Methods and devices for monitoring building deformation using artificial intelligence

Abstract

This invention discloses a method and apparatus for monitoring building deformation using artificial intelligence, relating to the technical field of building deformation monitoring. The method includes: interactively obtaining building information to be monitored, performing deformation sensitivity analysis, and establishing deformation-sensitive points of interest; parsing the building information, identifying prominent feature points, and establishing candidate feature comparison points; establishing a first adaptation result based on the observation adaptation analysis results; performing association matching of candidate feature comparison points to establish a second adaptation result; confirming the feature comparison points using the second adaptation result, and then using distributed monitoring sensors and feature comparison points to monitor building deformation. This invention solves the technical problems of existing technologies, such as difficulty in comprehensively covering key areas, significant susceptibility to environmental interference, resulting in limited accuracy, insufficient monitoring efficiency, and poor adaptability to complex building structures. It achieves the technical effect of improving the efficiency, accuracy, and adaptability of building deformation monitoring.

Description

Technical Field

[0001] This application relates to the technical field of building deformation monitoring, specifically to a method and device for monitoring building deformation using artificial intelligence. Background Technology

[0002] In the field of modern building construction, with the acceleration of urbanization and the continuous expansion of building scale, the demand for building safety monitoring is becoming increasingly urgent. However, traditional methods for monitoring building deformation, such as conventional leveling and total station monitoring, have many drawbacks, including low efficiency, limited accuracy, and difficulty in real-time and comprehensive monitoring. Leveling instruments monitor deformation by measuring the height difference between two points, but their operation is cumbersome, requiring manual point-by-point measurement, resulting in low efficiency and being greatly affected by the skill level of the observers, making it difficult to monitor large-area buildings in real time and comprehensively. Total station monitoring uses angle and distance measurements to determine the positional changes of target points. Although it has relatively high accuracy, in complex environments, such as urban areas with many high-rise buildings, the signal is easily blocked and interfered with, leading to missing or inaccurate monitoring data. Sensor network-based monitoring technologies, such as strain gauges and displacement gauges, can achieve a certain degree of automated monitoring, but the sensor deployment is limited, unable to cover all critical parts of the building, and has poor adaptability to buildings with complex structures and diverse deformation patterns, making it difficult to monitor subtle local deformations of buildings.

[0003] Therefore, current technologies suffer from several technical problems, including difficulty in fully covering key areas, significant susceptibility to environmental interference, resulting in limited accuracy and efficiency in building deformation monitoring, as well as poor adaptability to complex building structures. Summary of the Invention

[0004] This application provides a building deformation monitoring method and device that incorporates artificial intelligence, which solves the technical problems in the prior art, such as difficulty in fully covering key parts, high susceptibility to environmental interference, resulting in limited accuracy, insufficient monitoring efficiency, and poor adaptability to complex building structures. It achieves the technical effect of improving the efficiency, accuracy, and adaptability of building deformation monitoring.

[0005] This application provides a method for monitoring building deformation using artificial intelligence. The method includes: interactively obtaining building information to be monitored; performing deformation sensitivity analysis using the building information to establish deformation sensitivity concerns, wherein the deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis; parsing the building information; identifying prominent feature points based on the parsing results to establish candidate feature comparison points; obtaining observation adaptation analysis results of the candidate feature comparison points; establishing a first adaptation result based on the observation adaptation analysis results; performing association matching of the candidate feature comparison points based on the deformation sensitivity concerns; establishing a second adaptation result based on the association matching result and the first adaptation result; confirming the feature comparison points using the second adaptation result; distributing monitoring sensors; and using the distributed monitoring sensors and feature comparison points to monitor building deformation.

[0006] In a possible implementation, the artificial intelligence-integrated building deformation monitoring method further performs the following processing: extracting BIM data, surveying data, and design drawing data of the building information to establish a digital twin model; performing finite element simulation based on the digital twin model to simulate the structural response under varying loads and establishing a first deformation-sensitive concern; performing stress concentration analysis on the digital twin model to establish a second deformation-sensitive concern; performing influence analysis on the digital twin model based on the coupling effect of temperature, humidity, and vibration environment to establish a third deformation-sensitive concern; and establishing the deformation-sensitive concern using the first, second, and third deformation-sensitive concerns.

[0007] In a possible implementation, the building deformation monitoring method combined with artificial intelligence also performs the following processing: activating a physical simulation tool and using the physical simulation tool to simulate the temperature field, humidity field, and vibration field; performing simulation fitting of a digital twin model through a hybrid field composed of the temperature field, humidity field, and vibration field; performing coupled evaluation based on temperature-sensitive indicators, humidity-sensitive indicators, and vibration-sensitive indicators; and establishing the third deformation-sensitive concern.

[0008] In a possible implementation, the building deformation monitoring method combined with artificial intelligence further performs the following processes: extracting structural geometric relationships, component relationships, material parameters, and node location information from CAD drawings using natural language processing and computer vision; extracting geometric visual features using the structural geometric relationships and component relationships to establish a first recognition result; locating component intersection points, load-bearing nodes, connection nodes, edge points, and corner points using the component relationships and node location information, and establishing a second recognition result based on the location result; identifying material change points using the material parameters to establish a third recognition result; and establishing candidate feature comparison points based on the first recognition result, the second recognition result, and the third recognition result.

[0009] In a possible implementation, the artificial intelligence-integrated building deformation monitoring method further performs the following processes: generating deformation types and monitoring granularity based on the deformation sensitivity concerns; performing association matching of the candidate feature comparison points based on the deformation types and monitoring granularity to generate association matching results; using the association matching results and the first adaptation results to filter candidate feature points and establish filtering results; using the monitoring granularity to perform coverage evaluation of the association matching results and generating feature comparison point search instructions based on the coverage evaluation results; performing new searches for monitoring comparison points under the filtering results according to the feature comparison point search instructions, and establishing a second adaptation result based on the new search results and the filtering results.

[0010] In a possible implementation, the building deformation monitoring method combined with artificial intelligence also performs the following processing: reading the monitoring results of the monitoring sensors, performing deformation identification through the monitoring results and the feature comparison points, and generating a first deformation identification result; acquiring historical monitoring data, performing monitoring result deviation analysis based on feature comparison points using the historical monitoring data, and establishing a second deformation identification result; and reporting a deformation warning through the first deformation identification result and the second deformation identification result.

[0011] In a possible implementation, the building deformation monitoring method combined with artificial intelligence also performs the following processes: configuring an early warning response database; using the early warning response database to perform early warning trigger analysis on the first deformation identification result and the second deformation identification result; and calling the early warning scheme to issue a deformation early warning based on the early warning trigger analysis result.

[0012] This application also provides a building deformation monitoring device combining artificial intelligence. The device includes: a deformation sensitivity analysis module, used to interactively obtain building information to be monitored, and then use the building information to perform deformation sensitivity analysis and establish deformation sensitivity concerns. The deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis; a feature protrusion point identification module, used to parse the building information, identify feature protrusion points based on the parsing results, and establish candidate feature comparison points; a first adaptation result establishment module, used to obtain the observation adaptation analysis results of the candidate feature comparison points and establish a first adaptation result based on the observation adaptation analysis results; a second adaptation result establishment module, used to perform association matching of the candidate feature comparison points based on the deformation sensitivity concerns, and establish a second adaptation result based on the association matching results and the first adaptation result; and a building deformation monitoring module, used to confirm the feature comparison points using the second adaptation result, distribute monitoring sensors, and use the distributed monitoring sensors and feature comparison points to monitor building deformation.

[0013] The proposed method and device for monitoring building deformation using artificial intelligence involves interactively obtaining building information to be monitored, performing deformation sensitivity analysis to establish deformation-sensitive areas of interest, parsing the building information, identifying prominent feature points, and establishing candidate feature comparison points. Based on the observation and adaptation analysis results, a first adaptation result is established; the candidate feature comparison points are then matched to establish a second adaptation result; after confirming the feature comparison points using the second adaptation result, monitoring sensors are distributed, and building deformation is monitored using these distributed sensors and feature comparison points. This method solves the technical problems of existing technologies, such as difficulty in comprehensively covering key areas, significant susceptibility to environmental interference, resulting in limited accuracy, insufficient monitoring efficiency, and poor adaptability to complex building structures. It achieves the technical effect of improving the efficiency, accuracy, and adaptability of building deformation monitoring. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0015] Figure 1 This is a schematic diagram of the process for a building deformation monitoring method incorporating artificial intelligence, provided in an embodiment of this application.

[0016] Figure 2 This is a schematic diagram of the structure of a building deformation monitoring device that incorporates artificial intelligence, provided in an embodiment of this application.

[0017] Figure labeling: Deformation sensitivity analysis module 10, Feature protrusion point identification module 20, First adaptation result establishment module 30, Second adaptation result establishment module 40, Building deformation monitoring module 50. Detailed Implementation

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0021] This application provides a method for monitoring building deformation by incorporating artificial intelligence, such as... Figure 1 As shown, the method includes:

[0022] Step S100: After interactively obtaining the building information to be monitored, use the building information to perform deformation sensitivity analysis and establish deformation sensitivity focus. The deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis.

[0023] Preferably, the interactive acquisition of building information to be monitored may include: obtaining basic design information such as structural layout, dimensions, and materials used from architectural design drawings; obtaining actual conditions during construction, such as construction techniques and quality control, from construction records; obtaining geological conditions of the building's location, such as soil type, bearing capacity, and groundwater level, from geological exploration reports; and collecting environmental information around the building, such as temperature, humidity, and wind speed, through sensor data. Then, deformation sensitivity analysis is performed based on the building information, identifying key parts and factors prone to deformation. This deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis.

[0024] Preferably, structural weakness analysis includes structural component analysis, structural connection analysis, and structural system analysis. Structural component analysis refers to a thorough evaluation of major structural components such as beams, columns, and slabs based on architectural design drawings and construction records. Components with smaller dimensions or lower material strength have relatively weaker load-bearing capacity and are more prone to deformation under stress. For example, beams with excessively small cross-sectional areas may experience bending deformation under large loads, and columns with insufficient concrete strength may undergo compressive deformation under pressure. Connection node analysis involves analyzing the connection methods (such as welding and bolting), structural details, and construction quality of nodes to determine whether there are issues such as weak connections or stress concentration. Problems include loose or poorly constructed parts that are prone to loosening and cracking under stress, leading to structural deformation. For example, poor welding quality at steel structure joints may cause joint failure under repeated loading, thus affecting the stability of the entire structure. Structural system analysis refers to evaluating the overall structural system of a building, such as frame structures, shear wall structures, and tube structures. Different structural systems have different stress performances, and some structural systems may be more prone to deformation in specific directions or under loads. For example, frame structures have relatively low lateral stiffness and may produce large lateral displacements under horizontal loads (such as wind loads and seismic loads).

[0025] Preferably, the stress concentration analysis specifically includes load type analysis, load distribution analysis, and load combination analysis. Load type analysis refers to identifying the various loads acting on the building, including dead loads (such as structural self-weight and decoration weight), live loads (such as the weight of people, equipment, and furniture), wind loads, snow loads, and seismic loads. Different types of loads affect the structure in different ways and to varying degrees. For example, wind loads mainly exert horizontal thrust on the building's facade and upper floors, potentially causing lateral deformation, while seismic loads cause complex vibration responses, potentially affecting various parts of the structure. Load distribution analysis refers to... Analyzing the distribution of loads on a building structure helps identify areas of concentrated load. For example, in a large shopping mall, the floor slabs around the atrium may bear significant crowd loads, while in an industrial plant, beams and columns in equipment placement areas may bear substantial concentrated loads. If these areas of concentrated load are not adequately considered during the design phase, deformation problems are likely to occur. Load combination analysis assesses the simultaneous action of multiple loads. Different load combinations may cause certain parts of the structure to bear greater internal forces, thereby increasing the risk of deformation. For instance, under seismic loading, considering the combination of wind loads and live loads simultaneously may make the stress situation of the structure more complex.

[0026] Preferably, the environmental sensitivity analysis specifically includes temperature impact analysis, humidity impact analysis, and geological condition analysis. Among these, temperature changes cause thermal expansion and contraction of building materials, leading to structural deformation. For buildings that are more sensitive to temperature changes, such as large-span structures and ultra-long structures, the impact of temperature stress needs to be assessed in detail. For example, in the high temperatures of summer, concrete structures may crack due to expansion, while in the low temperatures of winter, steel structures may loosen at joints due to contraction. Humidity changes affect the performance of building materials, especially materials such as wood and masonry. At high humidity, wood may rot and deform, and masonry may suffer from dampness and weathering. At low humidity, materials may crack due to dryness. Geological conditions affect the foundation stability of buildings. If a building is located in an area with poor geological conditions, such as a soft soil foundation or an earthquake-prone area, structural deformation may occur due to foundation settlement or earthquakes. For example, uneven settlement of a soft soil foundation may cause the building to tilt and crack in the walls, while earthquakes may cause the foundation to shift, thus affecting the safety of the superstructure. After completing structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis, the results of each analysis are comprehensively evaluated. Based on the degree of influence of different factors on structural deformation, key parts and areas in the building that are prone to deformation are comprehensively identified, and deformation-sensitive monitoring is established to enable more targeted deformation monitoring of the building and timely detection of potential safety hazards.

[0027] Furthermore, step S100 also includes step S110, extracting BIM data, surveying data, and design drawing data of the building information to establish a digital twin model; step S120, performing finite element simulation based on the digital twin model to simulate the structural response under varying loads and establishing a first deformation-sensitive concern; step S130, performing stress concentration analysis on the digital twin model and establishing a second deformation-sensitive concern; step S140, performing influence analysis on the digital twin model based on the coupling effect of temperature, humidity, and vibration environment and establishing a third deformation-sensitive concern; and step S150, establishing the deformation-sensitive concern using the first, second, and third deformation-sensitive concerns.

[0028] Preferably, BIM data, surveying data, and design drawing data are extracted from building information. BIM data includes detailed data such as the building's 3D model, component information, and material properties; surveying data provides information such as the building's actual geographical location, topography, and actual dimensions; and design drawing data includes detailed architectural drawings such as floor plans, sections, and elevations. A digital twin model is then built using the extracted data. This model is a digital mapping of the real building and can accurately reflect the building's geometry, structural characteristics, and material properties. Then, finite element simulation analysis is performed based on the digital twin model. Finite element simulation analysis is a numerical calculation method that discretizes a continuous object into a finite number of elements for analysis. Specifically, by dividing the digital twin model into a large number of finite element elements and assigning corresponding material properties and boundary conditions to each element, the response of the building structure under various changing loads is simulated. During the simulation process, the stress, strain distribution and displacement changes of the structure are observed, and the parts that are prone to large deformations are identified as the first deformation sensitive focus. For example, when simulating seismic loads, it may be found that stress concentration is obvious and displacement changes are large at some weak layers or key nodes of the building.

[0029] Preferably, a stress concentration analysis is performed on the digital twin model. By analyzing the magnitude and distribution of loads borne by various parts of the model, areas of stress concentration are identified. For example, in buildings, the junctions of columns and beams, and the supports of large equipment, may bear large concentrated loads, making these areas prone to significant deformation. This leads to the establishment of a second deformation-sensitive area of ​​concern. Considering the coupling effects of temperature, humidity, and vibration, an impact analysis is performed on the digital twin model. Specifically, temperature changes cause thermal expansion and contraction of materials, humidity changes affect material properties and structural stability, and vibration can cause fatigue damage. The interaction of these three factors has a complex impact on the deformation of the building structure. By analyzing the structural response under different environmental conditions, key factors and sensitive areas with significant impact on structural deformation are identified, leading to the establishment of a third deformation-sensitive area of ​​concern. For example, in some industrial plants, vibrations from production equipment and changes in ambient temperature may cause loosening and deformation in certain connections. Finally, the first, second, and third deformation-sensitive areas of concern are comprehensively analyzed and integrated to obtain a final deformation-sensitive area of ​​concern. This facilitates focused monitoring and maintenance of deformation-sensitive areas to ensure the safety and stability of the building.

[0030] Furthermore, step S140 also includes step S141, activating the physical simulation tool and using the physical simulation tool to simulate the temperature field, humidity field, and vibration field; step S142, performing simulation fitting of the digital twin model through the mixed field composed of the temperature field, humidity field, and vibration field, and performing coupled evaluation based on temperature sensitivity index, humidity sensitivity index, and vibration sensitivity index to establish the third deformation sensitivity concern.

[0031] Preferably, a physical simulation tool is activated to perform temperature field simulation on the digital twin model of the building. This includes simulating the ambient temperature, sunlight conditions, thermal conductivity of building materials, and internal heat sources (such as electrical equipment and heat dissipation from people) of the building, simulating the temperature distribution of the building at different times and in different parts. For example, the simulation can determine how much the temperature of a wall exposed to direct sunlight will rise in summer, and the temperature differences between different rooms inside the building due to different insulation performance. Next, a humidity field simulation is performed, including simulating ambient humidity, waterproofing and moisture-proofing measures of the building, hygroscopicity of materials, and ventilation conditions, simulating humidity changes in different locations inside and outside the building. For example, the humidity may be higher in the basement due to poor ventilation, while the humidity is relatively lower in rooms near windows with good ventilation. Finally, a vibration field simulation is performed, including simulating external vibration sources that the building is subjected to, such as vehicle traffic on nearby roads, operation of industrial equipment, and seismic activity, as well as the vibration characteristics of the building's own structure. The simulation analysis can determine the vibration response of various parts of the building under the action of different vibration sources, including parameters such as vibration amplitude and frequency.

[0032] Preferably, a hybrid field consisting of temperature, humidity, and vibration fields is applied to a digital twin model for simulation fitting. This involves a coupled evaluation based on temperature-sensitive, humidity-sensitive, and vibration-sensitive indices. Specifically, temperature-sensitive indices include thermal stress and thermal deformation (calculating the stress and displacement of various parts of the building due to temperature changes and identifying areas with high temperature differences) and temperature gradient sensitivity (establishing evaluation indices for the impact of temperature gradients on local structures based on the thermal expansion coefficients of different components). Humidity-sensitive indices include hygroscopic expansion / contraction effects (assessing dimensional and stiffness changes of materials due to moisture absorption) and corrosion / hydration risk (using material aging models and humidity data to predict possible corrosion or hydration areas). Vibration-sensitive indices include natural frequency changes (monitoring the shift of the building structure's natural frequency under different environmental conditions) and dynamic response amplitude (recording the maximum acceleration and displacement of the vibration response and assessing local or overall vibration anomalies). A coupling model among temperature, humidity, and vibration is constructed by comprehensively considering coupling sensitivity indices to obtain coupling effect coefficients. Then, by combining time series data, the long-term cumulative effect is calculated to obtain the cumulative damage index, such as the degree of material fatigue damage under temperature and humidity interaction and the fatigue crack growth under vibration excitation. Finally, the deformation sensitivity prediction results are obtained, and a third deformation sensitivity concern is established.

[0033] Step S200: Analyze the building information, identify prominent features based on the analysis results, and establish candidate feature comparison points.

[0034] Preferably, the process involves parsing building information (building design drawings, BIM data, surveying data, etc.). For example, extracting information such as the building's structural type, dimensional parameters, and component layout from the design drawings; obtaining the building's 3D model information, component attribute information, and interrelationships from the BIM data; and obtaining information such as the building's actual geographical location, topography, actual dimensions, and spatial location from the surveying data. Then, identifying prominent feature points, i.e., finding points with significant characteristics that represent certain key attributes of the building or distinguish it from other buildings, can be key nodes in the building structure, such as beam-column connection points, wall corners, etc., or unique features of the building's appearance, such as building entrances, window corners, special roof shapes, etc., or the location of elevator shafts, the start and end points of stairs, etc. This forms candidate feature comparison points, which are used as reference points in subsequent comparison and identification tasks. For example, when monitoring building deformation, candidate feature comparison points can be used as monitoring points, and the deformation of the building can be judged by comparing the changes in their positions at different times.

[0035] Furthermore, step S200 also includes step S210, extracting structural geometric relationships, component relationships, material parameters, and node position information from CAD drawings using natural language processing and computer vision; step S220, extracting geometric visual features using the structural geometric relationships and component relationships to establish a first recognition result; step S230, locating component intersection points, load-bearing nodes, connection nodes, edge points, and corner points using the component relationships and node position information, and establishing a second recognition result based on the location results; step S240, identifying material change points using the material parameters to establish a third recognition result; and step S250, establishing candidate feature comparison points based on the first recognition result, the second recognition result, and the third recognition result.

[0036] Preferably, natural language processing is used to identify the text content in CAD drawings. Through text parsing and analysis, information such as the name, specifications, material, and dimensions of each component of the building can be extracted, such as beam numbers and dimensions like "Q235 steel," "C30 concrete," or "KL1(2A)250×500." Computer vision is used to identify lines, graphics, and other elements in CAD drawings to analyze the geometric shape and spatial relationships of the structure. For example, by identifying straight lines, curves, polygons, and other graphics, the outline and boundaries of components can be determined, and then the geometric parameters such as the length, width, height, and angle of the components and the relative positional relationships between different components can be calculated. It can also identify the connection methods and interrelationships between different components in CAD drawings, as well as node position information. For example, the position of beam-column nodes can be identified, thereby obtaining structural geometric relationships, component relationships, material parameters, and node position information.

[0037] Preferably, geometric visual features are extracted from structural geometric relationships and component relationships, including information such as the length, angle, area, and shape of components, as well as relationship features such as connection methods and relative positions between components. This establishes a first identification result for the building's geometry, used to describe the overall geometric form of the building structure and the spatial relationships between components. The relationships between building components and node location information are used to accurately locate key locations such as component intersections, load-bearing nodes, connection nodes, edge points, and corner points. By accurately identifying the locations of key nodes, a second identification result is established, used to analyze the mechanical properties and structural stability of the building structure. Material change points are identified based on material parameters, i.e., the boundaries and change areas of different materials in CAD drawings are found to understand the distribution of different materials in the building and the possible structural performance changes at the intersections of different materials. This establishes a third identification result, understanding the material distribution and changes in the building. Finally, the first, second, and third identification results are combined, and the extracted features and identified key locations are integrated and filtered to determine candidate feature comparison points for subsequent building structure analysis and deformation monitoring.

[0038] Step S300: Obtain the observation and adaptation analysis results of the candidate feature comparison points, and establish the first adaptation result based on the observation and adaptation analysis results.

[0039] Preferably, observation refers to the actual measurement and monitoring of relevant objects (such as buildings or building structures represented by CAD drawings) containing candidate feature comparison points. For example, measuring instruments are used to measure the actual dimensions and locations of the building, or digitized coordinates and dimensions are obtained from CAD drawings. Then, an adaptation analysis of the candidate feature comparison points is performed, including calculating the difference between observed and theoretical values, evaluating the consistency and accuracy of the data, and analyzing the impact of these differences on the overall structure. For example, the actual measured component length is compared with the length of the corresponding component in the CAD drawing, analyzing whether the error between the two is within the allowable range, and the potential impact of the error on the stability of the building structure, thereby determining whether the candidate feature comparison points can accurately reflect the characteristics and state of the building structure. Then, considering all the information obtained from the observation adaptation analysis, including difference data and consistency evaluation results, a comprehensive evaluation and judgment is made on the candidate feature comparison points. The first adaptation result is established based on the candidate feature comparison points that meet the requirements, which is used to detect whether the building has deformed and the degree and location of the deformation, thereby more accurately understanding and describing the characteristics and performance of the building structure.

[0040] Step S400: Based on the deformation-sensitive attention, perform association matching of the candidate feature comparison points, and establish a second adaptation result based on the association matching result and the first adaptation result.

[0041] Preferably, for each candidate feature comparison point, its changes under the influence of different deformation-sensitive factors are analyzed to establish the correlation between the candidate feature comparison point and various deformation-sensitive factors. Then, each candidate feature comparison point is matched with the corresponding deformation-sensitive factor and the type of deformation that may be generated. For example, for candidate feature comparison points at node locations that are easily affected by seismic forces and undergo shear deformation, they are matched with shear deformation caused by seismic forces to clarify the importance and observability of the point under this deformation condition. The correlation matching results and the first fitting results are comprehensively evaluated to gain a more comprehensive understanding of the characteristics and value of each candidate feature comparison point. Based on the comprehensive evaluation results, the candidate feature comparison points are screened and optimized. Those candidate feature comparison points that are closely related to important deformation-sensitive factors in the correlation matching and also perform well in the first fitting results are retained. Those points that are not highly correlated with deformation-sensitive factors or have obvious deficiencies in the first fitting results are eliminated. After screening and optimization, the remaining candidate feature comparison points are sorted and integrated to form the second fitting results, which can more accurately reflect the state changes of the structure when affected by deformation-sensitive factors.

[0042] Furthermore, step S400 also includes step S410, generating deformation types and monitoring granularity based on the deformation-sensitive attention; step S420, performing association matching of the candidate feature comparison points based on the deformation types and monitoring granularity to generate association matching results; step S430, using the association matching results and the first adaptation results to filter candidate feature points and establish filtering results; step S440, using the monitoring granularity to perform coverage evaluation of the association matching results and generating feature comparison point search instructions based on the coverage evaluation results; step S450, performing new search for monitoring comparison points under the filtering results according to the feature comparison point search instructions, and establishing a second adaptation result based on the new search results and the filtering results.

[0043] Preferably, based on the factors of concern regarding deformation sensitivity, the types of deformation that the building structure may experience are determined, such as tension, compression, bending, torsion, and shear deformation. For example, if the focus is on the impact of temperature changes on the building, the tensile or compressive deformation of components caused by thermal expansion and contraction may be considered. If the focus is on seismic action, complex deformations such as bending and shearing of the structure may be included. This clarifies the level of detail and accuracy required for deformation monitoring. The monitoring granularity is determined according to specific engineering needs and structural characteristics. For example, for critical parts or areas more sensitive to deformation, a higher monitoring granularity is required, meaning that even small deformations can be accurately measured. Candidate feature comparison points are then compared with different deformation types and monitoring parameters. The granularity is correlated and associated. Specifically, the sensitivity and correlation of each candidate feature comparison point under different deformation conditions are analyzed to determine whether it can effectively reflect a specific type of deformation and whether it meets the requirements of the monitoring granularity. For example, for a candidate feature comparison point located at a beam-column node, its displacement changes under bending and shear deformation are analyzed to assess whether it can accurately reflect these two types of deformation, while checking whether its accuracy meets the monitoring granularity. Then, the correlation matching results are obtained, that is, a list of matching relationships between each candidate feature comparison point and deformation type and monitoring granularity, which records the degree of correlation of each candidate feature comparison point with different deformation types and whether it meets the monitoring granularity requirements.

[0044] Preferably, the candidate feature comparison points are further screened by combining the correlation matching results and the first adaptation results. Candidate feature comparison points with low correlation to deformation types or that do not meet the monitoring granularity requirements, and which also perform poorly in the first adaptation results, are removed. The screening results are established, which include candidate feature comparison points with high potential and reliability in deformation monitoring. Then, according to the monitoring granularity requirements, the correlation matching results are evaluated to check whether the distribution of candidate feature comparison points under different deformation types can fully cover the areas and deformation conditions that need to be monitored. For example, if the monitoring granularity requires detailed monitoring of the deformation of a certain area of ​​the building structure, but the correlation matching results show that there are few candidate feature comparison points in that area, which cannot meet the coverage requirements of the monitoring granularity, based on the coverage evaluation results, it is determined whether new candidate feature comparison points need to be added and in which locations to search. A feature comparison point search instruction is generated, and the search scope, direction and target are specified to obtain more candidate feature comparison points that meet the requirements. Finally, following the feature comparison point search instructions, the system further searches relevant data such as building structures or CAD drawings based on the screening results to find new monitoring comparison points that meet the requirements. These new points are then added to the screening results to form a second matching result. This result can more accurately and comprehensively reflect the deformation of the building structure under complex working conditions, thereby improving the comprehensiveness and accuracy of building structure deformation monitoring.

[0045] Step S500: After confirming the feature comparison points using the second adaptation result, distributed monitoring sensors are used to monitor building deformation using the distributed monitoring sensors and feature comparison points.

[0046] Preferably, the feature comparison points are confirmed using the second fitting results. This involves selecting the points from the second fitting results that most accurately reflect the deformation characteristics of the building structure. These points are typically located in critical structural areas, such as corners, column tops, beam mid-spans, and foundation connections. Then, based on the characteristics of the feature comparison points and the specific requirements for building deformation monitoring, appropriate monitoring sensors are selected. For example, fiber optic strain gauges or resistance strain gauges can be used to monitor linear deformation; laser displacement sensors or tilt sensors can be used to measure displacement; and accelerometers can be used to measure vibration. The selected monitoring sensors are then distributed and installed on the building to ensure that each feature comparison point can be effectively monitored. The sensor placement must consider factors such as accuracy, range, accessibility of the installation location, and impact on the structure. For example, when installing strain sensors on columns, a suitable height and location must be chosen to avoid installation in areas with decorative layers or those susceptible to external interference. When installing displacement sensors on the roof, it must be ensured that they can accurately measure the roof's displacement changes under different operating conditions.

[0047] Preferably, distributed detection sensors are used to collect various physical quantity data at the feature comparison points in real time, such as strain, displacement, and acceleration, to reflect the state changes of the feature comparison points during the use of the building. For example, when the building is subjected to wind load, strain sensors installed on the walls will collect data on the changes in wall strain, and displacement sensors installed on the roof will collect data on the changes in horizontal and vertical displacement of the roof. Then, by analyzing the sensor data, the deformation of the feature comparison points is obtained, such as the amount of deformation, the rate of deformation, and the trend of deformation. Based on the analyzed and processed data, the deformation of the building is monitored and evaluated. For example, the deformation of the feature comparison points is compared with a preset threshold. If the amount of deformation exceeds the threshold, it indicates that there may be safety hazards in the building. The cause is analyzed and corresponding measures are taken to deal with the problem.

[0048] Furthermore, step S500 also includes step S510, reading the monitoring results of the monitoring sensor, performing deformation identification through the monitoring results and the feature comparison points, and generating a first deformation identification result; step S520, acquiring historical monitoring data, using the historical monitoring data to perform monitoring result deviation analysis based on feature comparison points, and establishing a second deformation identification result; step S530, reporting a deformation warning based on the first deformation identification result and the second deformation identification result.

[0049] Preferably, real-time monitoring data from various sensors on the building is read and combined with feature comparison points for deformation identification. This involves determining whether deformation has occurred at the feature comparison points, and the degree and direction of that deformation, based on changes in the monitoring data. For example, for strain sensors installed on beams, if the monitored strain value exceeds the normal range, and considering the beam's stress condition and the location of the feature comparison point, it can be determined that the beam has undergone a certain degree of bending deformation at that point, thus forming the first deformation identification result. This may include information such as the deformation amount, deformation type (e.g., tensile deformation, compressive deformation, shear deformation, etc.), and the time of deformation occurrence for each feature comparison point. Historical monitoring data of the target building is obtained from data records, including feature comparison data from different time periods and under different operating conditions. The system collects monitoring information from various points and then compares and analyzes the current monitoring results with historical monitoring data. The main focus is on assessing the deviation of the monitoring results from historical data. Specifically, by calculating indicators such as deviation values ​​and rates of change, it evaluates whether the current monitoring results exceed the normal range of change reflected by historical data. For example, if the displacement monitoring value of a certain feature comparison point suddenly increases recently and deviates from historical data under the same season and usage conditions, a second deformation identification result is established based on the deviation analysis results. This result includes not only the degree and nature of the deviation between the current monitoring results and historical data but may also provide a preliminary judgment on the cause of the deviation, such as whether it is due to increased external loads, accumulated structural damage, or changes in environmental factors.

[0050] Preferably, the first deformation identification result and the second deformation identification result are comprehensively evaluated, taking into full account the current deformation status of the feature comparison points and their deviation from historical data. Then, based on the comprehensive evaluation result, a deformation warning is issued according to a preset warning threshold. Specifically, if the first deformation identification result shows that the deformation of some feature comparison points is large, approaching or exceeding the safety limit, or if the second deformation identification result shows that the monitoring result deviates abnormally from historical data and there is a potential safety risk, a deformation warning is triggered, and a detailed warning report is generated, clearly indicating the location of the feature comparison point where the deformation warning occurred, the type and degree of deformation, the deviation from historical data, possible cause analysis, and recommended measures.

[0051] Furthermore, step S530 also includes step S531, configuring the early warning response database; step S532, using the early warning response database to perform early warning trigger analysis on the first deformation identification result and the second deformation identification result; and step S533, calling the early warning scheme to issue a deformation early warning based on the early warning trigger analysis result.

[0052] Preferably, an early warning response database is configured using deformation safety thresholds for different types of building structures at different locations, historical early warning records (including past early warning situations, corresponding deformation data, response measures taken, and final processing results), various parameters related to early warnings (such as sensor calibration parameters, relevant parameters of the building structure, etc.), and detailed information on early warning schemes (such as response procedures corresponding to different levels of early warnings). Then, the data in the first and second deformation identification results are compared and analyzed with the thresholds and other relevant information in the early warning response database. For example, the deformation amount of the feature comparison points in the first deformation identification result is compared with the corresponding safety threshold in the database to see if it exceeds the allowable range. Simultaneously, the degree to which the monitoring results in the second deformation identification result deviate from historical data is analyzed to determine if the early warning triggering conditions are met, thus obtaining the early warning triggering analysis results. If the early warning conditions are met, corresponding deformation early warnings are issued according to the early warning schemes stored in the early warning response database. For example, different notification methods may be used for deformation early warnings of different severity levels, while specifying the appropriate countermeasures, such as building repair, to ensure the safety of the building.

[0053] In the above text, refer to Figure 1 A method for monitoring building deformation incorporating artificial intelligence, according to embodiments of the present invention, is described in detail. Next, reference will be made to... Figure 2 This invention describes a building deformation monitoring device incorporating artificial intelligence according to an embodiment of the present invention.

[0054] The building deformation monitoring device incorporating artificial intelligence according to embodiments of the present invention addresses the technical problems in existing technologies, such as difficulty in comprehensively covering key areas, significant susceptibility to environmental interference, resulting in limited accuracy, insufficient monitoring efficiency, and poor adaptability to complex building structures. It achieves the technical effect of improving the efficiency, accuracy, and adaptability of building deformation monitoring. Figure 2 As shown, the building deformation monitoring device combined with artificial intelligence includes: a deformation sensitivity analysis module 10, a feature protrusion point identification module 20, a first adaptation result establishment module 30, a second adaptation result establishment module 40, and a building deformation monitoring module 50.

[0055] The deformation sensitivity analysis module 10 is used to interactively obtain the building information to be monitored, and then use the building information to perform deformation sensitivity analysis and establish deformation sensitivity concerns. The deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis. The feature protrusion identification module 20 is used to parse the building information, identify feature protrusions based on the parsing results, and establish candidate feature comparison points. The first adaptation result establishment module 30 is used to obtain the observation adaptation analysis results of the candidate feature comparison points and establish a first adaptation result based on the observation adaptation analysis results. The second adaptation result establishment module 40 is used to perform association matching of the candidate feature comparison points based on the deformation sensitivity concerns, and establish a second adaptation result based on the association matching results and the first adaptation result. The building deformation monitoring module 50 is used to distribute monitoring sensors after confirming the feature comparison points using the second adaptation results, and use the distributed monitoring sensors and feature comparison points to monitor building deformation.

[0056] The specific configuration of the deformation sensitivity analysis module 10 will be described in detail below. The deformation sensitivity analysis module 10 further includes: extracting BIM data, surveying data, and design drawing data of the building information to establish a digital twin model; performing finite element simulation based on the digital twin model to simulate the structural response under varying loads and establishing a first deformation sensitivity focus; performing stress concentration analysis on the digital twin model to establish a second deformation sensitivity focus; performing influence analysis on the digital twin model based on the coupling effect of temperature, humidity, and vibration environment to establish a third deformation sensitivity focus; and establishing the deformation sensitivity focus using the first, second, and third deformation sensitivity focuses.

[0057] The specific configuration of the deformation sensitivity analysis module 10 will be described in detail below. The deformation sensitivity analysis module 10 further includes: activating a physical simulation tool and using the physical simulation tool to simulate the temperature field, humidity field, and vibration field; performing simulation fitting of a digital twin model through a hybrid field composed of the temperature field, humidity field, and vibration field; performing coupled evaluation based on temperature sensitivity index, humidity sensitivity index, and vibration sensitivity index; and establishing the third deformation sensitivity focus.

[0058] The specific configuration of the feature protrusion recognition module 20 will be described in detail below. The feature protrusion recognition module 20 further includes: extracting structural geometric relationships, component relationships, material parameters, and node position information from CAD drawings using natural language processing and computer vision; extracting geometric visual features using the structural geometric relationships and component relationships to establish a first recognition result; locating component intersection points, load-bearing nodes, connection nodes, edge points, and corner points using the component relationships and node position information, and establishing a second recognition result based on the location results; identifying material change points using the material parameters to establish a third recognition result; and establishing candidate feature comparison points based on the first recognition result, the second recognition result, and the third recognition result.

[0059] The specific configuration of the second adaptation result establishment module 40 will be described in detail below. The second adaptation result establishment module 40 further includes: generating deformation types and monitoring granularity based on the deformation-sensitive concern; performing association matching of the candidate feature comparison points based on the deformation types and monitoring granularity to generate association matching results; using the association matching results and the first adaptation result to filter candidate feature points and establish filtering results; using the monitoring granularity to perform coverage evaluation of the association matching results, and generating feature comparison point search instructions based on the coverage evaluation results; performing new searches for monitoring comparison points under the filtering results according to the feature comparison point search instructions, and establishing the second adaptation result based on the new search results and the filtering results.

[0060] The specific configuration of the building deformation monitoring module 50 will be described in detail below. The building deformation monitoring module 50 further includes: reading the monitoring results from the monitoring sensors; performing deformation identification based on the monitoring results and the feature comparison points to generate a first deformation identification result; acquiring historical monitoring data; performing deviation analysis of the monitoring results based on the feature comparison points using the historical monitoring data to establish a second deformation identification result; and issuing a deformation warning based on the first deformation identification result and the second deformation identification result.

[0061] The specific configuration of the building deformation monitoring module 50 will be described in detail below. The building deformation monitoring module 50 further includes: configuring an early warning response database; using the early warning response database to perform early warning trigger analysis on the first deformation identification result and the second deformation identification result; and invoking an early warning scheme to issue a deformation early warning based on the early warning trigger analysis result.

[0062] The building deformation monitoring device combined with artificial intelligence provided in this embodiment of the invention can execute the building deformation monitoring method combined with artificial intelligence provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0063] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0064] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for monitoring building deformation using artificial intelligence, characterized in that, The method includes: After interactively obtaining the building information to be monitored, deformation sensitivity analysis is performed using the building information to establish deformation sensitivity concerns. The deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis. The building information is analyzed, and based on the analysis results, prominent feature points are identified, and candidate feature comparison points are established. Obtain the observation and adaptation analysis results of candidate feature comparison points, and establish the first adaptation result based on the observation and adaptation analysis results; Based on the deformation-sensitive attention, the candidate feature comparison points are associated and matched, and a second adaptation result is established based on the association matching result and the first adaptation result; After confirming the feature comparison points using the second adaptation result, distributed monitoring sensors are used to monitor building deformation using the distributed monitoring sensors and feature comparison points. The process of using the building information to perform deformation sensitivity analysis and establish deformation sensitivity concerns includes: Extract the BIM data, surveying data, and design drawing data of the building information to establish a digital twin model; Finite element simulation is performed based on the digital twin model to simulate the structural response under varying loads and establish a first deformation-sensitive concern. Force concentration analysis is performed on the digital twin model to establish a second deformation-sensitive focus; The influence analysis of the digital twin model is carried out based on the coupling effect of temperature, humidity and vibration environment, and a third deformation sensitive concern is established; The deformation-sensitive concern is established using the first deformation-sensitive concern, the second deformation-sensitive concern, and the third deformation-sensitive concern.

2. The method for monitoring building deformation using artificial intelligence as described in claim 1, characterized in that, The influence analysis of the digital twin model based on the coupling effect of temperature, humidity, and vibration environment establishes a third deformation-sensitive concern, including: Activate the physical simulation tool and use it to simulate the temperature field, humidity field, and vibration field. The digital twin model is simulated and fitted using a hybrid field consisting of temperature, humidity, and vibration fields. Coupled evaluation is performed based on temperature-sensitive, humidity-sensitive, and vibration-sensitive indices to establish the third deformation-sensitive concern.

3. The method for monitoring building deformation using artificial intelligence as described in claim 1, characterized in that, The process of parsing the building information, identifying prominent feature points based on the parsing results, and establishing candidate feature comparison points includes: Natural language processing and computer vision are used to extract structural geometric relationships, component relationships, material parameters, and node location information from CAD drawings; Geometric visual features are extracted using the structural geometric relationships and component relationships to establish a first recognition result; Using the component relationships and node location information, the component intersection points, load-bearing nodes, connection nodes, edge points, and corner points are located, and a second identification result is established based on the location results; The material parameters are used to identify material change points, and a third identification result is established. Candidate feature comparison points are established based on the first identification result, the second identification result, and the third identification result.

4. The method for monitoring building deformation using artificial intelligence as described in claim 1, characterized in that, Based on the deformation-sensitive attention, the candidate feature comparison points are associated and matched. A second adaptation result is established based on the association matching result and the first adaptation result, including: Based on the aforementioned deformation sensitivity focus, deformation types and monitoring granularity are generated; Based on the deformation type and monitoring granularity, the candidate feature comparison points are associated and matched to generate association matching results; The candidate feature points are filtered using the association matching results and the first adaptation results to establish a filtering result; The coverage evaluation of the association matching results is performed using the monitoring granularity, and a feature comparison point search instruction is generated based on the coverage evaluation results; Based on the feature comparison point search instruction, a new search for monitoring comparison points is performed under the filtering results, and a second adaptation result is established based on the new search results and the filtering results.

5. The method for monitoring building deformation using artificial intelligence as described in claim 1, characterized in that, The method of monitoring building deformation using distributed monitoring sensors and feature comparison points also includes: Read the monitoring results from the monitoring sensor, and perform deformation identification based on the monitoring results and the feature comparison points to generate a first deformation identification result; Acquire historical monitoring data, and use the historical monitoring data to perform a monitoring result deviation analysis based on feature comparison points to establish a second deformation recognition result; A deformation warning is issued based on the first deformation recognition result and the second deformation recognition result.

6. The method for monitoring building deformation using artificial intelligence as described in claim 5, characterized in that, The step of reporting a deformation warning based on the first deformation recognition result and the second deformation recognition result includes: Configure the early warning response database; The warning response database is used to perform warning trigger analysis on the first deformation identification result and the second deformation identification result; Based on the analysis results of the early warning trigger, the early warning scheme is invoked to issue a deformation early warning.

7. A building deformation monitoring device incorporating artificial intelligence, characterized in that, The device is used to implement the building deformation monitoring method combining artificial intelligence as described in any one of claims 1 to 6, the device comprising: The deformation sensitivity analysis module is used to interactively obtain the building information to be monitored, and then use the building information to perform deformation sensitivity analysis and establish deformation sensitivity concerns. The deformation sensitivity analysis includes structural weakness analysis, stress concentration analysis, and environmental sensitivity analysis. The feature protrusion identification module is used to parse the building information, identify feature protrusions based on the parsing results, and establish candidate feature comparison points; The first adaptation result establishment module is used to obtain the observation adaptation analysis results of candidate feature comparison points and establish the first adaptation result based on the observation adaptation analysis results; The second adaptation result establishment module is used to perform association matching of the candidate feature comparison points based on the deformation-sensitive attention, and establish a second adaptation result based on the association matching result and the first adaptation result. The building deformation monitoring module is used to distribute monitoring sensors after confirming the feature comparison points using the second adaptation result, and to monitor building deformation using the distributed monitoring sensors and feature comparison points.

Images