to ensure data consistency and availability;

[0069] The data analysis and simulation module applies machine learning and artificial intelligence algorithms to deeply analyze and simulate the data, including prediction models, clustering algorithms, and time series analysis;

[0070] The real-time monitoring and feedback module uses holographic display and digital twin technologies to monitor the running state of the system in real time and provides real-time feedback and countermeasure suggestions.

[0071] In one or more embodiments of the present invention, the partition management system further includes:

[0072] The system deployment and maintenance module is used to deploy the system into the actual environment and perform continuous maintenance and optimization to ensure the efficient operation and long-term stability of the system;

[0073] The system deployment and maintenance module includes deploying the system on a cloud platform or a local server, regularly monitoring the system performance, performing optimization and upgrading to ensure the efficient operation and long-term stability of the system.

[0074] Through the above technical solutions, the present invention has the following beneficial effects:

[0075] 1. Through machine learning and artificial intelligence algorithms, this application can deeply analyze and simulate and predict urban data, provide scientific decision-making support for urban managers, optimize urban planning and management, and monitor and manage urban infrastructure and public services in real time, such as intelligent transportation, intelligent water services, etc., optimize resource allocation, and improve resource utilization efficiency.

[0076] 2. Through data classification and partition management, it can effectively integrate data from different sources and types, improve data utilization efficiency, and support complex urban management and decision-making.

[0077] 3. Through correlation calculation, data can be effectively partitioned to reduce data redundancy and duplication, and improve the efficiency of data storage and management. The partitioned data can be queried and processed more quickly, improving system performance. Placing highly correlated data in the same partition can reduce data access delays, increase data access speed, and optimize data reading and writing efficiency, especially when processing large-scale data.

[0078] 4. By calculating the comprehensive correlation, highly correlated data can be aggregated in the same partition to improve the accuracy and consistency of data analysis, provide more comprehensive and accurate data support for city managers, and help them make scientific decisions. Reasonable partition management can effectively balance the system load, reduce system bottlenecks, and improve system stability and reliability.

[0079] Other features and advantages of the present invention will be described in the following description, and part of them will become obvious from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a flow chart of the partition management planning method of the present invention;

[0081] Figure 2 It is a schematic diagram of the partition management system of the present invention. DETAILED DESCRIPTION

[0082] The following will disclose multiple embodiments of the present invention with the accompanying drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. In other words, in some embodiments of the present invention, these practical details are not necessary. And if it is possible to implement, the features of different embodiments can be applied interchangeably.

[0083] Unless otherwise defined, all words (including technical and scientific terms) used herein have their usual meanings, which can be understood by those familiar with this field. Furthermore, the definitions of the above words in commonly used dictionaries should be interpreted as the meanings consistent with the relevant fields of the present invention in the content of this specification. Unless otherwise clearly defined, these words will not be interpreted as idealized or overly formal meanings.

[0084] like Figure 1 As shown, the present invention provides a data partition management planning method for an AI smart city, which is used to perform partition management planning on urban data, establish the correlation between the data of each partition, and break down the barriers between data partitions.

[0085] In one embodiment, the zoning management planning method includes the following steps:

[0086] Data classification: Classify the initial data of the smart city according to different types and uses, including traffic data, environmental data, energy data, and social governance data;

[0087] Data zoning: Based on the data type, associate the data, calculate the geographical zoning correlation, time zoning correlation, and theme zoning correlation of multiple sub-data in the same category of data respectively. When carrying out data zoning management planning, perform zoning management based on the correlation between the data;

[0088] Data collection and perception: Real-time collect various data of the city through Internet of Things devices, and use digital twin technology to informatize and digitalize the physical entities of the city;

[0089] Data processing and integration: Use data processing and integration technology to clean, transform, and integrate data from different sources and types to ensure data consistency and availability;

[0090] Data analysis and simulation: Apply machine learning and artificial intelligence algorithms to predict the change data types corresponding to the construction of urban planning facilities, predict the change amount of the change data types corresponding to urban facilities based on urban facility planning data, and determine the final variables in combination with geographical zoning correlation, time zoning correlation, and theme zoning correlation;

[0091] Real-time monitoring and feedback: Use holographic display technology to real-time monitor the operation status of the city and provide real-time feedback and countermeasure suggestions.

[0092] In an implementable manner, through data classification and zoning management, it is possible to effectively integrate data from different sources and types, improve the utilization efficiency of urban data, support complex urban management and decision-making based on the correlation between urban data zones. Since there are time and geographical differences in urban data, calculating the correlation degree of sub-data in different types of data can ensure the differences corresponding to sub-data at different times, locations, and themes.

[0093] Based on multi-dimensions of geography, time, and theme for data zoning management, realize the refined management and optimization of data, improve data access and processing efficiency. For the future planning and construction of the city, the impacts that can be generated will ultimately be reflected in the data. And since different types of data are calculated with different zoning correlations, the data corresponding to the planning and construction can also reflect the expected effects through the correlation degree.

[0094] Furthermore, it provides decisions for urban management, and when carrying out construction, it is also possible to determine the construction time and construction period status of facilities based on time zoning.

[0095] In one embodiment, the data source is analyzed to determine the format and content of the initial data, the initial data is marked, the format and features of the initial data are obtained, the initial data is classified based on the format, and based on the type of the initial data, the initial data is respectively stored in the storage system;

[0096] The initial data format is divided into structured, semi-structured, and unstructured;

[0097] Retrieve the initial data formats corresponding to traffic data, environmental data, energy data, and social governance data, and determine the initial data types.

[0098] In an implementable manner, the initial data is distinguished by various formats. When determining the initial data type, it is only necessary to search for the content of the initial data in the data of the corresponding format, which improves the convenience of data classification. Exemplarily, the initial data formats include:

[0099] Structured data has a fixed format and defined data, and is queried and operated through SQL language. The data has a row and column organization method, and each field has a predefined type.

[0100] Tabular data: such as spreadsheets, database tables.

[0101] Numeric data: such as financial data, sensor readings.

[0102] Well-defined data types: such as dates, times, strings, etc.

[0103] Semi-structured data has a certain structure but does not fully follow the relational database format. Tags or markers are used to organize and identify data elements, which has flexibility and scalability.

[0104] JSON (JavaScript Object Notation): Widely used for API data exchange and web applications.

[0105] XML (eXtensible Markup Language): Commonly used for configuration files, document storage, etc.

[0106] YAML (Yet Another Markup Language): Used for configuration files and data serialization.

[0107] HTML (HyperText Markup Language): Used for web page content representation.

[0108] Unstructured data refers to data without a predefined structure and cannot be stored and queried through traditional relational databases. It includes various formats such as text, images, audio, video, etc.

[0109] Text data: such as emails, reports, documents, social media posts.

[0110] Image data: such as photos, scanned documents, graphics.

[0111] Audio data: such as recordings, music files, podcasts.

[0112] Video data: such as videos, movie files, video streams.

[0113] In one embodiment, the relevance calculation includes geographical partition relevance calculation, time partition relevance calculation, and topic partition relevance calculation. The calculation steps are as follows:

[0114] Geographical partition relevance calculation:

[0115] Obtain sub-data and perform sub-data cleaning to ensure the integrity and consistency of the sub-data;

[0116] Add geographical location information to each sub-data point and calculate the geographical distance between sub-data points;

[0117] Apply a clustering algorithm to group geographically close data points into the same partition;

[0118] According to the clustering results, calculate the average distance of sub-data points within each partition as the geographical partition relevance;

[0119] Time partition relevance calculation:

[0120] Obtain the cleaned sub-data;

[0121] Add timestamp information to each sub-data point and calculate the time interval between sub-data points;

[0122] Apply time series analysis methods to group data points that are close in time into the same partition;

[0123] According to the time series analysis results, calculate the average time interval of data points within each partition as the time partition relevance;

[0124] Topic partition relevance calculation:

[0125] Obtain the cleaned sub-data;

[0126] Analyze the sub-data text, extract topic keywords, and calculate the topic similarity between sub-data points;

[0127] Apply a clustering algorithm to group sub-data points with similar topics into the same partition;

[0128] According to the clustering results, calculate the average similarity or other statistics of the data points within each partition as the relevance of the topic partition.

[0129] In an implementable way, by calculating different sub - data, the geographical partition relevance, time - partition relevance, and topic - analysis relevance of different sub - data can be obtained. Calculate the sub - data of multiple types to perform the relevance of sub - data partitioning.

[0130] Among them, since the geographical locations, times, and topics of multiple sub - data are inconsistent, when calculating the partition relevance and performing data scheduling, multiple relevance partitions corresponding to the sub - data can be scheduled.

[0131] In one embodiment, the partition relevance calculation formula is as follows:

[0132] The calculation formula for the geographical distance d is:

[0133]

[0134] where d is the geographical distance between two points, r is the radius of the earth, and are the latitudes of two points is the difference in latitudes between two points, and Δλ is the difference in longitudes between two points;

[0135] The calculation formula for the time interval Δt is:

[0136] Δt = t 2 - t 1 ;

[0137] where t1 and t2 are the timestamps of two data points;

[0138] The topic similarity coaine s The calculation formula is:

[0139]

[0140] where A and B are two vectors, A·B represents the dot product of vector A and vector B, and ||A|| and ||B|| represent the magnitudes of vector A and vector B respectively.

[0141] In an implementable way, through the relevance calculation, data can be effectively partitioned, reducing data redundancy and duplication, improving the efficiency of data storage and management. Placing data with high relevance in the same partition can reduce the latency of data access and improve the data access speed.

[0142] Among them, partition management enables centralized storage of data of the same type and high correlation, improves the accuracy and reliability of data analysis, helps identify patterns and trends in data, and provides a better foundation for data analysis. Partitioned data can be queried and processed more quickly, improving system performance.

[0143] In addition, since the partition correlation of each sub-data is calculated, multiple sub-data that may be associated can be associated and potential connections can be explored, thereby ensuring that when retrieving sub-data, data with higher correlation can be retrieved, avoiding incomplete data retrieval due to the existence of barriers when retrieving sub-data from different partitions.

[0144] By calculating the geographic distance between data points and grouping geographically close data points into the same partition, the efficiency of geographic data access can be improved to support applications such as smart transportation and emergency response.

[0145] By calculating the time intervals between data points and grouping data points that are close in time into the same partition, time series analysis of historical data can be achieved, supporting applications such as environmental monitoring and trend forecasting.

[0146] By calculating the topic similarity between data points and grouping data points with similar topics into the same partition, the efficiency of retrieval and analysis of topic data can be improved, supporting applications such as social governance and public services.

[0147] In one embodiment, when planning data partition management, the partition management is performed by integrating the geographical partition relevance, the time partition relevance and the subject partition relevance:

[0148] Weight allocation: According to actual needs, weights are allocated to geographical, time and subject partition relevance. The comprehensive relevance k is calculated as follows:

[0149] k=w g ·d+w t ·Δt+w s ·coaine s ;

[0150] Comprehensive score: Calculate the comprehensive score of each sub-data point in three dimensions;

[0151] Final partitioning: Divide the child data points into the most appropriate partitions based on the comprehensive scores.

[0152] In one feasible method, by calculating the comprehensive correlation, data with high correlation can be aggregated in the same partition to improve the speed of data retrieval and query. By comprehensively considering the geographical, temporal and subject correlation, storage and computing resources can be allocated more reasonably, thus avoiding resource waste and improving the utilization of storage space.

[0153] Aggregating highly relevant data together can improve the accuracy and consistency of data analysis. For example, in traffic data analysis, traffic flow data in the same area and during the same time period can be grouped in the same partition to more accurately identify traffic patterns and trends.

[0154] Implementing refined management and partitioning of data contributes to the refined governance of smart cities. For example, in social governance, by combining the theme relevance, event data of similar types can be grouped in the same partition to improve the efficiency of event handling and management.

[0155] Comprehensive relevance calculation can provide more comprehensive and accurate data support for urban managers, helping them make scientific decisions. For example, in urban planning, by combining data analysis of multi-dimensional relevance, urban infrastructure can be more reasonably laid out and resource allocation optimized.

[0156] In one embodiment, according to the data types related to urban planning facilities and historical change data, a prediction model is established to predict future change data types and change amounts:

[0157] The calculation formula for the change amount y is as follows:

[0158] y = β 0 + β 1 x 1 + β 2 x 2 +…+ β n x n ;

[0159] where βn is the regression coefficient and xn is the influencing factor;

[0160] The predicted value Y of time t t The calculation formula is as follows:

[0161] Y t = c + φ 1 Y t-1 + φ 2 Y t-2 +…+ φ p Y t-p + θ 1 ∈ t-1 + θ 2 ∈ t-2 +…+ θ q ∈ t-q + ∈ t ;

[0162] where φ and θ are model parameters, ∈ is the error term, and c is the constant term.

[0163] In one implementable way, by calculating the change amount, analysis and prediction can be carried out based on actual data, providing scientific decision-making support for the planning and construction of urban facilities, avoiding relying on experience and subjective judgment, analyzing long-term trends and periodic changes in urban data, and helping urban planners make more reasonable and long-term layout decisions.

[0164] By predicting future demand changes, resources can be allocated more reasonably to ensure that sufficient facilities are provided at the required time and location, avoiding resource waste and shortages. The prediction of the change amount can identify potential problems and risks in advance, quickly take preventive measures, and improve the response speed and ability of urban management.

[0165] Exemplarily, by predicting the change amount of traffic flow, traffic facilities such as roads and bridges can be reasonably planned and expanded to relieve traffic congestion and improve travel efficiency.

[0166] Predict the change amount of population growth and population density, and plan and construct public facilities such as schools, hospitals, and parks in advance to meet the needs of citizens.

[0167] Predict the change amount of electricity consumption and energy demand, optimize the power grid layout and energy dispatching, improve energy utilization efficiency, and ensure energy supply.

[0168] By predicting the change amount of water demand, optimize the water supply network and water treatment facilities, improve the water supply guarantee ability, and reduce water resource waste.

[0169] In one embodiment, in combination with the comprehensive correlation degree and the prediction model output, calculate the final variable FV of each sub-data point i :

[0170] FV i = α·k + β·y;

[0171] Where α and β are weight coefficients.

[0172] In one implementable way, the calculation of the final variable helps to achieve the refined management and intelligent operation of the smart city, improve the management level and service ability of the city, and enhance the quality of life of citizens.

[0173] Among them, the weight coefficients of α and β are calculated through a weighted regression model.

[0174] In one embodiment, based on the analysis results and the prediction model, provide real-time feedback and countermeasure suggestions:

[0175] Automated response: For preset events, automatically execute the preset emergency plan;

[0176] Decision support: Provide data-driven decision support for urban managers;

[0177] Intelligent early warning: For potential risks and problems, give early warnings and provide suggestions on countermeasures.

[0178] As Figure 2 shown, the present application also provides a data partition management system for the above data analysis and management planning method. The partition management system includes:

[0179] A data classification module, which is used to classify the data of the smart city according to different types and uses, including traffic data, environmental data, energy data, and social governance data;

[0180] A partition strategy determination module, which selects at least one of the strategies of geographical partition, time partition, and theme partition to manage the data according to the characteristics of the data;

[0181] A data collection and storage module, which uses Internet of Things devices and sensors to collect various data of the city in real time and stores the data in a cloud storage or a distributed database;

[0182] A data processing and integration module, which uses ETL tools to clean, transform, and integrate the data to ensure the consistency and availability of the data;

[0183] A data analysis and simulation module, which applies machine learning and artificial intelligence algorithms to deeply analyze and simulate the data, including prediction models, clustering algorithms, and time series analysis;

[0184] A real-time monitoring and feedback module, which uses holographic display and digital twin technologies to monitor the running state of the system in real time and provides real-time feedback and countermeasure suggestions.

[0185] In one embodiment, the partition management system further includes:

[0186] A system deployment and maintenance module, which is used to deploy the system to the actual environment and perform continuous maintenance and optimization to ensure the efficient operation and long-term stability of the system;

[0187] The system deployment and maintenance module includes deploying the system on a cloud platform or a local server, regularly monitoring the system performance, and performing optimization and upgrade to ensure the efficient operation and long-term stability of the system.

[0188] Although the present invention is disclosed in combination with the above embodiments, it is not intended to limit the present invention. Any person skilled in this art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be defined by the appended claims.

Claims

1. A data partition management planning method for an AI smart city, characterized in that: The following steps are involved: Data classification: Classify the initial data of smart cities according to different types and uses, including traffic data, environmental data, energy data and social governance data; Data partitioning: Data is associated based on data type, and the geographic partition association, time partition association, and subject partition association of multiple sub-data in the same type of data are calculated. When planning data partition management, partition management is performed based on the association between data. Data collection and perception: Various data of the city are collected in real time through IoT devices, and the physical entities of the city are informatized and digitized using digital twin technology; Data processing and integration: Use data processing and integration technologies to clean, transform and integrate data from different sources and types to ensure data consistency and availability; Data analysis and simulation: Apply machine learning and artificial intelligence algorithms to predict the type of change in the data corresponding to the construction of urban planning facilities, and predict the change in the type of change in the data corresponding to urban facilities based on urban facility planning data, and determine the final variable by combining the correlation of geographical partitions, time partitions, and subject partitions; Real-time monitoring and feedback: Utilize holographic display technology to monitor the city’s operating status in real time and provide real-time feedback and countermeasures.

2. According to the data partition management planning method of an AI smart city according to claim 1, it is characterized in that: Analyze the data source, determine the format and content of the initial data, mark the initial data, obtain the format and characteristics of the initial data, classify the initial data based on the format, and store the initial data in the storage system based on the initial data type; The initial data formats are divided into structured, semi-structured and unstructured; Retrieve the initial data formats corresponding to traffic data, environmental data, energy data, and social governance data, and determine the initial data type.

3. According to the data partition management planning method of an AI smart city according to claim 2, it is characterized in that: The correlation calculation includes geographical partition correlation calculation, time partition correlation calculation and topic partition correlation calculation. The calculation steps are as follows: Geographical partition correlation calculation: Obtain sub-data and clean it to ensure the integrity and consistency of sub-data; Add geographic location information to each sub-data point and calculate the geographic distance between the sub-data points; Apply clustering algorithms to group geographically close data points into the same partition; According to the clustering results, the average distance of the sub-data points in each partition is calculated as the geographical partition association degree; Time partition correlation calculation: Get the cleaned sub-data; Add timestamp information to each sub-data point and calculate the time interval between sub-data points; Apply time series analysis methods to group data points that are close in time into the same partition; According to the time series analysis results, the average time interval of data points in each partition is calculated as the time partition correlation; Topic partition correlation calculation: Get the cleaned sub-data; Analyze sub-data text, extract topic keywords and calculate topic similarity between sub-data points; Apply clustering algorithms to group sub-data points with similar topics into the same partition; According to the clustering results, the average similarity or other statistics of the data points in each partition are calculated as the topic partition association.

4. According to the data partition management planning method of an AI smart city according to claim 3, it is characterized in that: The partition association calculation formula is as follows: The calculation formula of geographic distance d is: Where d is the geographical distance between two points, r is the radius of the earth, and is the latitude of the two points is the difference between the latitudes of the two points, and Δλ is the difference between the longitudes of the two points; The time interval Δt is calculated as: Δt=t2-t1; Among them, t1 and t2 are the timestamps of two data points; Topic similarity coefficient s The calculation formula is: Among them, A and B are two vectors. A·B represents the dot product of vector A and vector B. ||A|| and ||B|| represent the modulus lengths of vector A and vector B respectively.

5. According to claim 4, a data partition management planning method for an AI smart city is characterized in that: When planning data partition management, partition management is performed based on the geographical partition correlation, time partition correlation, and subject partition correlation: Weight allocation: According to actual needs, weights are allocated to geographical, time and subject partition relevance. The comprehensive relevance k is calculated as follows: k=w g ·d+w t ·Δt+w s ·coaine s ; Comprehensive score: Calculate the comprehensive score of each sub-data point in three dimensions; Final partitioning: Divide the child data points into the most appropriate partitions based on the comprehensive scores.

6. The data partition management planning method for an AI smart city according to claim 5 is characterized in that: Based on the data types and historical change data related to urban planning facilities, a prediction model is established to predict the type and amount of future changes: The calculation formula of the change y is as follows: y=β0+β1x1+β2x2+…+β n x n ; Among them, βn is the regression coefficient and xn is the influencing factor.

7. The data partition management planning method for an AI smart city according to claim 6 is characterized in that: Combine the comprehensive correlation and prediction model output to calculate the final variable FV for each sub-data point i : FV i =α·k+β·y; Among them, α and β are weight coefficients.

8. The data partition management planning method for an AI smart city according to claim 7 is characterized in that: Provide real-time feedback and countermeasures based on analysis results and prediction models: Automated response: Automatically execute preset emergency plans for preset events; Decision support: Provide data-driven decision support for city managers; Intelligent early warning: Provide early warning of potential risks and problems and provide suggestions for response measures.

9. A data partition management system, used in the data partition management planning method according to any one of claims 1 to 8, characterized in that: include: Data classification module, which is used to classify smart city data according to different types and uses, including traffic data, environmental data, energy data, and social governance data; A partition strategy determination module selects at least one strategy among geographic partition, time partition and subject partition to partition the data according to the characteristics of the data; Data collection and storage module, which uses IoT devices and sensors to collect various data of the city in real time and stores the data in cloud storage or distributed databases; Data processing and integration module, using ETL tools to clean, transform and integrate data to ensure data consistency and availability; Data analysis and simulation module, which applies machine learning and artificial intelligence algorithms to conduct in-depth analysis and simulation of data, including prediction models, clustering algorithms, and time series analysis; The real-time monitoring and feedback module uses holographic display and digital twin technology to monitor the system's operating status in real time and provide real-time feedback and countermeasures.

10. A data partition management system according to claim 9, characterized in that: The partition management system also includes: System deployment and maintenance module, used to deploy the system into the actual environment and perform continuous maintenance and optimization to ensure efficient operation and long-term stability of the system; The system deployment and maintenance module includes deploying the system on a cloud platform or local server, regularly monitoring system performance, optimizing and upgrading the system, and ensuring efficient operation and long-term stability of the system.

Images