Urban update district full life cycle monitoring and optimizing system based on digital twinning

Abstract

The invention discloses a city update district full life cycle monitoring and optimizing system based on digital twinning, and the system comprises a data collection module which collects physical world data in real time through a plurality of sensors; the digital twinborn model building module is used for receiving the data and building a model containing three-dimensional geographic information, a building information model and crowd behavior simulation; the data analysis module processes original data and feeds back the original data for model updating; the dynamic monitoring module is used for displaying a state and early warning based on a model and analysis data; the optimization decision-making module is used for generating a decision-making scheme through analog simulation and an optimization algorithm; and the system interaction module provides a multi-terminal interface and collects user feedback for scheme adjustment. According to the method, the problems of data islands, monitoring lag and decision-making experience in traditional city updating can be solved, and complete-cycle data integration, dynamic monitoring and intelligent optimization are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of urban planning, in particular to a full-life-cycle monitoring and optimization system for urban renewal areas based on digital twins. Background Art

[0002] With the deepening of the urbanization process, urban renewal has become the core task of enhancing urban functions and improving the living environment. However, traditional urban renewal has obvious shortcomings, especially in the monitoring link: for example, building safety monitoring relies on manual inspections, and environmental monitoring relies on manual fixed-point sampling. These monitoring methods not only have limited coverage, but also have a lag in risk warning for key indicators such as building safety and environmental changes. In addition, the monitoring data of each stage is stored分散ly and the standards are inconsistent, forming an "information island", making it difficult to achieve full-cycle data linkage. Moreover, urban renewal decisions rely more on experience, lacking quantitative simulation of the effects of the plan, which is likely to lead to waste of resources or disconnection from actual needs. Although digital twin technology provides the possibility to solve the above problems, existing applications are mostly limited to a single link (such as only used for building operation and maintenance monitoring), with insufficient model accuracy and real-time performance, making it difficult to reflect the dynamic changes of the physical world; and lacking multi-terminal adaptation ability and low public participation, there is no mature solution covering the full life cycle of urban renewal. Therefore, there is an urgent need to build a digital twin system that can integrate full-cycle data, achieve dynamic monitoring and intelligent optimization. Summary of the Invention

[0003] The purpose of the present invention is to provide a full-life-cycle monitoring and optimization system for urban renewal areas based on digital twins. The present invention can solve the problems of data islands, monitoring lags, and decision-making based on experience in traditional urban renewal, and achieve full-cycle data integration, dynamic monitoring, and intelligent optimization.

[0004] The technical solution of the present invention: A full-life-cycle monitoring and optimization system for urban renewal areas based on digital twins includes: A data acquisition module for real-time collecting physical world data of the urban renewal area through various types of sensors; A digital twin model construction module connected to the data acquisition module for receiving and processing the collected data and constructing a digital twin model including three-dimensional geographic information, building information model, and crowd behavior simulation; A data analysis module respectively connected to the data acquisition module and the digital twin model construction module for cleaning, fusing, and analyzing the collected raw data, and feeding the processed data back to the digital twin model construction module for model update; A dynamic monitoring module connected to the digital twin model construction module and the data analysis module for displaying the status of the urban renewal area based on the digital twin model and real-time analysis data and issuing warnings; The optimization decision-making module, connected to the dynamic monitoring module and the digital twin model construction module, is used to receive monitoring data and generate optimization decision schemes through simulation and optimization algorithms. The system interaction module is connected to the dynamic monitoring module and the optimization decision module respectively. It is used to provide a multi-terminal interactive interface to display monitoring and decision information, and to collect user feedback information and transmit it to the optimization decision module for scheme adjustment.

[0005] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system includes a data acquisition module with sensors such as temperature and humidity sensors, air quality sensors, geomagnetic sensors, and cameras. The data acquisition module possesses self-calibration and adaptive adjustment functions, used to automatically calibrate measurement data according to environmental changes and adaptively adjust the acquisition frequency based on data change frequency. The sensor acquisition frequency is calculated as follows: ; In the formula: The adjusted sampling frequency, The initial sampling frequency, This is the sensitivity coefficient. The standard deviation of the data over a 5-minute period. This is the average of the data over a 5-minute period.

[0006] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system transmits physical world data from the data acquisition module via 5G or NB-IoT networks. Asymmetric encryption algorithms are used to encrypt the transmitted data, specifically: an RSA algorithm is used to generate a public key and a private key. The public key is used to encrypt the raw data, and the private key is used for decryption. The encryption formula is as follows: ,in, For encrypted data, The original data, It is a natural constant. It is the modulus.

[0007] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system includes a digital twin model construction module comprising: The three-dimensional geographic model construction unit uses three-dimensional laser scanning and oblique photogrammetry technology to acquire topographic data; BIM model fusion unit, used to fuse building information model with 3D geographic model; Behavioral model building unit, which simulates human behavior patterns based on machine learning algorithms; The model lightweighting unit is used to automatically adjust the model accuracy based on the performance of the terminal device.

[0008] The aforementioned urban renewal area full life cycle monitoring and optimization system based on digital twins uses a spatiotemporal hash algorithm in its digital twin model construction module to achieve high-precision mapping between physical entities and digital models. The calculation formula is as follows: ; In the formula: For mapping accuracy, For digital model timestamps, For physical entity timestamps, For the spatial coordinates of the digital model, For physical entity spatial coordinates, and These are the time and space maximum ranges, respectively.

[0009] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system includes the following data analysis module: The data preprocessing module is used to denoise and convert the format of the raw data; The data fusion module is used to merge multi-source heterogeneous data into a unified data resource pool; The analysis and mining module employs an improved Isolation Forest algorithm for anomaly detection. This algorithm introduces a time decay factor to dynamically adjust the weights of historical data, as shown in the following formula: , in, As the attenuation factor, For the current moment, for Time-based data weights It is a natural constant.

[0010] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system includes a dynamic monitoring module comprising: The visualization unit is used to visualize the dynamic monitoring results. The monitoring indicator management unit is used to set thresholds for key indicators and trigger early warnings; The multi-level early warning push unit pushes warnings to different levels of management personnel according to the warning level.

[0011] The aforementioned digital twin-based urban renewal area full life cycle monitoring and optimization system includes an optimization decision module comprising: The simulation module is used to simulate the impact of different urban renewal schemes; The multi-objective optimization module uses a three-dimensional objective function for optimization, which is expressed as follows: ; In the formula, For the set of decision variables; For construction cost function; Let resident satisfaction be the function; For environmental impact functions, , and These are the dynamic weights of the corresponding functions; The case matching module uses a cosine similarity algorithm to match historical cases.

[0012] The aforementioned digital twin-based urban renewal area full lifecycle monitoring and optimization system includes the following system interaction module: Multi-terminal interactive interface, supporting PC, mobile and large screen display terminals; The voice interaction unit supports voice recognition and voice broadcasting functions; The feedback processing unit is used to transform user feedback into a basis for model optimization.

[0013] Compared with existing technologies, this invention breaks down the data barriers of traditional urban renewal, achieving effective integration and sharing of data throughout the entire lifecycle, avoiding the phenomenon of "information silos," and providing complete data support for each stage of the work. Through digital twin models and real-time monitoring technology, this invention can promptly grasp the status changes of urban renewal areas, quickly warn of potential risks, and solve the problem of monitoring lag. This invention utilizes simulation and intelligent algorithms to provide quantitative analysis and scientific basis for urban renewal decisions, reducing the subjectivity and reliance on experience in decision-making and improving the rationality of decisions. This invention overcomes the limitations of existing digital twin applications, achieving full coverage of the urban renewal process, and improves system usability and public participation through multi-terminal adaptation and other designs. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a structural diagram of the digital twin model building module; Figure 3 This is a structural diagram of the data analysis module; Figure 4 This is a structural diagram of the dynamic monitoring module; Figure 5 This is a structural diagram of the optimization decision-making module; Figure 6 This is a structural diagram of the system interaction module. Detailed Implementation

[0015] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this should not be construed as limiting the present invention.

[0016] Example: A digital twin-based urban renewal area full lifecycle monitoring and optimization system, such as... Figure 1 As shown, it includes: The data acquisition module is used to collect physical world data of the urban renewal area in real time through various types of sensors; A digital twin model construction module, connected to the data acquisition module, is used to receive and process the acquired data to construct a digital twin model that includes three-dimensional geographic information, building information model, and crowd behavior simulation. The data analysis module is connected to both the data acquisition module and the digital twin model construction module. It is used to clean, fuse, and analyze the acquired raw data, and then feed the processed data back to the digital twin model construction module for model updates. The dynamic monitoring module, connected to the digital twin model construction module and the data analysis module, is used to display the status of the urban renewal area and issue early warnings based on the digital twin model and real-time analysis data. The optimization decision-making module, connected to the dynamic monitoring module and the digital twin model construction module, is used to receive monitoring data and generate optimization decision schemes through simulation and optimization algorithms. The system interaction module is connected to the dynamic monitoring module and the optimization decision module respectively. It is used to provide a multi-terminal interactive interface to display monitoring and decision information, and to collect user feedback information and transmit it to the optimization decision module for scheme adjustment.

[0017] Specifically, the data acquisition module serves as the "sensing entry point" of the digital twin-based urban renewal area full lifecycle monitoring and optimization system. It is responsible for establishing a data connection between the physical world and the digital twin model. Its core function is to collect physical world data of the urban renewal area in real time, accurately, and securely through diverse technological means, providing fundamental support for subsequent model construction, analysis, and decision-making. The module achieves comprehensive perception of the physical world through various types of sensors, covering multi-dimensional data such as environment, traffic, and pedestrian flow, specifically including: Temperature and humidity sensor: Collects air temperature and humidity data within the area to reflect the microclimate environment. Air quality sensor: monitors air quality indicators such as PM2.5, PM10, and harmful gases (such as formaldehyde and CO); Geomagnetic sensors: used for traffic flow monitoring (such as parking space occupancy and vehicle traffic volume), sensing the presence of vehicles through changes in the magnetic field; Cameras: Collect video image data and support crowd density statistics, behavior analysis, building exterior monitoring, etc.

[0018] To ensure data accuracy and acquisition efficiency, the module features self-calibration and adaptive adjustment functions. The self-calibration function automatically adjusts the measurement data based on environmental changes (such as temperature drift and equipment aging), reducing sensor errors and ensuring the accuracy of the raw data. The adaptive acquisition frequency adjustment dynamically adjusts based on data activity, avoiding data redundancy while ensuring information density during critical periods. The acquisition frequency is calculated as follows: ; In the formula: The adjusted sampling frequency, The initial sampling frequency, This is the sensitivity coefficient. The standard deviation of the data over a 5-minute period. This is the average of the data over a 5-minute period.

[0019] The principle behind the formula for calculating the sampling frequency is that when data fluctuates greatly (such as during morning and evening rush hours or sudden environmental changes). Increase the sampling frequency to capture details; when the data is stable (e.g., environmental parameters at night). Reduce costs and save resources.

[0020] The collected physical world data is transmitted via 5G or NB-IoT networks. 5G networks are suitable for high-bandwidth, low-latency scenarios (such as video data); NB-IoT networks are suitable for low-power, wide-coverage scenarios (such as periodic sensor data), meeting the transmission needs of different data types. During transmission, an asymmetric encryption algorithm (RSA) is used to ensure security. The specific process is as follows: A public key and a private key are generated using the RSA algorithm. The public key is used to encrypt the raw collected data, and the private key is used for decryption. The encryption formula is... ,in, For encrypted data, The original data, It is a natural constant. The modulus is used. This mechanism prevents data from being stolen or tampered with during transmission, ensuring data integrity.

[0021] Preferably, such as Figure 2 As shown, the digital twin model construction module is responsible for transforming the physical world data acquired by the data acquisition module into an interactive and analyzable digital twin model, and dynamically updating the model through real-time data feedback, providing a precise digital mapping carrier for subsequent monitoring and decision-making. Its core function is to construct a multi-dimensional model integrating geospatial information, building information, and crowd behavior, while ensuring high-precision synchronization between the physical entity and the digital model. Specific features are as follows: The 3D geographic model building unit uses 3D laser scanning and oblique photogrammetry to acquire topographic data. 3D laser scanning collects dense point clouds of terrain, building facades, etc. by emitting laser beams, and calculates distances using the time difference of laser reflection to generate 3D point cloud data with millimeter-level accuracy. Oblique photogrammetry captures images of the area from different angles using multi-view cameras, and combines aerial triangulation technology to restore surface texture and elevation information. Finally, the point cloud and image data are integrated to generate a 3D geographic model that accurately restores the topography and spatial layout of the area.

[0022] The model fusion unit is used to integrate the building information model (BIM) with the 3D geographic model. The BIM contains detailed data such as the size, material, and equipment parameters of building components. Through a unified spatial coordinate system (such as the urban coordinate system), the coordinates of the BIM model and the 3D geographic model are matched and calibrated. At the same time, the attribute data of the two are associated (such as the building location and geographical terrain, building components and surrounding roads), realizing the integrated digital mapping of macro-geography and micro-building, and solving the problem of separation between traditional geographic models and building models.

[0023] The behavior model building unit simulates crowd behavior patterns based on machine learning algorithms. By collecting data such as crowd density, movement trajectory, and dwell time from cameras and geomagnetic sensors, the behavior model building unit uses clustering algorithms (such as DBSCAN) to discover crowd activity hotspots and time series analysis algorithms (such as LSTM) to predict travel patterns. Finally, the behavior characteristics are transformed into mathematical model parameters (such as crowd movement speed and turning probability), so that the digital twin model can not only restore the physical space, but also simulate the distribution and flow trends of crowds at different times, reflecting the interaction between people and space.

[0024] The lightweight model unit is used to automatically adjust the model's accuracy based on the performance of the terminal device. This unit addresses the computing power and display capabilities of different terminals (such as mobile phones, PCs, and large screens) by simplifying the number of polygons in the model, reducing texture resolution, and retaining core attribute data. This reduces the amount of model data without affecting the display of key information (e.g., retaining building outlines and location information on mobile devices, and displaying complete textures and component details on large screens), ensuring smooth loading and interaction of the model on various terminals.

[0025] Furthermore, to achieve "real-time synchronization" between the digital model and the physical entity, the module uses a spatiotemporal hash algorithm to calculate the mapping accuracy and dynamically correct deviations. The core principle is as follows: By quantifying the differences between the digital model and the physical entity in both time and space, the consistency between the model's state and the physical world is ensured. The calculation formula is: ; In the formula: For mapping accuracy, For digital model timestamps, For physical entity timestamps, For the spatial coordinates of the digital model, For physical entity spatial coordinates, and These are the time and space maximum ranges, respectively.

[0026] In this formula, the time dimension is compared with the timestamps of the digital model ( ) and physical entity data collection timestamp ( The time difference is calculated; the smaller the difference, the higher the time synchronization. The spatial dimension is determined by comparing the spatial coordinates of the digital model (…). ) and the actual spatial coordinates of the physical entity ( ), calculate the spatial deviation; the smaller the deviation, the higher the spatial matching degree; finally, through the time and space maximum range ( , The difference is standardized to obtain the mapping accuracy. (Values ​​range from 0 to 1) When When the data falls below a certain threshold, the model automatically corrects the timestamp and spatial coordinates based on real-time data to ensure the spatiotemporal consistency between the digital model and the physical entity.

[0027] The digital twin model building module maintains the model's dynamism through a data feedback loop. It receives real-time physical data (such as building structure monitoring data and crowd flow data) from the data acquisition module and fused data (such as cleaned environmental data and anomaly detection results) from the data analysis module. It triggers model updates in response to data changes: if building structure data changes (such as renovation construction), the component parameters of the BIM model are updated; if crowd behavior data changes (such as the shift in pedestrian flow due to the addition of a new business district), the prediction parameters of the behavior model are adjusted. The mapping accuracy of the updated model is verified through a spatiotemporal hash algorithm to ensure that the adjusted model still maintains a precise match with the physical world.

[0028] Preferably, such as Figure 3 As shown, the data analysis module is a data processor that connects the raw collected data with the digital twin model and decision-making system. Its core principle is to transform scattered, redundant, and heterogeneous raw data into high-quality, reusable, and effective information, providing data support for digital twin model updates and risk warnings. The data analysis module includes: The data preprocessing module removes outliers from sensor-acquired data (such as outliers caused by equipment malfunctions or fluctuations caused by signal interference) using smoothing filtering algorithms (such as moving averages) to preserve the true trend of the data. It also converts heterogeneous data from different sensors (such as numerical data from temperature and humidity sensors and image data from cameras) into structured data that the system can recognize (such as datasets with numerical values, timestamps, and spatial coordinates), laying the foundation for subsequent fusion analysis.

[0029] The data fusion module is used to integrate multi-source heterogeneous data into a unified data resource pool. In this module, due to the diverse sources of data collected in urban renewal areas (such as temperature and humidity, air quality, pedestrian flow, and traffic), and the differences in format and dimensions (heterogeneous data), this module maps data from different sources to the same spatiotemporal framework through unified data standards (such as timestamp alignment and spatial coordinate association). For example, it can associate "temperature and humidity data," "pedestrian density data," and "air quality data" of a certain area with the same timestamp and spatial coordinates, merging them into a comprehensive "environment-pedestrian flow" dataset for that area. Ultimately, this forms a unified data resource pool covering the entire area, solving the problem of traditional "information silos."

[0030] The analysis and mining module, the core functional unit of the data analysis module, is primarily responsible for extracting valuable information (such as anomalies and potential patterns) from cleaned and fused standardized data, providing data support for dynamic monitoring and risk warning of urban renewal areas. Its core objective is to identify anomalous patterns in the data (such as sudden changes in environmental parameters, equipment failures, and abnormal crowd gatherings) through algorithmic models, and to provide a basis for subsequent decision-making. The core technical feature of this submodule is the use of an improved Isolation Forest algorithm for anomaly detection. Compared to traditional algorithms, its innovation lies in introducing a "time decay factor," making the anomaly detection results more closely reflect the dynamic changes in urban renewal areas (such as the weakening impact of short-term events on long-term data over time), improving the timeliness and accuracy of detection. The Isolation Forest algorithm is a highly efficient anomaly detection algorithm. Its core principle is that anomalous data possesses "scarcity" and "uniqueness" in the feature space, making it easier to isolate quickly (i.e., separating it from other data with fewer partitioning steps). Traditional Isolation Forest constructs multiple isolated trees by randomly partitioning data, calculating the average path length of samples (the difficulty of isolation), and determining whether they are outliers. To address the "temporal correlation" of urban renewal area data (e.g., recent data has higher reference value for the current state), the improved algorithm adds a time decay factor to dynamically adjust the weight of historical data, as shown in the following formula: , in, As the attenuation factor, For the current moment, for Time-based data weights The time decay factor is a natural constant. It assigns higher weight to recent data (which has a greater impact on anomaly detection), while the weight of older data gradually decreases over time (reducing the interference of older data on current judgments). For example, if a certain area experiences a short-term surge in pedestrian traffic (an anomaly) due to a temporary exhibition, the weight of this historical data will decay over time after the exhibition ends, and the algorithm will not treat it as a "normal" occurrence affecting subsequent judgments about pedestrian traffic in that area. Conversely, if an area experiences persistent pedestrian traffic anomalies (such as chronic congestion), the high weight of recent data will reinforce this anomaly, ensuring the algorithm can identify and issue warnings promptly. The improved algorithm better aligns with the dynamic changes in urban renewal areas, avoiding the problem of "old data interfering with current decisions" in traditional algorithms. It improves the detection accuracy of sudden anomalies (such as equipment failure or environmental changes) and persistent anomalies (such as chronic traffic congestion), providing a more reliable basis for risk warnings in the dynamic monitoring module.

[0031] Therefore, by processing the raw data, the data analysis module provides accurate updates for the digital twin model, ensuring consistency between the model and the physical world; on the other hand, it identifies risks in advance through anomaly detection, providing a data-driven basis for dynamic monitoring and optimization decisions.

[0032] Preferably, the dynamic monitoring module integrates a digital twin model with real-time analysis data to transform the dynamic changes of the physical area into visualized information, and uses preset rules to provide timely warnings of abnormal states, offering immediate basis for management decisions. For example... Figure 4 As shown, the dynamic monitoring module includes: The visualization unit uses the 3D geographic information and building information model generated by the digital twin model building module as the basic carrier. It maps the real-time data output by the data analysis module (such as temperature, humidity, air quality, population density, equipment status, etc.) to the corresponding positions in the model according to spatial coordinates and timestamps. Through color marking (such as red indicating exceeding the standard and green indicating normal), dynamic icons (such as population flow arrows indicating movement direction), data dashboards (such as real-time numerical display), etc., the abstract data is transformed into an intuitive visualization scene, realizing an intuitive mapping from "physical area status → dynamic display of digital model", allowing managers to quickly grasp the overall situation of the area.

[0033] The monitoring indicator management unit pre-sets thresholds for core monitoring indicators in the urban renewal area (such as the upper limit of PM2.5 concentration, the safe threshold for pedestrian density, and the operating temperature range of equipment). These thresholds can be dynamically adjusted according to the characteristics of the area (such as residential areas and commercial areas) and management needs. The unit receives processed data from the data analysis module in real time, compares it with the corresponding indicator thresholds, and determines whether the current status is normal (e.g., PM2.5 concentration > 75 μg / m³ is judged as "light pollution"), providing a basis for triggering early warnings.

[0034] The multi-level early warning push unit classifies early warnings into different levels (e.g., general warning, important warning, emergency warning) based on the degree to which monitored indicators exceed thresholds (e.g., slight exceedance, severe exceedance) and the scope of impact (e.g., single-point anomaly, area-wide anomaly). Different levels correspond to preset push rules—for example: general warnings (e.g., a parking space sensor malfunction) are pushed to area maintenance personnel; important warnings (e.g., air quality exceeding standards in a localized area) are pushed to area management personnel; and emergency warnings (e.g., potential safety risks due to crowd congestion) are pushed to emergency command personnel. This level-based classification ensures the accuracy of early warning information transmission, avoids information overload, and guarantees that high-priority risks are addressed promptly.

[0035] The dynamic monitoring module receives real-time updated models (as a visualization carrier) from the digital twin model construction module and processed data (such as real-time status data and anomaly detection results) from the data analysis module. It matches the real-time data with the digital twin model space and displays it through the visualization unit. At the same time, the monitoring indicator management unit compares the data with thresholds, triggers corresponding level warnings, pushes the visualized monitoring results to the system interaction module, and transmits warning information to the optimization decision-making module to provide a basis for subsequent decisions. Simultaneously, the warning information is sent to the corresponding personnel through the multi-level warning push unit.

[0036] Preferably, the optimization decision-making module integrates dynamic monitoring data with digital twin models, combines simulation and intelligent algorithms to generate optimization schemes adapted to the actual needs of the area, and continuously iterates based on historical experience and user feedback, providing quantitative support for decision-making throughout the entire lifecycle of urban renewal. Figure 5 As shown, the optimization decision module includes: The simulation module uses the 3D geographic information, building information, and crowd behavior models generated by the digital twin model building module as a virtual test field. It transforms urban renewal plans to be evaluated (such as traffic flow adjustments, greening renovations, and facility additions) into model-recognizable parameters (such as road width, vegetation coverage, and facility location coordinates). Algorithms simulate the changes in the area's state after the plan's implementation—for example, simulating the impact of adding bus stops on surrounding pedestrian flow, or adjusting building volume ratio on changes in lighting and ventilation. The simulation process incorporates historical data from the data analysis module (such as past pedestrian flow patterns and environmental parameter trends) to ensure the simulation results closely resemble real-world scenarios, achieving a closed loop of "plan pre-setting → digital model simulation → effect prediction," thus avoiding the implementation risks caused by traditional "experience-based decision-making."

[0037] The multi-objective optimization module uses a three-dimensional objective function for optimization, which is expressed as follows: ; In the formula, For the set of decision variables; For construction cost function; Let resident satisfaction be the function; For environmental impact functions, , and These are the dynamic weights of the corresponding functions; for example, during the construction period, they can be 0.6, 0.2, and 0.2 respectively; while during the operation period, they can be 0.2, 0.5, and 0.3 respectively.

[0038] This algorithm iteratively calculates the objective function value under different combinations of decision variables and selects the scheme with the best overall benefits (such as a balanced scheme that is cost-controllable, satisfies residents, and is environmentally friendly), thus solving the problem of single-objective priority and multi-objective conflict in traditional decision-making.

[0039] The case matching module uses the cosine similarity algorithm to match historical urban renewal cases. The principle is to transform the core features of the current area (such as area, population density, and renewal goals) and the problems to be solved (such as traffic congestion and poor environment) into feature vectors, calculate the similarity with the feature vectors of historical cases in the database (the closer the value is to 1, the more similar the cases are), and extract the successful experiences (such as transformation measures and implementation steps) of highly similar cases as a reference for the current plan.

[0040] Therefore, the optimization decision-making module realizes virtual verification of the scheme's effectiveness through a digital twin model, balances multi-dimensional demands through a three-dimensional objective function, and optimizes decision-making accuracy through historical cases and real-time feedback. Ultimately, it transforms urban renewal decision-making from "experience-based judgment" into a scientific process of "quantitative analysis + simulation verification," thereby improving the feasibility and sustainability of the scheme.

[0041] Preferably, the system interaction module provides diverse interaction methods to achieve efficient transmission of monitoring information and decision-making solutions, while collecting user feedback to optimize decision-making, ensuring a precise match between system functions and user needs. The system interaction module includes: The multi-terminal interactive interface is designed to adapt to the hardware characteristics (such as screen size, operation method, and computing power) of different terminal devices (PC, mobile, and large-screen display). For example, the large-screen display adopts an immersive 3D visualization interface to highlight the overall monitoring data and early warning information of the area, adapting to the macro-decision-making scenario of the command center; the mobile terminal adopts a lightweight interface, focusing on key indicators (such as personal surrounding environment and push notifications) and simple operations (such as feedback submission), adapting to the mobile office needs of managers or the public's need for quick query; the PC terminal provides a full-function interface, supporting complex operations such as viewing data details and comparing and analyzing solutions, adapting to refined management scenarios. All terminals are connected to the core module of the system through a unified data interface to ensure that monitoring data and decision-making solutions are synchronized in real time across multiple terminals, avoiding information inconsistency issues. The voice interaction unit uses speech recognition technology to convert user voice commands (such as "Query air quality in XX area" or "Display recent warning records") into text. It then uses semantic parsing algorithms to extract the core requirements of the commands (such as the target area and query content), automatically invokes the corresponding system functions (such as retrieving real-time data for the area or displaying warning history), and uses speech synthesis technology to provide the results to the user in voice format. This unit lowers the operational threshold and is particularly suitable for scenarios where managers can quickly obtain information or for the public to conveniently query information, thus improving interaction efficiency.

[0042] The feedback processing unit collects user feedback information (such as opinions on optimization plans, new monitoring needs, and actual implementation problems) submitted through interactive interfaces (e.g., form submissions, ratings, text feedback) or voice interaction. It uses natural language processing algorithms to transform unstructured feedback (e.g., text descriptions) into structured data (e.g., "the plan cost is too high" corresponding to the "cost optimization need" tag) and associates it with corresponding parameters in the optimization decision module (e.g., adjusting the cost weight α in the multi-objective optimization function). For example, if most residents report "low satisfaction with the greening renovation plan," the feedback processing unit will transmit this information to the optimization decision module, prompting it to increase the weight β of the resident satisfaction function S(X) in subsequent plans, thereby optimizing the plan.

[0043] Therefore, the system interaction module reduces the information access threshold through multi-terminal adaptation and voice interaction, ensuring that different users (managers and the public) can use the system efficiently; through the feedback processing mechanism, it transforms the actual needs of users into the basis for system optimization, upgrading the decision-making scheme from "algorithm generation" to dynamic optimization of "algorithm + human-computer feedback", ultimately improving the practicality of the system and the implementation of the decision-making scheme.

[0044] Furthermore, this embodiment uses a certain old urban renewal area (covering residential areas, commercial streets, and public squares, with a total area of ​​approximately 1.2 square kilometers) as an example to illustrate the specific application process of this system and the functional implementation of each module. This area faces problems such as traffic congestion, lagging environmental monitoring, and a lack of quantitative assessment of the renovation plan. This system enables dynamic monitoring and intelligent optimization throughout the entire life cycle.

[0045] I. Data Acquisition Module Sensor Deployment and Data Acquisition: Deploy multiple types of sensors within the area according to functional zones: Install temperature and humidity sensors in residential areas (initial value of sampling interval). =5 minutes), air quality sensors (monitoring PM2.5 and formaldehyde); geomagnetic sensors (monitoring parking space occupancy) and high-definition cameras (statistics on pedestrian density) are deployed in commercial streets; noise sensors and light sensors are added to public squares.

[0046] Adaptive data collection frequency adjustment: During the morning peak (8:00-10:00), the pedestrian flow data in the commercial street fluctuates greatly, so the collection frequency is automatically increased to once every 4 minutes; at night (0:00-5:00), the data is stable, so the collection frequency is reduced to once every 10 minutes to reduce resource consumption.

[0047] Data transmission and encryption: Camera video data (high bandwidth) is transmitted via 5G network, while periodic data such as temperature, humidity, and geomagnetism are transmitted via NB-IoT network to reduce power consumption. Asymmetric encryption algorithms are used to encrypt the transmitted data. II. Digital Twin Model Construction Module 3D geographic model construction unit: The point cloud data of the building facades in the area is obtained by 3D laser scanning (accuracy ±2mm), and combined with oblique photogrammetry (6 perspectives) to generate a 3D geographic model containing surface texture, restoring the outlines of roads, green spaces and buildings.

[0048] Model fusion unit: Aligns the BIM models (including unit types, pipelines, and building material information) of the eight residential buildings in the residential area with the three-dimensional geographic model (unified to the city coordinate system) to achieve integrated display of geographic space and architectural details.

[0049] Behavioral model building unit: Based on 30 days of pedestrian flow trajectory data collected by cameras, the model is trained by LSTM algorithm to simulate the flow patterns of people on weekdays and weekends (e.g., the peak pedestrian flow in the commercial street reaches 800 people / 100㎡ from 18:00 to 20:00 on weekdays, and the pedestrian flow in the square is concentrated from 10:00 to 16:00 on weekends).

[0050] The digital twin model building module uses a spatiotemporal hash algorithm to calculate the mapping accuracy: when the digital model timestamp With physical entity timestamp The difference is 2 seconds (maximum time range) =30 seconds), digital model spatial coordinates With physical entity coordinates Deviation 0.5 meters (maximum spatial range) =5 meters), and the mapping accuracy is calculated to be 92.9% according to the formula, which meets the real-time synchronization requirements.

[0051] III. Data Analysis Module Data preprocessing module: Abnormal values ​​of air quality sensors (such as "PM2.5=500μg / m³" caused by equipment failure) are removed using the moving average method, while retaining the normal fluctuation trend; the image data from the camera is converted into structured data of "time-space-human density".

[0052] Data fusion module: It associates the "temperature and humidity (25℃, 60%)", "human density (120 people / 100㎡)" and "PM2.5 (65μg / m³)" of the same area with timestamp (2024-12-20 15:00) and spatial coordinates (X=120.1°, Y=30.2°) to form a comprehensive dataset, which is then stored in a unified resource pool.

[0053] Analysis and Mining Module: An improved Isolation Forest algorithm is used for anomaly detection, and a time decay factor is introduced. ,in, The value is set to 0.1. When detecting pedestrian flow data in a commercial street at a certain time, the weight of abnormal data from 3 days ago (such as peak pedestrian flow caused by temporary promotions) decays over time (time decay factor ≈ 0.0007), and its impact on the current detection is negligible; while the weight of continuous abnormal pedestrian flow from 1 hour ago (such as passage blockage) is higher (time decay factor ≈ 0.905), and it is identified by the algorithm as an abnormal event that needs to be warned.

[0054] IV. Dynamic Monitoring Module Visualization Unit: The visualization unit uses a 3D model as a carrier on the large screen, with red heat maps marking densely populated areas of the commercial street (>150 people / 100㎡), green marking areas of good environmental standards in the square (PM2.5 <50μg / m³), and dynamic arrows showing the direction of crowd movement.

[0055] Monitoring indicator management unit: The safe threshold for pedestrian density is 120 people / 100㎡, and the warning threshold for PM2.5 is 75μg / m³. When the pedestrian density in a certain area of ​​the commercial street reaches 160 people / 100㎡ (exceeding the threshold by 33%), a warning judgment is triggered.

[0056] Multi-level early warning push unit: When the pedestrian density in a certain area of ​​the commercial street reaches 160 people / 100㎡, the abnormal pedestrian flow event is judged as an "important warning". The warning is automatically pushed through the multi-level early warning push unit: "Pedestrian congestion in XX section, temporary diversion is recommended" is sent to the area maintenance personnel (mobile terminal); detailed data including 3D positioning is pushed to the management personnel (PC terminal); the warning area is flashed on the large screen to prompt the command center to pay attention.

[0057] V. Optimize the decision-making module Simulation module: Addressing congestion issues, two solutions are simulated: Solution A (widening the sidewalk by 1.5 meters and reducing lanes) and Solution B (adding temporary traffic control barriers to guide and divert traffic). The construction cost of Solution A is simulated using a digital twin model. =800,000 yuan, resident satisfaction =85 points (out of 100), Environmental Impact =10 (Low Impact); Cost of Option B =300,000 yuan, satisfaction level =70 points, Environmental Impact =5.

[0058] The multi-objective optimization module uses a three-dimensional objective function for optimization, taking operational options weight. =0.2, =0.5, =0.3, the scores are calculated as follows: Score of Scheme A = 0.2×80+0.5×85+0.3×10=59.5; Score of Scheme B = 0.2×30+0.5×70+0.3×5=42.5, Scheme A is preferred.

[0059] The case matching module uses the cosine similarity algorithm to match historical cases and found that a similar business district alleviated congestion by "widening the sidewalk and adding rest seats", with a similarity of 0.89. Therefore, "adding 2 rest points" was added to the solution A, and the satisfaction score was improved to 90 points after optimization.

[0060] VI. System Interaction Module A multi-terminal interactive interface displays the simulation effect and 3D model of Solution A on a large screen; administrators receive solution summaries via mobile devices. The voice interaction unit allows residents to view the plans and submit feedback via a WeChat mini-program (lightweight interface). After giving the voice command "Compare the costs of plans A and B", the system automatically replies "Plan A costs 500,000 yuan more than Plan B, but the satisfaction score is 15 points higher".

[0061] The feedback processing unit collected resident feedback that "Scheme A may affect the passage of fire trucks," marked it as a "safety-related requirement," and transmitted it to the optimization decision-making module. The system re-simulated, retaining the fire truck access width (≥4 meters) in Scheme A. The adjusted scheme was approved and implemented.

[0062] As a result, through the application of this system, the old city renewal area has achieved improved data collection efficiency, shortened response time to abnormal events, and reduced the incidence of pedestrian congestion in the commercial street and increased resident satisfaction after the renovation plan was implemented.

[0063] In summary, this invention breaks down the data barriers of traditional urban renewal, achieving effective integration and sharing of data throughout the entire lifecycle, avoiding the phenomenon of "information silos," and providing complete data support for each stage of the work. Through digital twin models and real-time monitoring technology, this invention can promptly grasp the status changes of urban renewal areas, quickly warn of potential risks, and solve the problem of monitoring lag. By leveraging simulation and intelligent algorithms, this invention provides quantitative analysis and scientific basis for urban renewal decisions, reducing subjectivity and reliance on experience in decision-making, and improving the rationality of decisions. This invention overcomes the limitations of existing digital twin applications, achieving full coverage of the urban renewal process, and improves system usability and public participation through multi-terminal adaptation and other designs.

Claims

1. A digital twin-based urban renewal area full lifecycle monitoring and optimization system, characterized by: include: The data acquisition module is used to collect physical world data of the urban renewal area in real time through various types of sensors; A digital twin model construction module, connected to the data acquisition module, is used to receive and process the acquired data to construct a digital twin model that includes three-dimensional geographic information, building information model, and crowd behavior simulation. The data analysis module is connected to both the data acquisition module and the digital twin model construction module. It is used to clean, fuse, and analyze the acquired raw data, and then feed the processed data back to the digital twin model construction module for model updates. The dynamic monitoring module is connected to the digital twin model construction module and the data analysis module, and is used to display the status of the urban renewal area and issue early warnings based on the digital twin model and real-time analysis data; The optimization decision-making module, connected to the dynamic monitoring module and the digital twin model construction module, is used to receive monitoring data and generate optimization decision schemes through simulation and optimization algorithms. The system interaction module is connected to the dynamic monitoring module and the optimization decision module respectively. It is used to provide a multi-terminal interactive interface to display monitoring and decision information, and to collect user feedback information and transmit it to the optimization decision module for scheme adjustment.

2. The urban renewal area full life cycle monitoring and optimization system based on digital twins as described in claim 1, characterized in that: The data acquisition module includes sensors such as a temperature and humidity sensor, an air quality sensor, a geomagnetic sensor, and a camera. The data acquisition module has self-calibration and adaptive adjustment functions, used to automatically calibrate measurement data according to environmental changes and adaptively adjust the acquisition frequency based on the frequency of data changes. The sensor acquisition frequency is calculated as follows: ； In the formula: The adjusted sampling frequency, The initial sampling frequency, This is the sensitivity coefficient. The standard deviation of the data over a 5-minute period. This is the average of the data over a 5-minute period.

3. The urban renewal area full life cycle monitoring and optimization system based on digital twins as described in claim 1, characterized in that: The physical world data acquired by the data acquisition module is transmitted via a 5G or NB-IoT network. An asymmetric encryption algorithm is used to encrypt the transmitted data; specifically, an RSA algorithm is used to generate a public key and a private key. The public key is used to encrypt the raw acquired data, and the private key is used for decryption. The encryption formula is as follows: ,in, For encrypted data, The original data, It is a natural constant. It is the modulus.

4. The urban renewal area full life cycle monitoring and optimization system based on digital twins as described in claim 1, characterized in that: The digital twin model construction module includes: The three-dimensional geographic model construction unit uses three-dimensional laser scanning and oblique photogrammetry technology to acquire topographic data; BIM model fusion unit, used to fuse building information model with 3D geographic model; Behavioral model building unit, which simulates human behavior patterns based on machine learning algorithms; The model lightweighting unit is used to automatically adjust the model accuracy based on the performance of the terminal device.

5. The urban renewal area full life cycle monitoring and optimization system based on digital twins according to claim 4, characterized in that: The digital twin model construction module uses a spatiotemporal hash algorithm to achieve high-precision mapping between physical entities and digital models. The calculation formula is as follows: ； In the formula: For mapping accuracy, For digital model timestamps, For physical entity timestamps, For the spatial coordinates of the digital model, For physical entity spatial coordinates, and These are the time and space maximum ranges, respectively.

6. The urban renewal area full life cycle monitoring and optimization system based on digital twins according to claim 1, characterized in that: The data analysis module includes: The data preprocessing module is used to denoise and convert the format of the raw data; The data fusion module is used to merge multi-source heterogeneous data into a unified data resource pool; The analysis and mining module employs an improved Isolation Forest algorithm for anomaly detection. This algorithm introduces a time decay factor to dynamically adjust the weights of historical data, as shown in the following formula: ， in, As the attenuation factor, For the current moment, for Time-based data weights It is a natural constant.

7. The urban renewal area full life cycle monitoring and optimization system based on digital twins according to claim 1, characterized in that: The dynamic monitoring module includes: The visualization unit is used to visualize the dynamic monitoring results. The monitoring indicator management unit is used to set thresholds for key indicators and trigger early warnings; The multi-level early warning push unit pushes warnings to different levels of management personnel according to the warning level.

8. The urban renewal area full life cycle monitoring and optimization system based on digital twins according to claim 1, characterized in that: The optimization decision module includes: The simulation module is used to simulate the impact of different urban renewal schemes; The multi-objective optimization module uses a three-dimensional objective function for optimization, which is expressed as follows: ； In the formula, For the set of decision variables; For construction cost function; Let resident satisfaction be the function; For environmental impact functions, , and These are the dynamic weights of the corresponding functions; The case matching module uses a cosine similarity algorithm to match historical cases.

9. The urban renewal area full life cycle monitoring and optimization system based on digital twins according to claim 1, characterized in that: The system interaction module includes: Multi-terminal interactive interface, supporting PC, mobile and large screen display terminals; The voice interaction unit supports voice recognition and voice broadcasting functions; The feedback processing unit is used to transform user feedback into a basis for model optimization.

Images