Urban physical examination improvement scheme optimization method and system based on artificial intelligence

Abstract

The invention discloses a city physical examination improvement scheme optimization method and system based on artificial intelligence, and relates to the technical field of smart cities, and the method comprises the steps: obtaining social media text data and city planning GIS data, and carrying out the time-space alignment processing, and obtaining a structured data set; the structured data set is analyzed, building stress distribution and road bearing capacity are calculated, and a physical health thermodynamic diagram is obtained; and performing sentiment analysis on social media text data by using a BERT model, obtaining a citizen emotion density distribution diagram in combination with a geocoding technology, constructing a physical constraint neural network model, and embedding a physical health thermodynamic diagram into a network loss function to obtain a three-dimensional structure health diagnosis report. According to the method, the physical constraint neural network model is constructed, the building stress distribution and the citizen emotion density are innovatively coupled and analyzed, the PDE loss function is utilized, multi-modal deep fusion of structural health and social emotion is realized, and a high stress-high emotion coupling region can be accurately identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart city, in particular to an improved scheme optimization method and system for city physical examination based on artificial intelligence. BACKGROUND

[0002] At present, the field of smart city has gradually adopted multi-source data fusion technology for city health status evaluation, vibration sensor network is deployed to collect building structure response data, stress distribution is calculated combined with finite element analysis, and a physical health heat map is generated using a spatial interpolation algorithm, some research attempts to introduce social media data to assist decision-making, an emotional analysis model is used to classify the emotions of citizen feedback, and an emotion distribution map is generated based on kernel density estimation, the existing technology has initially realized parallel analysis of physical monitoring data and social perception data, and formed a certain standardized process in data acquisition and visualization.

[0003] The existing technology usually adopts the mode of independent analysis and then superimposed presentation, and fails to establish a dynamic correlation mechanism between the physical state and the emotions of citizens, the physical health heat map only reflects the structural mechanics state, without considering the potential impact of high stress areas on the emotions of citizens, the emotional analysis result lacks interactive feedback with engineering parameters, resulting in difficulty in quantifying the gain effect of emotional improvement on structural reinforcement benefits when optimizing the planning scheme, when high stress concentration and negative emotions gather in a certain area at the same time, the existing technology cannot quantify the coupling effect of the two through a mathematical model, and can only rely on artificial experience to adjust the scheme, which is easy to cause resource allocation deviation. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides an improved scheme optimization method for city physical examination based on artificial intelligence, which solves the suboptimal problem of planning scheme caused by the split analysis of physical state and citizen emotions in the prior art.

[0006] To solve the above technical problems, the present application provides the following technical scheme: In a first aspect, the present application provides an improved scheme optimization method for city physical examination based on artificial intelligence, which comprises obtaining social media text data and city planning GIS data for spatiotemporal alignment processing to obtain a structured data set; analyzing the structured data set to calculate building stress distribution and road bearing capacity to obtain a physical health heat map; using a BERT model to perform emotional analysis on the social media text data, obtaining a citizen emotion density distribution map combined with geocoding technology, constructing a physical constraint neural network model, embedding the physical health heat map into a network loss function, and obtaining a three-dimensional structure health diagnosis report; Based on the three-dimensional structure health diagnosis report, the Cityengine planning module is used to generate a set of feasible reconstruction schemes; The set of feasible reconstruction schemes is imported into the AnyLogic simulation module for social benefit simulation, an economic benefit prediction curve evaluation report is obtained, an emotion propagation partial differential equation is established, negative emotion density is predicted according to a citizen emotion density distribution map, the public space configuration proportion of the set of feasible reconstruction schemes is dynamically adjusted, and an optimized scheme is obtained.

[0007] As a preferred scheme of the city health examination improvement scheme optimization method based on artificial intelligence, wherein: the social media text data and the city planning GIS data are obtained and time-space alignment processing is performed, to obtain a structured data set, including the following steps, The social media text data is obtained through an API interface, the coordinate conversion module is used to unify the space reference of the social media text data, the city planning GIS data is obtained through a municipal geographic information public service module using a WFS service, and the building vibration sensor time series data is combined; The building vibration sensor time series data, the social media text data and the city planning GIS data are time-space aligned using a dynamic time warping algorithm to obtain a structured data set.

[0008] As a preferred scheme of the city health examination improvement scheme optimization method based on artificial intelligence, wherein: the structured data set is analyzed, building stress distribution and road bearing capacity are calculated, and a physical health heat map is obtained, including the following steps, The vibration signal in the structured data set is subjected to Butterworth band-pass filtering to obtain denoised acceleration time series data, the building quality is obtained from the material density and building volume in the city planning GIS data, the dynamic load is calculated, and the filtered dynamic load time series data is obtained; The building GIS contour is extracted from the city planning GIS data, a three-dimensional finite element model is established in ANSYS based on the filtered dynamic load time series data and the building GIS contour, the load sequence in the filtered dynamic load time series data is applied to the floor slab nodes of the three-dimensional finite element model, and the building stress distribution is calculated; According to the vibration data in the city planning GIS data, the road deflection value is calculated; The road deflection value and the building stress distribution are superimposed and rendered in ArcGIS to obtain a physical health heat map.

[0009] As a preferred scheme of the city health examination improvement scheme based on artificial intelligence optimization method, wherein: the BERT model is used for sentiment analysis of social media text data, the emotion density distribution map of citizens is obtained by combining the geographic coding technology, the physical constraint neural network model is constructed, the physical health heat map is embedded into the network loss function, and a three-dimensional structure health diagnosis report is obtained, including the following steps, The BERT model is used for emotion classification of social media text data to obtain labeled data with emotion scores. The city area is extracted in the planning drawing for grid division and building GIS contour alignment, the emotion density value of each grid is calculated, and the emotion density distribution map of citizens is obtained. Based on the labeled data with emotion scores, the emotion and physical health heat map is used for pixel-level spatial alignment to obtain a registered stress-emotion dual-channel image, a physical constraint neural network model is constructed, and the physical constraint neural network model is trained combined with the PDE loss function. The registered stress-emotion dual-channel image is input into the trained physical constraint neural network model to predict the global stress field and mark the high emotion-high stress coupling area, and a three-dimensional structure health diagnosis report is obtained.

[0010] As a preferred scheme of the city health examination improvement scheme based on artificial intelligence optimization method, wherein: based on the three-dimensional structure health diagnosis report, the Cityengine planning module is used to generate a set of feasible reconstruction schemes, including the following steps, Based on the three-dimensional structure health diagnosis report, a double-threshold condition is set for voxel-level traversal screening, and the DBSCAN algorithm is used to merge adjacent high-risk areas to obtain a high-risk area boundary coordinate list. Based on the high-risk area boundary coordinate list and the building GIS contour, a CGA script is written in Cityengine to define dynamic reinforcement logic to form an alternative scheme. The cost, risk score and emotion improvement rate of the alternative scheme are calculated, the NSGA-II algorithm is used to solve the optimal solution, and an optimization scheme is obtained. The optimization scheme is imported into Unity3D to simulate human flow and structure response to generate a set of feasible reconstruction schemes.

[0011] As a preferred scheme of the city health examination improvement scheme based on artificial intelligence optimization method, wherein: the set of feasible reconstruction schemes is imported into the AnyLogic simulation module for social benefit simulation to obtain an evaluation report of economic benefit prediction curve, including the following steps, Convert the building model in the set of feasible reconstruction schemes into GLTF format, the road network into GeoJSON format, build a three-dimensional simulation environment in AnyLogic, set the dynamic behavior logic of citizen agents and vehicle agents, and obtain an initialized simulation model; According to the citizen emotion density distribution map, and configure the agent behavior rules; Load the initialized simulation model and agent behavior rules into the AnyLogic cloud computing module, run Monte Carlo simulation, record the data of pedestrian flow, business income, and public facility usage rate, obtain the original simulation data, calculate the business income growth rate and facility usage Gini coefficient, and obtain the evaluation report of the economic benefit prediction curve.

[0012] As a preferred scheme of the city health check improvement scheme optimization method based on artificial intelligence, wherein: an emotion propagation partial differential equation is established, the negative emotion concentration is predicted according to the citizen emotion density distribution map, the public space configuration proportion of the set of feasible reconstruction schemes is dynamically adjusted, and an optimized scheme is obtained, including the following steps, Based on the citizen emotion density map, a reaction-diffusion partial differential equation is established, and the emotion evolution model is obtained by solving the equation by the finite difference method; The emotion evolution model and the social media text data are used to solve the equation by the finite difference method, and the negative emotion hotspot boundary is obtained; According to the hotspot area in the negative emotion hotspot boundary, the proportion of public space is dynamically increased, and a reconstruction scheme is obtained; The reconstruction scheme is imported into the AnyLogic simulation, the emotion improvement rate is verified, and an optimized scheme is obtained.

[0013] In the second aspect, the application provides a city health check improvement scheme optimization system based on artificial intelligence, which comprises a data acquisition module, social media text data and city planning GIS data are acquired and spatio-temporal alignment processing is performed, and a structured data set is obtained; The data analysis module analyzes the structured data set, calculates the building stress distribution and road bearing capacity, and obtains a physical health heat map; The health diagnosis module uses the BERT model to perform sentiment analysis on the social media text data, combines the geocoding technology to obtain a citizen emotion density distribution map, constructs a physical constraint neural network model, embeds the physical health heat map into a network loss function, and obtains a three-dimensional structure health diagnosis report; The scheme reconstruction module uses the Cityengine planning module based on the three-dimensional structure health diagnosis report to generate a set of feasible reconstruction schemes; An optimization module imports the set of feasible reconstruction schemes into the AnyLogic simulation module to simulate social benefits, obtain an evaluation report of the economic benefit prediction curve, establish an emotion propagation partial differential equation, predict the negative emotion density according to the emotion density distribution map, dynamically adjust the public space configuration proportion of the set of feasible reconstruction schemes, and obtain an optimized scheme.

[0014] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the method for optimizing an improvement scheme for urban health examination based on artificial intelligence according to the first aspect of the present application.

[0015] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the method for optimizing an improvement scheme for urban health examination based on artificial intelligence according to the first aspect of the present application.

[0016] The present application has the following beneficial effects: the present application couples and analyzes building stress distribution and citizen emotion density by constructing a physically constrained neural network model, realizes multi-modal deep fusion of structure health and social emotion by using a PDE loss function, can accurately identify high stress-high emotion coupling areas, and simultaneously establishes a dynamic feedback mechanism based on an emotion propagation partial differential equation to convert the negative emotion density prediction value into a public space configuration optimization scheme, thereby forming a closed loop from data perception to scheme optimization. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0018] Fig. 1 The flowchart of the method for optimizing an improvement scheme for urban health examination based on artificial intelligence.

[0019] Fig. 2 The schematic diagram of the system for optimizing an improvement scheme for urban health examination based on artificial intelligence.

[0020] Fig. 3 The schematic diagram of data acquisition and alignment.

[0021] Fig. 4 The schematic diagram of the set of feasible reconstruction schemes. DETAILED DESCRIPTION

[0022] In order to make the above objectives, characteristics and advantages of the present application more apparent, a detailed description of the specific embodiments of the present application will be given below with reference to the accompanying drawings.

[0023] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. The present application, however, can be practiced in a variety of ways other than those specifically described herein, and the scope of the present application is not limited to the specific embodiments described herein.

[0024] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments.

[0025] Reference Figs. 1-4 For one embodiment of the present application, the embodiment provides an artificial intelligence-based urban health check improvement scheme optimization method, comprising the following steps: S1, obtaining social media text data and urban planning GIS data for spatio-temporal alignment processing to obtain a structured data set.

[0026] S1.1, obtain social media text data through an API interface, use a coordinate conversion module to unify the spatial reference of the social media text data, use a municipal geographic information public service module to obtain urban planning GIS data using WFS service, and combine building vibration sensor time series data.

[0027] Further, the social media text data containing geographic location information is obtained by calling the API interface provided by the social media platform, and the geographic coordinates in the social media text data are uniformly converted to the WGS84 coordinate system using a coordinate conversion module; the urban planning GIS data containing building outlines and road networks is obtained through the WFS service interface provided by the municipal geographic information public service platform, and the acceleration time series data recorded by the vibration sensors deployed at the key nodes of the building are synchronously collected.

[0028] S1.2, use a dynamic time warping algorithm to perform spatio-temporal alignment on the building vibration sensor time series data, the social media text data and the urban planning GIS data to obtain a structured data set.

[0029] Further, the acceleration sampling time sequence in the building vibration sensor time sequence data is extracted, the timestamp of the social media text data is obtained, and the version update time of the urban planning GIS data is obtained; the dynamic time warping algorithm is used to calculate the time offset of the building vibration sensor time sequence data and the social media text data, and the time offset of the building vibration sensor time sequence data and the urban planning GIS data, the time reference of the social media text data and the urban planning GIS data is compensated and corrected according to the calculated time offset, and the physical coordinates of the building vibration sensor installation position, the geographic coordinates in the social media text data and the spatial reference system in the urban planning GIS data are unified to the WGS84 coordinate system, and a structured data set containing time alignment and unified spatial reference is generated.

[0030] S2, analyzing the structured data set, calculating the building stress distribution and the road carrying capacity, and obtaining the physical health heat map.

[0031] S2.1, the vibration signal in the structured data set is subjected to Butterworth band-pass filtering to obtain denoised acceleration time sequence data, the building quality is obtained through the material density and the building volume in the urban planning GIS data, the dynamic load is calculated, and the filtered dynamic load time sequence data is obtained.

[0032] Specifically, the expression is, ; Among them, is the dynamic load of the building structure at time , is the building quality, is the second-order differential of displacement, is the square of time differentiation, is the displacement, is the time.

[0033] It should be noted that the original vibration signal of the building vibration sensor time sequence data record in the structured data set is subjected to frequency domain filtering processing by using the Butterworth band-pass filter, the environmental noise below the building natural frequency and the high-frequency interference above the sampling frequency are filtered out, the effective frequency band signal reflecting the structure dynamic characteristics is retained, and the denoised acceleration time sequence data is obtained; according to the building material density parameters and the building volume information recorded in the urban planning GIS data, the building quality is calculated by the product of volume and density, based on the denoised acceleration time sequence data and the building quality, the dynamic load containing time sequence is finally output according to Newton's second law, wherein the acceleration value takes the second-order differential result of the acceleration time sequence data.

[0034] S2.2, extracting the building GIS contour in the urban planning GIS data, establishing a three-dimensional finite element model in ANSYS based on the filtered dynamic load time series data and the building GIS contour, applying the load sequence in the filtered dynamic load time series data to the three-dimensional finite element model floor node, and calculating the building stress distribution.

[0035] Specifically, the expression is, ; Wherein, is the building stress distribution, is the normal stress of the building structure in the direction, is the normal stress of the building structure in the direction, is the shear stress of the building structure in the direction and direction plane.

[0036] S2.3, according to the vibration data in the urban planning GIS data, inverse calculation of road deflection value.

[0037] Specifically, the expression is, ; Wherein, is the road deflection value, is the dynamic load, is the equivalent radius, is the elastic modulus, is the Poisson's ratio.

[0038] S2.4, superimpose and render the road deflection value and the building stress distribution in ArcGIS to obtain the physical health heat map.

[0039] Further, a spatial layer containing road network and building contour is created in the ArcGIS platform, the calculated road deflection value is spatially interpolated according to the road centerline position to generate a continuous road health degree raster surface; at the same time, the building stress distribution calculation result is spatially associated with the building GIS contour to generate a building stress intensity surface layer, and a multi-band rendering technology is used to spatially superimpose the road health degree raster surface and the building stress intensity surface layer, wherein the road deflection value is represented by a green to red gradient color scale to represent the health degree, and the building stress distribution is represented by a blue to yellow gradient color scale to represent the stress level, the transparency parameter is set to maintain the visibility of the two layers when superimposed and displayed, and finally the physical health heat map reflecting the road bearing state and the building structure stress is generated.

[0040] S3, sentiment analysis of social media text data using BERT model, combined with geocoding technology to obtain citizen emotion density distribution map, construct physically constrained neural network model, embed physical health heat map into network loss function, obtain three-dimensional structure health diagnosis report.

[0041] S3.1, sentiment classification of social media text data using BERT model, obtain labeled data with emotion score.

[0042] Further, the pre-trained BERT model is used for emotion classification processing of social media text data contained in the structured data set. Each piece of social media text data is input into the BERT model to obtain a text feature vector, and then the feature vector is mapped to an emotion classification label space through a fully connected layer, and the corresponding emotion category and confidence score of each piece of social media text data are output. Emotion categories are divided into positive, neutral and negative three categories, and confidence scores range from 0 to 1, indicating emotion intensity. The classification results are associated with the geographical position information and time stamp of the original social media text data to form labeled data containing emotion labels, emotion scores, geographical coordinates and time stamps. The labeled data retains the same space-time reference as the structured data set.

[0043] S3.2, extract city area in planning drawing for grid division and building GIS contour alignment, calculate emotion density value of each grid, obtain citizen emotion density distribution map.

[0044] Specifically, the expression is, ; Wherein, Emotion density value, is the sentiment analysis result of the th text, is the bandwidth, is the distance from the th text to the target point, is the total number of texts, is the text index.

[0045] S3.3, based on emotion and physical health heat map with emotion score labeled data, pixel-level spatial alignment is performed to obtain a registered stress-emotion dual-channel image, and a physically constrained neural network model is constructed. Train the physically constrained neural network model combined with the PDE loss function.

[0046] Specifically, the expression is, ; Wherein, PDE loss function, is the three-dimensional stress field predicted by the neural network, stress value measured by the optical fiber sensor, Laplacian of the stress field, material stiffness coefficient, emotion weight matrix, PDE constraint weight, emotion weight constraint strength.

[0047] S3.4, input the registered stress-emotion dual-channel image into the trained physical constraint neural network model, predict the global stress field and mark the high emotion-high stress coupling area, and obtain a three-dimensional structure health diagnosis report.

[0048] Further, the stress-emotion dual-channel image that has completed pixel-level spatial alignment is input as input data into the trained physical constraint neural network model for processing; the physical constraint neural network model first extracts stress distribution features and emotion density features through a convolutional neural network, then uses a loss function containing PDE constraints to calculate layers to analyze the features, and outputs a global three-dimensional stress field prediction result; at the same time, based on the product of the stress field prediction value and the emotion density value, a coupling coefficient is calculated, high emotion-high stress coupling areas with coupling coefficients exceeding a set threshold are automatically identified and marked, and a three-dimensional structure health diagnosis report containing a three-dimensional stress field cloud map, coupling area spatial coordinates and coupling intensity values is generated.

[0049] S4, based on the three-dimensional structure health diagnosis report, use the Cityengine planning module to generate a set of feasible reconstruction schemes.

[0050] S4.1, based on the three-dimensional structure health diagnosis report, set a double-threshold condition for voxel-level traversal screening, and use the DBSCAN algorithm to merge adjacent high-risk areas to obtain a high-risk area boundary coordinate list.

[0051] Further, based on the stress field values and emotion coupling coefficients recorded in the three-dimensional structure health diagnosis report, two screening conditions of stress threshold and emotion coupling threshold are set; each voxel unit in the diagnosis report is detected, and when the stress field value of the voxel unit exceeds the stress threshold and the emotion coupling coefficient exceeds the emotion coupling threshold, the voxel is marked as a potential high-risk unit; the spatial coordinates of all marked potential high-risk units are input, and a DBSCAN spatial clustering algorithm based on density is applied for processing; the DBSCAN algorithm identifies unit clusters with spatial density by calculating the Euclidean distance between units, and merges unit clusters with a mutual distance less than a neighborhood radius and containing a minimum number of units into continuous high-risk areas; finally, a list containing three-dimensional boundary coordinates of each high-risk area is output.

[0052] S4.2, based on the high-risk area boundary coordinate list and building GIS contour, write CGA script in CityEngine to define dynamic reinforcement logic and form the alternative scheme.

[0053] Further, in the CityEngine three-dimensional modeling environment, import the high-risk area boundary coordinate list and building GIS contour data, write CGA rule script according to the building structure type and risk level. The CGA script first analyzes the spatial position relationship between the high-risk area boundary coordinates and the building GIS contour, and defines differentiated reinforcement rules for different risk level areas: for frame structure buildings, set the generation logic of adding shear walls or steel supports, and for masonry structure buildings, set the generation rule of adding ring beam columns; at the same time, adjust the reinforcement member size parameters according to the stress distribution gradient in the three-dimensional structure health diagnosis report, and automatically increase the reinforcement section in the building stress concentration area; after executing the CGA script, a set of three-dimensional models containing multiple reinforcement schemes is generated, each model is attached with engineering parameters such as material consumption and construction range, forming a complete alternative scheme.

[0054] S4.3, calculate the cost, risk score and emotion improvement rate of the alternative scheme, use NSGA-II algorithm to solve the optimal solution, and get the optimization scheme.

[0055] Specifically, the expression is, ; Wherein, is the optimization scheme, is the reinforcement volume, is the unit price of materials; S4.4, import the optimization scheme into Unity3D to simulate human flow and structure response, and generate a set of feasible reconstruction schemes.

[0056] Specifically, the expression is, ; Wherein, is the set of feasible reconstruction schemes, is the original acceleration, is the acceleration after reconstruction.

[0057] S5, import the set of feasible reconstruction schemes into the AnyLogic simulation module to simulate social benefits and get the economic benefit prediction curve evaluation report.

[0058] S5.1, convert the building model in the set of feasible reconstruction schemes to GLTF format and the road network to GeoJSON format, build a three-dimensional simulation environment in AnyLogic, set the dynamic behavior logic of citizen agents and vehicle agents, and get the initialized simulation model.

[0059] Further, the three-dimensional building model containing the feasibility reconstruction scheme is exported to a GLTF format file through a format conversion tool, and the road network data is converted to a GeoJSON format; a three-dimensional scene is newly created in an AnyLogic simulation platform, the GLTF format building model and the GeoJSON format road network data are imported to construct a basic environment; the initial spatial distribution of the citizen agent is set based on the citizen emotion density distribution map, and the state machine logic containing the behaviors such as commuting, leisure and shopping is configured for the citizen agent, wherein the behavior selection probability is associated with the emotion score; the path planning algorithm based on the road network and the car following model parameters are defined for the vehicle agent, the interaction rules of the citizen agent and the vehicle agent are realized through the agent programming interface provided by AnyLogic, including the pedestrian crossing behavior and the bus riding decision, and the initialization simulation model containing the three-dimensional scene, the agent behavior logic and the interaction rules is generated.

[0060] S5.2, according to the citizen emotion density distribution map, and configure the behavior rules of the agent.

[0061] Further, in the AnyLogic simulation module, the spatial grid data of the citizen emotion density distribution map is matched with the initial position of the agent, each citizen agent is given a corresponding emotion score attribute, the behavior rules of the citizen agent are set based on the emotion score, the emotion density distribution percentile of the high emotion-high stress coupling area marked in the three-dimensional structure health diagnosis report is used to determine the emotion score setting threshold, the citizen agent with an emotion score lower than the emotion score setting threshold increases the probability of going to the park leisure place, the citizen agent with an emotion score higher than the emotion score setting threshold increases the commuting route change frequency and the commercial facility stay time, and the emotion propagation mechanism between the citizen agents is configured, when two citizen agents meet within a set distance range, the emotion value interaction is carried out according to the parameters of the emotion propagation partial differential equation, the speed adjustment rules of the vehicle agent based on the road network state are set, the emotion density threshold of the road is set based on the 85 percentile of the road network area emotion value in the citizen emotion density distribution map, and the driving speed is automatically reduced when the emotion density of the road exceeds the emotion density threshold of the road.

[0062] S5.3, the initialization simulation model and the behavior rules of the agent are loaded to the AnyLogic cloud computing module, the Monte Carlo simulation is run, the data of the passenger flow, the commercial revenue and the public facility usage rate are recorded, the original simulation data is obtained, the commercial revenue growth rate and the facility usage Gini coefficient are calculated, and the evaluation report of the economic benefit prediction curve is obtained.

[0063] Specifically, the expression is, ; wherein, is the commercial revenue growth rate, The original income, The income after the transformation.

[0064] ; Wherein, The facility use Gini coefficient, The number of facilities, The number of facilities in the simulation cycle, The number of facilities in the simulation cycle, The facility index.

[0065] It should be noted that the initialization simulation model containing three-dimensional scene and agent behavior logic is uploaded to the AnyLogic cloud computing module, the Monte Carlo simulation parameters are configured and multiple rounds of simulation are started, the real-time recording of the flow data of the citizen agent in the commercial area, the income change data of the commercial facility and the use data of the public facilities such as the park in the simulation process is completed, the simulation results of each round are summarized, the commercial income growth rate before and after the implementation of the transformation scheme is calculated, the facility use Gini coefficient is calculated based on the public facility use data, the data of the commercial income growth rate and the facility use Gini coefficient changing with time is visualized, and the evaluation report reflecting the dynamic changes of economic benefit and social benefit is generated.

[0066] S6, establish an emotion propagation partial differential equation, predict the negative emotion concentration according to the citizen emotion density distribution map, dynamically adjust the public space configuration proportion of the set of feasible transformation schemes, and obtain the optimization scheme.

[0067] S6.1, based on the citizen emotion density map, a reaction-diffusion partial differential equation is established, and after solving by finite difference method, an emotion evolution model is obtained.

[0068] Specifically, the expression is, ; Wherein, The differential change of emotion concentration, The differential change of time, The diffusion coefficient, and the social public opinion weight, The social public opinion, The emotion concentration.

[0069] S6.2, utilize the emotion evolution model and the social media text data, solve the equation by finite difference method, and obtain the negative emotion hotspot boundary.

[0070] Further, the emotion evolution model is matched and calibrated with the emotion classification results of social media text data, and the emotion evolution equation is discretely solved on the city space grid using the explicit finite difference method, wherein the spatial derivative is discretized using the five-point difference format, and the time derivative is discretized using the forward Euler format; the sentiment analysis results of the social media text data are input as the emotion source term into the finite difference method solving equation, and the emotion concentration distribution at multiple time periods in the future is obtained through iterative calculation, the contour lines of the spatial distribution of the emotion concentration are extracted, the continuous area boundary with an emotion concentration exceeding a set critical value is identified, and the negative emotion hotspot boundary data containing the geometric boundary of the hotspot area and the peak emotion value are output.

[0071] S6.3. According to the hotspot area in the negative emotion hotspot boundary, the proportion of public space is dynamically increased to obtain a reconstruction scheme.

[0072] Specifically, the expression is, ; wherein, is the adjusted proportion of public space, is the original proportion of public space, is the intensity of controlling the influence of the emotion hotspot on the configuration of public space, is the total area of the reconstruction area, is the area of the negative emotion hotspot.

[0073] S6.4. The reconstruction scheme is introduced into the AnyLogic simulation, and the emotion improvement rate is verified to obtain an optimized scheme.

[0074] Specifically, the expression is, ; wherein, is the emotion improvement rate, is the negative emotion concentration before reconstruction, is the negative emotion concentration after reconstruction.

[0075] The embodiment also provides an urban health check improvement scheme optimization system based on artificial intelligence, which comprises: an acquisition data module, which acquires social media text data and city planning GIS data for spatiotemporal alignment processing to obtain a structured data set; a data analysis module, which analyzes the structured data set, calculates building stress distribution and road bearing capacity, and obtains a physical health heat map; a health diagnosis module, which uses a BERT model to perform sentiment analysis on the social media text data, combines geographic coding technology to obtain a citizen emotion density distribution map, constructs a physical constraint neural network model, embeds the physical health heat map into a network loss function, and obtains a three-dimensional structure health diagnosis report; A scheme reconstruction module generates a set of feasible reconstruction schemes based on the three-dimensional structure health diagnosis report using a Cityengine planning module; An optimization module imports the set of feasible reconstruction schemes into an AnyLogic simulation module to simulate social benefits and obtain an evaluation report of an economic benefit prediction curve, establishes an emotion propagation partial differential equation, predicts negative emotion density according to a citizen emotion density distribution map, dynamically adjusts a public space configuration ratio of the set of feasible reconstruction schemes, and obtains an optimized scheme.

[0076] The embodiment also provides a computer device suitable for the AI-based urban health check improvement scheme optimization method, which includes a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the AI-based urban health check improvement scheme optimization method proposed in the above embodiment.

[0077] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0078] The embodiment also provides a storage medium on which a computer program is stored, the program being executed by a processor to implement the method for optimizing an improvement scheme of city health examination based on artificial intelligence as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0079] To sum up, the application innovatively couples the building stress distribution and the citizen emotion density for analysis by constructing a physically constrained neural network model, and realizes the multi-modal deep fusion of structure health and social emotion by using a PDE loss function, which can accurately identify the high stress-high emotion coupling area, and a dynamic feedback mechanism is established based on the emotion propagation partial differential equation to convert the negative emotion concentration prediction value into a public space configuration optimization scheme, forming a closed loop from data perception to scheme optimization.

[0080] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. An artificial intelligence-based urban health check improvement scheme optimization method, characterized in that: comprising, obtain social media text data and urban planning GIS data for spatio-temporal alignment processing to obtain a structured data set; analyze the structured data set to calculate building stress distribution and road bearing capacity to obtain a physical health heat map; use a BERT model to perform sentiment analysis on the social media text data, combine geocoding technology to obtain a citizen emotional density distribution map, construct a physically constrained neural network model, embed the physical health heat map into the network loss function, and obtain a three-dimensional structural health diagnosis report; based on the three-dimensional structural health diagnosis report, use the Cityengine planning module to generate a set of feasible reconstruction schemes; import the set of feasible reconstruction schemes into the AnyLogic simulation module to simulate social benefits and obtain an evaluation report of the economic benefit prediction curve, establish an emotional propagation partial differential equation, predict the negative emotional concentration based on the citizen emotional density distribution map, dynamically adjust the public space configuration proportion of the set of feasible reconstruction schemes, and obtain an optimized scheme. 2.The method of claim 1, wherein the method comprises: obtain social media text data and urban planning GIS data for spatio-temporal alignment processing to obtain a structured data set, including the following steps, obtain social media text data through an API interface, use a coordinate conversion module to unify the spatial reference of the social media text data, use a WFS service to obtain urban planning GIS data through a municipal geographic information public service module, and combine building vibration sensor time series data; use a dynamic time warping algorithm to perform spatio-temporal alignment on the building vibration sensor time series data, social media text data, and urban planning GIS data to obtain a structured data set. 3.The method of claim 2, wherein the method further comprises: determining a first score of the first improvement scheme based on the first evaluation result; determining a second score of the second improvement scheme based on the second evaluation result; and selecting the first improvement scheme or the second improvement scheme based on the first score and the second score. analyze the structured data set to calculate building stress distribution and road bearing capacity to obtain a physical health heat map, including the following steps, perform Butterworth band-pass filtering on the vibration signals in the structured data set to obtain denoised acceleration time series data, obtain building mass from material density and building volume in the urban planning GIS data, calculate dynamic load, and obtain filtered dynamic load time series data; extract building GIS contours from the urban planning GIS data, establish a three-dimensional finite element model in ANSYS based on the filtered dynamic load time series data and the building GIS contours, apply the load sequence in the filtered dynamic load time series data to the three-dimensional finite element model floor nodes, and calculate building stress distribution; calculate road deflection values based on vibration data in the urban planning GIS data; superimpose and render the road deflection values and building stress distribution in ArcGIS to obtain a physical health heat map. 4.The method of claim 3, wherein the method further comprises: determining a first score of the first improvement scheme based on the first evaluation result; determining a second score of the second improvement scheme based on the second evaluation result; and selecting the first improvement scheme or the second improvement scheme based on the first score and the second score. use a BERT model to perform sentiment analysis on the social media text data, combine geocoding technology to obtain a citizen emotional density distribution map, construct a physically constrained neural network model, embed the physical health heat map into the network loss function, and obtain a three-dimensional structural health diagnosis report, including the following steps, use a BERT model to perform sentiment classification on the social media text data to obtain labeled data with sentiment scores; extract urban areas in the planning drawing for grid division and building GIS contour alignment, calculate the emotional density value of each grid, and obtain a citizen emotional density distribution map; Based on the labeled data with emotional scores and physical health heat maps, pixel-level spatial alignment is performed to obtain a stress-emotion dual-channel image after registration, a physically constrained neural network model is constructed, and the physically constrained neural network model is trained in combination with a PDE loss function; The registered stress-emotion dual-channel image is input into the trained physically constrained neural network model to predict the global stress field and mark the high-emotion-high-stress coupling area, and a three-dimensional structural health diagnosis report is obtained. 5.The method of claim 4, wherein the method further comprises: determining a first score of the first improvement scheme based on the first evaluation result; determining a second score of the second improvement scheme based on the second evaluation result; and selecting the first improvement scheme or the second improvement scheme based on the first score and the second score. Based on the three-dimensional structural health diagnosis report, a set of feasible reconstruction schemes are generated using the Cityengine planning module, including the following steps, Based on the three-dimensional structural health diagnosis report, a set of feasible reconstruction schemes are generated using the Cityengine planning module, including the following steps, Based on the three-dimensional structural health diagnosis report, a set of feasible reconstruction schemes are generated using the Cityengine planning module, including the following steps, Based on the three-dimensional structural health diagnosis report, a set of feasible reconstruction schemes are generated using the Cityengine planning module, including the following steps, The optimization scheme is imported into Unity3D to simulate human flow and structural response, and a set of feasible reconstruction schemes are generated. 6.The method of claim 5, wherein the method further comprises: The set of feasible reconstruction schemes is imported into the AnyLogic simulation module to simulate social benefits, and an evaluation report of the economic benefit prediction curve is obtained, including the following steps, The building models in the set of feasible reconstruction schemes are converted to GLTF format, and the road network is converted to GeoJSON format. A three-dimensional simulation environment is built in AnyLogic, and the dynamic behavior logic of citizen agents and vehicle agents is set, to obtain an initialized simulation model; According to the citizen emotion density distribution map, and configure the agent behavior rules; The initialized simulation model and agent behavior rules are loaded into the AnyLogic cloud computing module, and Monte Carlo simulation is run to record the data of human flow, business income, and public facility usage rate, to obtain the original simulation data. The business income growth rate and facility usage Gini coefficient are calculated to obtain the economic benefit prediction curve evaluation report. 7.The method of claim 6, wherein the method further comprises: determining a first score of the first improvement scheme based on the first evaluation result; determining a second score of the second improvement scheme based on the second evaluation result; and selecting the first improvement scheme or the second improvement scheme based on the first score and the second score. Establish an emotion propagation partial differential equation to predict the negative emotion concentration according to the citizen emotion density distribution map, and dynamically adjust the public space configuration proportion of the set of feasible reconstruction schemes to obtain an optimization scheme, including the following steps, Based on the citizen emotion density map, a reaction-diffusion partial differential equation is established, and the emotion evolution model is obtained by solving the equation using the finite difference method; Using the emotion evolution model and social media text data, the equation is solved by the finite difference method to obtain the negative emotion hotspot boundary; According to the hotspot area in the negative emotion hotspot boundary, the proportion of public space is dynamically increased to obtain a reconstruction scheme; The reconstruction scheme is imported into AnyLogic simulation to verify the emotion improvement rate to obtain an optimization scheme.

8. The system for optimizing the improvement scheme of the city health check based on artificial intelligence according to any one of claims 1-7, characterized in that: including, The data acquisition module acquires social media text data and urban planning GIS data for spatio-temporal alignment processing to obtain a structured data set; The data analysis module analyzes the structured data set to calculate the building stress distribution and road carrying capacity, and obtains a physical health heat map; The health diagnosis module uses a BERT model to perform sentiment analysis on social media text data, obtains a citizen emotion density distribution map by combining geocoding technology, constructs a physically constrained neural network model, embeds a physical health heat map into a network loss function, and obtains a three-dimensional structure health diagnosis report. The scheme reconstruction module uses a Cityengine planning module based on the three-dimensional structure health diagnosis report to generate a set of feasible reconstruction schemes. The optimization module imports the set of feasible reconstruction schemes into an AnyLogic simulation module to simulate social benefits, obtains an evaluation report of an economic benefit prediction curve, establishes an emotion propagation partial differential equation, predicts negative emotion concentration based on the citizen emotion density distribution map, dynamically adjusts the public space configuration proportion of the set of feasible reconstruction schemes, and obtains an optimized scheme. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to implement the steps of the method for optimizing an improvement scheme of a city physical examination based on artificial intelligence according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the method for optimizing an improvement scheme of a city physical examination based on artificial intelligence according to any one of claims 1-7.

Images