Digital twin autonomous optimization system based on building structure quality monitoring

Abstract

The application discloses a digital twin autonomous optimization system based on building structure quality monitoring. The system comprehensively collects structure state data, environmental influence data, material performance data and historical record data through a multi-source data acquisition module; extracts characteristic data such as structure deterioration rate and material aging coefficient through an edge computing preprocessing module; constructs a digital twin through a multi-level fusion module, wherein geometric layer fusion realizes crack hierarchical labeling, physical layer fusion corrects parameters such as material elastic modulus, and behavior layer fusion updates structure damping ratio to reproduce dynamic behavior; outputs structure health index, residual life prediction and maintenance suggestions through an autonomous optimization decision module; and through effectiveness evaluation and parameter adjustment, the dynamic iterative updating module ensures that the system continuously adapts to structure changes. The application improves the accuracy of building structure quality evaluation and the scientificity of maintenance decision, and is suitable for the whole life cycle management of various buildings.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of digital twinning, and relates to a digital twinning autonomous optimization system based on building structure quality monitoring. BACKGROUND

[0002] With the accelerated digital transformation of the construction industry, the importance of structure life cycle quality control is increasingly prominent. Current building structure quality monitoring still faces many challenges, among which structure defect identification lag, poor environmental adaptability and insufficient dynamic adaptation in the whole life cycle are particularly prominent, which has become a key bottleneck restricting the development of intelligent construction.

[0003] The existing building structure quality monitoring method can meet the basic needs, but has obvious limitations: on the one hand, the traditional method relies on manual inspection and single-point sensors, data collection is fragmented, it is difficult to realize global real-time monitoring, and the identification accuracy of hidden defects such as cracks and steel corrosion is insufficient, which easily leads to missed judgment of safety hazards; on the other hand, the existing monitoring model is mostly static, cannot effectively fuse environmental data such as temperature, humidity and load, and the evaluation accuracy is greatly reduced in complex environments, and lacks a dynamic iteration mechanism, making it difficult to adapt to material aging and structure changes in the whole life cycle of buildings.

[0004] In view of this, in order to solve the problems raised in the background art, a building structure quality autonomous optimization method based on digital twinning is proposed. SUMMARY

[0005] The purpose of the application can be achieved by the following technical solutions: the application provides a digital twinning autonomous optimization system based on building structure quality monitoring, which comprises a multi-source data acquisition module, an edge computing preprocessing module, a multi-level fusion module, an autonomous optimization decision module and a dynamic iteration updating module.

[0006] The multi-source data acquisition module acquires multi-element data of the target building structure by marking the specified building as the target building, including structure state data, environmental influence data, material performance data and historical record data;

[0007] The edge computing preprocessing module extracts feature data by performing edge computing preprocessing on the multi-element data;

[0008] The multi-level fusion module constructs a digital twinning body that is synchronously mapped with the physical building by multi-level fusion of the feature data;

[0009] The autonomous optimization decision module generates quality grade determination and maintenance suggestions by making autonomous optimization decisions based on the constructed digital twinning body;

[0010] The dynamic iteration updating module performs dynamic iteration updating on the digital twinning body and autonomous optimization decisions through monitoring feedback.

[0011] Preferably, collecting the structure state data comprises: collecting stress values of each monitoring point in real time through the fiber grating stress sensor; collecting crack length, width and area through high-definition industrial cameras combined with image processing algorithms ; determining the total area S of the crack area through the laser range finder

[0012] Preferably, collecting the environmental influence data comprises: collecting environmental temperature and relative humidity through the temperature and humidity sensor; collecting the concentration of corrosive components in the air through the chloride ion sensor; collecting real-time load through the pressure sensor; collecting ultraviolet intensity through the ultraviolet sensor

[0013] Preferably, collecting the material performance data comprises: obtaining the ratio of steel corrosion area to total surface area through the electromagnetic induction type steel corrosion detector; obtaining the concrete surface hardness through the rebound hammer to convert to the measured strength of the concrete ; obtaining the design strength of the concrete in the design drawing ; detecting the connection pre-tightening force through the torque wrench; fitting the curves of the carbonation depth of the concrete and the corrosion thickness of the steel with time based on historical detection data to obtain the aging rate

[0014] A structure health index HI is defined by the crack parameters and stress values, specifically: wherein, is the average stress value, is the material yield stress, is the crack area, and S is the total area of the monitored region.

[0015] Preferably, the step of extracting feature data by the edge computing preprocessing module comprises:

[0016] Data cleaning, spatio-temporal alignment and format conversion are performed on the multi-source data to obtain preprocessed data

[0017] Based on the preprocessed data, the structure deterioration rate is calculated as ;

[0018] wherein, is the initial structure health index, is the current health index, is the interval time

[0019] The material aging coefficient is ;

[0020] wherein, is the measured strength of the concrete, is the design strength of the concrete.

[0021] Preferably, the step of the multi-level fusion, constructing the digital twin mapped synchronously with the physical building, comprises:

[0022] S1, geometric layer fusion is performed to fuse the building three-dimensional model and the structural defect data in the structure state data to generate a geometric twin with defect characteristics, cracks are graded through double thresholds of crack length and crack width into micro cracks, medium cracks and severe cracks, and the graded and labeled crack data is integrated with the original three-dimensional model to form a geometric twin that fully reflects the geometric shape and defect distribution of the physical building;

[0023] The threshold of the micro crack is: crack length ≤ 50mm and crack width ≤ 0.2mm;

[0024] The threshold of the medium crack is: 50mm < crack length ≤ 200mm and 0.2mm < crack width ≤ 0.5mm;

[0025] The threshold of the severe crack is: crack length > 200mm or crack width > 0.5mm;

[0026] S2, physical layer fusion is performed to combine the material performance data with the geometric twin to construct a physical twin with physical properties, specifically: taking the "geometric twin with defect characteristics" generated by geometric layer fusion as the basic framework, fusing the material performance data, giving the geometric model physical properties, and constructing a physical twin with physical properties;

[0027] The material physical parameters are corrected through field measurement data, and the correction formula of the material elastic modulus E is: Wherein, is the design elastic modulus of the material, is the aging coefficient, is the type coefficient of the material; S3, behavior layer fusion is performed to fuse the structural dynamic response data in the structure state data and the environmental influence data to construct a behavior twin that can reproduce the dynamic behavior of the structure, specifically: taking the "physical twin with physical properties" generated by physical layer fusion as the basis, fusing the structural dynamic response data in the structure state data and the environmental influence data, constructing a behavior twin that can reproduce the dynamic behavior of the structure, and realizing virtual simulation of the dynamic response of the physical building under different environmental conditions;

[0028] The structural damping ratio of the behavior twin is updated in real time by the following formula: Wherein, The initial damping ratio of the structure is a fixed reference value, is the temperature influence coefficient, is the frequency influence coefficient, is a load influence coefficient, wherein, 、 、 、 is a change amount of the ambient temperature relative to the initial temperature, is a structural natural frequency, and F is a real-time load size.

[0029] S4, integrating the geometric twin, the physical twin and the behavior twin, to obtain the digital twin;

[0030] Preferably, the autonomous optimization decision module is based on the digital twin to generate a quality level judgment and a maintenance suggestion, specifically:

[0031] According to the building type, the digital twin is loaded with an adaptive intelligent optimization model;

[0032] The parameters of the digital twin are input into the intelligent optimization model for reasoning, and a structure health index HI and a remaining life prediction are output;

[0033] The remaining life prediction is based on crack propagation rate calculation, specifically: wherein, is an initial crack length, is a critical crack length, which is of the cross-sectional thickness of the component; a material constant, a shape factor Y = 1.12, and a fatigue index m = 3.2; is a stress intensity factor amplitude;

[0034] According to the structure health index, a quality level of the building structure is determined based on a preset threshold value:

[0035] HI≥0.8: determined as first level;

[0036] 0.5≤HI<0.8: determined as second level;

[0037] HI<0.5: determined as third level;

[0038] For the quality level of the early warning state or the dangerous state, a targeted maintenance suggestion is generated;

[0039] An optimization benefit value is calculated, which is a quantitative value of the comprehensive maintenance cost and the structure safety improvement, specifically: wherein is a difference value of the health index before and after maintenance, and B>0 indicates that the maintenance scheme is feasible.

[0040] Preferably, the step of dynamic iterative updating comprises:

[0041] An effectiveness coefficient E is calculated and its effect is evaluated, specifically:

[0042] When E is greater than or equal to 0.2: determined as level one;

[0043] When 0.1 is less than E and E is less than 0.2: determined as level two;

[0044] When E is less than 0.1: determined as level three;

[0045] When the building structure is changed, the geometry, physics and behavior parameters of the digital twin are automatically adjusted to maintain synchronization with the physical building.

[0046] Preferably, the digital twin specifically comprises: integrating the geometry twin, the physics twin and the behavior twin to obtain the digital twin;

[0047] The synchronization mapping accuracy of the digital twin is Calculated once a day:

[0048] Wherein, M is the total number of parameters participating in verification, is the twin parameter, is the entity parameter.

[0049] Preferably, a computer device comprises a memory and a processor, and the algorithm iteration step length Satisfies the convergence condition, specifically: And Wherein, is the model parameter, is the objective function, which ensures stable convergence of the iteration process.

[0050] Technical effects and advantages of the present application:

[0051] 1. Improve the accuracy of quality evaluation: through multi-source data acquisition and high-precision mapping of digital twin, combined with structure health index, crack classification and other quantitative indicators, the objective evaluation of the building structure state is realized, the problem of strong subjectivity and insufficient precision of traditional manual detection is solved, and the mapping accuracy can reach more than 90%.

[0052] 2. Optimize the scientific nature of maintenance decision: based on the dynamic simulation and residual life prediction of the digital twin, combined with the effectiveness evaluation of maintenance suggestion, targeted schemes can be generated for different quality levels (good, early warning, dangerous), which can avoid excessive maintenance or insufficient maintenance and reduce 15%-20% of the maintenance cost.

[0053] 3. Realize the dynamic adaptation of the whole life cycle: through the dynamic iteration update module, the data such as structure change and environmental change are integrated in real time, the model parameters and decision logic are continuously optimized, and the system can adapt to the performance changes of the whole life cycle of the building for a long time, which provides continuous and reliable technical support for intelligent construction and operation and maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the basic field, other drawings can also be obtained without creative labor on the basis of these drawings.

[0055] Figure 1 For the schematic diagram of the module connection of the present application.

[0056] Figure 2 For the schematic diagram of the flow implementation of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0058] The specific flow is according to the flow schematic diagram, and the flow schematic diagram is as shown in Figure 2 , and includes the following steps:

[0059] The multi-source data acquisition module acquires the multi-element data of the target building structure by marking the specified building as the target building, including the structure state data, the environmental influence data, the material performance data and the historical record data.

[0060] In this embodiment, the structure state data reflects the current mechanical properties and defect distribution of the building, and is acquired through a distributed stress and strain monitoring unit. Fiber Bragg grating stress sensors are arranged at key stress parts such as beam spans, column bottoms and nodes in the target building, the sampling frequency is set to 1 Hz, and the stress values (unit: MPa) of each monitoring point are acquired in real time. The average stress value of each monitoring area is automatically calculated daily as a basic parameter for structure health assessment; a high-definition industrial camera (resolution ≥ 5 million pixels) is used to scan the building surface, and the crack length, width and area are extracted by combining the image processing algorithm . At the same time, the total area S of the crack area is determined by a laser range finder to provide data for structure health index calculation.

[0061] According to the formula: wherein, is the yield stress of the building material (such as C30 concrete , Q355 steel ), which needs to be input into the system in advance according to the material type of the main structure of the building. The average stress value of the monitoring area Let S be the total area of ​​all cracks in the monitored area. For example, the monitored area of ​​a concrete beam... , , If S = 100000 mm², then: .

[0062] Furthermore, environmental factors directly affect the durability and mechanical performance of building structures, requiring real-time monitoring through environmental impact sensing units. Temperature and humidity sensors should be installed on the building facade and key indoor areas (such as basements and roofs) to collect ambient temperature (T, unit: °C) and relative humidity (RH, unit: %) at a sampling frequency of 0.5 Hz, used to analyze the impact of temperature changes on structural deformation. Chloride ion sensors should be deployed on concrete surfaces to collect the concentration of corrosive components in the air (C, unit: mg / m³) at a sampling frequency of once per hour, assessing the risk of material corrosion. Pressure sensors should be installed on load-bearing components such as floor slabs and roofs to collect real-time loads (F, unit: kN / m²) at a sampling frequency of 1 Hz, focusing on monitoring load changes under conditions of high population density and equipment stacking. Ultraviolet sensors should be installed on the building roof to collect ultraviolet radiation intensity (UV, unit: W / m²), recording the cumulative radiation daily to assess the aging rate of exterior wall materials.

[0063] Furthermore, material performance data reflects the inherent quality of the building structure. This data is collected periodically by a material performance sensing unit. An electromagnetic induction rebar corrosion detector is used to measure the ratio of the corroded area of ​​the rebar to the total surface area on beam and column members, conducted quarterly to assess the degree of degradation in the rebar's mechanical properties. Concrete surface hardness is measured on-site using a rebound hammer, and the actual concrete strength is calculated from this data. Simultaneously retrieve the concrete design strength from the design drawings. As the basis for calculating the material aging coefficient, a torque wrench is used to test the connection preload (unit: N·m) for key parts such as steel structure bolt connections and concrete joints, and the test is conducted every six months to ensure the reliability of joint force transmission. Based on historical test data, the curves of concrete carbonation depth and steel corrosion thickness over time are fitted to obtain the aging rate (k, unit: mm / year), which is used to predict the long-term trend of material performance changes.

[0064] Further, the historical data provides longitudinal comparison basis for structural performance analysis. Through the integration and storage of historical data terminals, including concrete pouring strength, steel reinforcement ratio, curing days, construction temperature, etc., which are extracted from construction logs, the initial quality state of the structure is traced back; the time of past crack repair, the type of reinforcement material (such as carbon fiber cloth, epoxy resin), the replacement of component models, etc. are recorded as a reference for the current maintenance scheme; the structural damage assessment report after disasters such as earthquakes, typhoons, and heavy rains is collected, including crack development, component deformation data, etc., which is used to verify the simulation accuracy of the digital twin in disaster response.

[0065] The edge computing preprocessing module extracts feature data by performing edge computing preprocessing on the multi-element data;

[0066] The data preprocessing steps include data cleaning, space-time alignment, and format conversion.

[0067] It should be noted that the data cleaning is specifically based on the criterion of eliminating abnormal samples with stress values exceeding 2.5 times the standard deviation of historical data (i.e., stress > 5σ), crack width > 5mm (judged as sensor failure); for temperature and humidity, load, etc. data, eliminate values outside the physical reasonable range (such as temperature > 60℃ or <-20℃); when the data missing rate is <5%, use the mean value of adjacent time data to fill in; when the missing rate is ≥5%, mark it as an invalid period and need to re-collect data. For historical data standard deviation, crack width > 5mm (judged as sensor failure); for temperature and humidity, load, etc. data, eliminate values outside the physical reasonable range (such as temperature > 60℃ or <-20℃); when the data missing rate is <5%, use the mean value of adjacent time data to fill in; when the missing rate is ≥5%, mark it as an invalid period and need to re-collect data.

[0068] It should be noted that the space-time alignment is specifically space alignment and time alignment. The space alignment is to convert all sensor data to the building global coordinate system (with the center point of the foundation bottom as the origin, X axis as the building length direction, Y axis as the width direction, Z axis as the height direction), and eliminate the installation position deviation of the sensor through the coordinate conversion matrix; the time alignment is to take the stress sensor timestamp as the reference, and synchronize the data of different frequencies such as temperature and humidity, load temperature and humidity, and load through linear interpolation, to ensure that the timestamp deviation of all data is ≤0.1s.

[0069] It should be noted that the format conversion is to convert image data (such as crack image) into pixel coordinate matrix, with a resolution of 1024x1024; convert sensor data into JSON format, including parameter name, value, timestamp, collection position, etc. field, which is convenient for subsequent fusion calculation.

[0070] Further, the feature data extraction is based on the preprocessed data, and the characteristic parameters reflecting the structural state and material performance are calculated to provide input structural deterioration rate and material aging coefficient for digital twin construction, specifically:

[0071] The structural deterioration rate is ​​; wherein is the structure health index at the initial time, is the health index at the current time, is the interval time. For example, the initial time of a certain building is 0, , 2 years , then: ;

[0072] Further, it is indicated that the structure health state decreases by 0.1 per year on average.

[0073] The material aging coefficient is ; wherein is the measured strength of the concrete, is the design strength of the concrete. For example, the design strength of C30 concrete is , the measured strength is , then: ;

[0074] Further, it is indicated that the material strength has been attenuated by 10%.

[0075] The multi-level fusion module constructs a digital twin body that is mapped synchronously with the entity building by fusing multi-level feature data, and the step of constructing a digital twin body that is mapped synchronously with the entity building includes:

[0076] S1, geometric layer fusion is performed to fuse the building three-dimensional model and the structure defect data in the structure state data to generate a geometric twin body with defect features;

[0077] S2, physical layer fusion is performed to combine the material performance data with the geometric twin body to construct a physical twin body with physical properties;

[0078] S3, behavior layer fusion is performed to fuse the structure dynamic response data in the structure state data with the environmental influence data to construct a behavior twin body that can reproduce the dynamic behavior of the structure;

[0079] S4, the geometric twin body, the physical twin body and the behavior twin body are integrated to construct the digital twin body;

[0080] It should be noted that the geometric layer fusion combines the building three-dimensional model with the structure defect data to generate a geometric twin body with defect features, which directly presents the structure appearance and defect distribution, and specifically: basic model construction: taking the building BIM model as the basic framework, the geometric size, spatial position and other information of the components such as beams, columns and plates are included, and the precision is controlled within ±5mm; the defects are classified according to the double thresholds of crack length and width, and the specific standards are:

[0081] Micro crack: crack length ≤ 50mm and crack width ≤ 0.2mm, mainly harmless cracks caused by material shrinkage;

[0082] Medium crack: 50mm < crack length ≤ 200mm and 0.2mm < crack width ≤ 0.5mm, need to pay attention to its development trend;

[0083] Severe crack: crack length > 200mm or crack width > 0.5mm, may affect the load bearing capacity of the structure, and needs to be treated immediately.

[0084] Further, through coordinate matching technology, the graded crack data (position, length, width, grade) is mapped to the corresponding position of the BIM model, and different colors are used to identify the crack grade (micro crack is blue, medium crack is yellow, and severe crack is red), to generate a geometric twin. For example, a frame column has a crack with a length of 150mm and a width of 0.3mm, which is marked on the corresponding position of the column body in the geometric twin with a yellow line and with crack parameter information.

[0085] It should be noted that the physical layer fusion combines material performance data with the geometric twin, giving the virtual model physical properties, making it have consistent mechanical properties with the physical structure. Specifically, based on the material performance data, the physical parameters of each component in the geometric twin are corrected, including the elastic modulus, which is: wherein, is the design elastic modulus of the material (e.g. C30 concrete E_design=3.0×10 4 MPa, Q355 steel E_design=2.06×10 5 MPa); is the material aging coefficient; is the material type coefficient (1.0 for concrete, 1.1 for steel structure, and 0.9 for composite materials). For example, the , , of a certain concrete beam is corrected to: Further, the Poisson's ratio is set according to the material type, with 0.2 for concrete and 0.3 for steel.

[0086] Ultimate strength: based on the reinforcement corrosion rate and the measured strength of the concrete, the ultimate tensile and compressive strength of the component is corrected, for example, for every 10% increase in reinforcement corrosion rate, the tensile strength is reduced by 8%.

[0087] Further, the corrected physical parameters are bound to the corresponding components of the geometric twin, such as the elastic modulus and ultimate strength in the beam component attributes, and the connection stiffness in the node attributes, so that the virtual model can truly reflect the stress characteristics of the physical structure.

[0088] It should be noted that the behavior layer fusion is to combine the structural dynamic response data and the environmental impact data to construct a behavior twin that can reproduce the dynamic behavior of the structure, simulate the response of the structure under different environments, specifically:

[0089] Based on the environmental data, the structural dynamic characteristic parameters are updated in real time, and the core parameter is the structural damping ratio, and the calculation formula is: Wherein, 、 、 、 , is the change of the ambient temperature relative to the initial temperature, is the natural frequency of the structure, is the change of the ambient temperature relative to the initial temperature, and F is the real-time load size.

[0090] The initial temperature of a building is 25℃, the current temperature is 35℃ (ΔT=10℃), the natural frequency f=1.5Hz, and the real-time load F=3kN / m², then That is, the damping ratio increases from the initial 0.02 to 0.028, reflecting the change of the energy dissipation capacity of the structure under the action of temperature rise and load.

[0091] Further, the updated damping ratio, elastic modulus and other parameters are input into the structural dynamics equation to simulate the dynamic response (such as displacement amplitude, vibration frequency change) of the structure under the action of vibration, temperature deformation and load. For example, under the action of strong wind load, the vibration displacement of the roof is simulated by the behavior twin, and compared with the actual displacement collected by the sensor to verify the simulation accuracy.

[0092] It should be noted that the digital twin integration is to integrate the geometry, physics and behavior twins to form a complete digital twin, realize the full-dimensional mapping of the physical building, and specifically: data association, model integration and precision verification.

[0093] Further, the data association is to align the spatial coordinates, time stamps of the three types of twins based on the global coordinate system of the building, to ensure that the geometry defect position, physical parameters and dynamic response are one-to-one corresponding in space and time. For example, the crack position marked in the geometry twin needs to be associated with the elastic modulus attenuation in the region of the physics twin and the stress concentration phenomenon at the position of the behavior twin.

[0094] Further, the model integration is to integrate the model data, parameter data and dynamic response data of the three types of twins into a unified database through the digital twin fusion platform, supporting three-dimensional visualization and parameter query. Users can click on any component in the virtual interface while viewing its geometry (including cracks), physical parameters (such as elastic modulus) and dynamic response (such as real-time stress).

[0095] Further, the precision verification: daily calculation of the synchronization mapping precision of the digital twin, to ensure the consistency of the virtual model and the entity building, specifically: Wherein M is the total number of parameters participating in the verification (such as stress, displacement, frequency, etc.). When the precision ≥ 0.8, it is determined that the digital twin is effective; otherwise, the sensor data and model parameters need to be recalibrated until the precision requirement is met.

[0096] An autonomous optimization decision module makes autonomous optimization decisions based on the obtained digital twin, generates quality level judgments and maintenance suggestions;

[0097] It should be noted that the autonomous optimization decision is based on the autonomous optimization decision of the digital twin, which realizes the structure quality level judgment and maintenance suggestion generation through intelligent model reasoning, provides precise guidance for building operation and maintenance, including: intelligent optimization model loading, model reasoning and parameter output, and quality level judgment and maintenance suggestion.

[0098] Further, the intelligent optimization model loading is to select the appropriate intelligent optimization model according to the building type, specifically:

[0099] Super high-rise building: load "seismic response optimization model", focus on analyzing the dynamic response of the structure under seismic load;

[0100] Large-span bridge: load "fatigue damage optimization model", analyze the fatigue life of components based on vehicle load;

[0101] Industrial plant: load "corrosion environment optimization model", combined with the concentration of corrosive components to predict the material corrosion rate.

[0102] Further, the model reasoning and parameter output is to input the feature parameters of the digital twin into the intelligent optimization model, and output the key evaluation indexes, specifically:

[0103] Structural health index (HI): calculated according to the formula above, range [0, 1], the higher the value, the better the structure state.

[0104] Residual life prediction (RUL): based on crack propagation rate calculation, specifically: Wherein, is the initial crack length (unit: m), is the critical crack length (unit: m, taking 1 / 3 of the cross-section thickness of the component); material constant , shape factor Y = 1.12, fatigue index m = 3.2; Stress intensity factor amplitude.

[0105] For example, the initial crack length of a beam component , critical crack , , then: It indicates that the crack needs about 83 years to expand to the critical state.

[0106] Optimization benefit value (B): the quantitative value of the comprehensive maintenance cost and the improvement of the structure safety, specifically: Wherein is the difference value of the health index before and after maintenance, and B>0 indicates that the maintenance scheme is feasible.

[0107] Further, the quality level determination and maintenance suggestion, specifically: determining the quality level based on the preset threshold value of the structure health index HI, including:

[0108] HI≥0.8: determined as first level, indicating good, stable structure performance, no special maintenance is needed, only routine inspection is needed;

[0109] 0.5≤HI<0.8: determined as second level, indicating early warning, the structure has local defects and needs targeted reinforcement;

[0110] HI<0.5: determined as third level, indicating danger, the structure has insufficient bearing capacity and needs to be immediately disabled and a major repair plan is developed.

[0111] According to the determination result, a maintenance suggestion is generated, specifically: for the "early warning" state: if the crack is a micro crack, epoxy resin grouting repair is recommended; if the steel corrosion rate is >10%, it is recommended to brush rust inhibitor; if the joint pre-tightening force is insufficient, it is recommended to re-tighten the bolt; for the "dangerous" state: if the stress in the middle of the beam is over-standard, it is recommended to paste carbon fiber cloth for reinforcement; if the column bottom crack is serious, it is recommended to increase the section for reinforcement; if the foundation settlement is over-standard, it is recommended to use anchor rod static pressure pile to correct deviation.

[0112] A dynamic iterative updating module updates the digital twin and autonomous optimization decision dynamically and iteratively through monitoring feedback;

[0113] It should be noted that the dynamic iterative update continuously optimizes the digital twin and the decision model through monitoring feedback, ensuring that the system adapts to changes in the structure and environment in the long term, specifically including: maintenance suggestion effectiveness evaluation, model adjustment when the structure changes, and long-term data optimization.

[0114] Further, the maintenance suggestion effectiveness evaluation is to calculate the effectiveness coefficient E and evaluate its effect after the implementation of the maintenance scheme, specifically: When E≥0.2: determined as first level, indicating significant effectiveness, the maintenance scheme is reasonable, and the current strategy is maintained;

[0115] When 0.1≤E<0.2: determined as second level, indicating partial effectiveness, and local parameters need to be optimized (such as adjusting the amount of reinforcing material);

[0116] When E < 0.1: determined as level three, indicating invalid, need to redesign maintenance scheme (such as replacement of reinforcement method).

[0117] For example, the HI of a building before maintenance is 0.6, and the HI after maintenance is 0.75, then: Determined as significantly effective, continue to use this type of maintenance scheme.

[0118] It should be noted that when the building structure is changed (such as adding a layer, removing components, replacing materials), the digital twin is automatically updated, specifically: the geometric twin needs to modify the size and position of the changed components in the BIM model, delete the removed parts, and add new components; the physical twin needs to update the material parameters of the changed components (such as the elastic modulus and Poisson's ratio of the newly added steel structure); the behavior twin needs to recalculate the dynamic parameters such as the natural frequency and damping ratio of the structure to ensure consistency with the changed physical structure.

[0119] It should be further noted that this method can be based on long-term operation data (at least 1 year) to periodically optimize system parameters, specifically: adjust the structure deterioration rate and material aging coefficient in the decision model through machine learning algorithm, for example, increase the weight of the concentration of corrosive components in the corrosion environment; use new monitoring data to retrain the intelligent optimization model and update parameters such as crack propagation coefficient and fatigue index to improve prediction accuracy; the algorithm iteration step needs to meet the convergence condition: and where, is the model parameter, is the objective function to ensure stable convergence in the iteration process.

[0120] For example, suppose there is a key parameter (e.g. correction factor of material aging coefficient) in the digital twin of a building structure, with an initial value The system optimizes this parameter through iteration, with the goal of minimizing the prediction error of the updated model, and the iteration rules are as follows: after each iteration, the difference between the new parameter and the old parameter must meet: The specific process is as follows: 1st iteration: old parameter , calculate the new parameter ; the difference is |0.800005-0.8|=0.000005≤0.00001, which meets the condition, and the iteration is valid.

[0121] 2nd iteration: old parameter , calculate the new parameter , the difference is |0.800008-0.800005|=0.000003≤0.00001, which meets the condition, and the iteration continues.

[0122] If the calculation of a certain time , the old parameter ; the difference is |0.80002-0.800005|=0.000015>0.00001, not meet the conditions, need to reduce the adjustment range, recompute (for example, adjust to 0.800012, the difference 0.000007, meet the conditions).

[0123] Through this control, the amplitude of each update of the parameter is small, which ensures that the model converges to the optimal value step by step and avoids the prediction fluctuation caused by large step length.

[0124] Through the cooperative operation of the above modules, the method can realize the whole life cycle autonomous optimization of the building structure quality. In actual application, the mapping accuracy of the digital twin and the physical building can reach more than 90%, the structure health assessment accuracy is improved by 25%, and the maintenance cost is reduced by 15% to 20%. For example, a high-rise residential building early warned the crack expansion risk of the basement beam body through the system, and timely used grouting reinforcement to avoid the crack further developing into a structure safety hidden danger.

[0125] The embodiment of the present application comprehensively integrates all parameters and formulas in the claim by the closed-loop design of multi-source data acquisition, edge computing preprocessing, multi-level fusion, autonomous optimization decision and dynamic iterative updating, realizes the intelligent and precise control of the building structure quality, and provides effective technical support for smart construction and operation.

[0126] Secondly: the drawings in the disclosed embodiment of the present application only involve the structure involved in the disclosed embodiment of the present application, other structures can refer to the usual design, in the case of no conflict, the same embodiment and different embodiments of the present application can be combined with each other;

[0127] Finally: the above only for the preferred embodiment of the present application, and not for limiting the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A digital twin autonomous optimization system based on building structure quality monitoring, characterized by, The method comprises the following steps: A multi-source data acquisition module acquires multi-source data of a target building structure by marking a specified building as a target building, including structure state data, environmental influence data, material performance data, and historical record data; An edge computing preprocessing module extracts feature data by performing edge computing preprocessing on the multi-source data; A multi-level fusion module constructs a digital twin that is synchronously mapped with the entity building by fusing the feature data at multiple levels; The step of constructing a digital twin that is synchronously mapped with the entity building comprises: S1. Geometric layer fusion is performed to fuse the building three-dimensional model and the structure defect data in the structure state data, to generate a geometric twin with defect features, the defects are classified into micro-cracks, medium cracks, and severe cracks by double thresholds of crack length and crack width, the classified and labeled crack data is integrated with the original three-dimensional model, and a geometric twin that fully reflects the geometric shape and defect distribution of the entity building is formed; The threshold of the micro-cracks is that the crack length is less than or equal to 50 mm and the crack width is less than or equal to 0.2 mm; The threshold of the medium cracks is that 50 mm is less than the crack length and less than or equal to 200 mm, and 0.2 mm is less than the crack width and less than or equal to 0.5 mm; The threshold of the severe cracks is that the crack length is greater than 200 mm or the crack width is greater than 0.5 mm; S2. Physical layer fusion is performed to combine the material performance data with the geometric twin, to construct a physical twin with physical properties, specifically, the geometric twin with defect features generated in the geometric layer fusion is taken as a basic framework, the material performance data is fused, the geometric model is given physical properties, and a physical twin with physical properties is constructed; The material physical parameters are corrected by field measurement data, and a correction formula of the material elastic modulus E is: ； wherein, is the design modulus of elasticity of the material, is the aging coefficient, is the type coefficient of the material; S3. Behavior layer fusion is performed to fuse the structure dynamic response data in the structure state data and the environmental influence data, to construct a behavior twin that can reproduce the structure dynamic behavior, specifically, the physical twin with physical properties generated in the physical layer fusion is taken as a basis, the structure dynamic response data in the structure state data and the environmental influence data are fused, a behavior twin that can reproduce the structure dynamic behavior is constructed, and virtual simulation of the dynamic response of the entity building under different environmental conditions is realized; The structural damping ratio of the behavior twin is updated in real time using the following equation: ； wherein, the initial damping ratio of the structure is a fixed reference value, is a temperature influence coefficient, is a frequency influence coefficient, is a load influence coefficient, wherein, , , , is the change amount of the ambient temperature relative to the initial temperature, is the natural frequency of the structure, and F is the real-time load size; S4. The geometric twin, the physical twin, and the behavior twin are integrated, and the digital twin is constructed; An autonomous optimization decision module performs autonomous optimization decision based on the constructed digital twin, to generate quality level judgment and maintenance suggestions; The autonomous optimization decision module generates quality level judgment and maintenance suggestions based on the digital twin, specifically: An intelligent optimization model is loaded into the digital twin according to the building type; The parameters of the digital twin are input into the intelligent optimization model for reasoning, to output a structure health index HI and a remaining life prediction; The remaining life prediction is calculated based on the crack propagation rate, specifically: ； wherein, is the initial crack length, is the critical crack length, taken as ; material constant , shape factor Y = 1.12, fatigue exponent m = 3.2; is the stress intensity factor amplitude; According to the structure health index, the quality level of the building structure is determined based on a preset threshold: HI≥0.8: first level. 0.5≤HI<0.8: determined as secondary; HI<0.5: determined as tertiary; For the quality level of early warning state or dangerous state, a targeted maintenance suggestion is generated; An optimization benefit value is calculated, which is a quantitative value of comprehensive maintenance cost and structure safety improvement, specifically: ； wherein To maintain the difference between the health index before and after maintenance, B>0 indicates that the maintenance scheme is feasible; A dynamic iterative updating module performs dynamic iterative updating on the digital twin and autonomous optimization decision through monitoring feedback; Algorithm iteration step size satisfies a convergence condition, specifically: and ; wherein, are model parameters, is a target function, ensuring stable convergence of the iterative process.

2. The digital twin autonomous optimization system based on building structural quality monitoring of claim 1, wherein, The acquisition of the structural state data includes: real-time acquisition of stress values ​​at each monitoring point using a fiber optic stress sensor; and extraction of crack length, width, and area using a high-definition industrial camera combined with image processing algorithms. The total area S of the region where the crack is located is determined by a laser rangefinder. The environmental influence data is collected, including: collecting environmental temperature and relative humidity through a temperature and humidity sensor; collecting the concentration of corrosive components in the air through a chloride ion sensor; collecting real-time load through a pressure sensor; collecting ultraviolet intensity through an ultraviolet sensor; The material performance data is collected, including: obtaining the ratio of steel bar corrosion area to total surface area by an electromagnetic induction type steel bar corrosion detector; obtaining the concrete surface hardness by a rebound hammer, and converting to obtain the measured concrete strength ; obtaining the concrete design strength by calling the design drawing ; detecting the connection pretightening force by a torque wrench; based on historical detection data, fitting the change curves of concrete carbonation depth and steel bar corrosion thickness with time to obtain the aging rate; A structure health index HI is defined through crack parameters and stress values, specifically: ； wherein, is the average stress value, is the material yield stress, is the crack area, S is the total area of the monitored region.

3. The building structure quality monitoring based digital twin autonomous optimization system of claim 1, wherein, The steps of the edge computing preprocessing module for extracting feature data include: The multi-source data is subjected to data cleaning, time-space alignment and format conversion to obtain preprocessed data; Based on the preprocessed data, the following are calculated: The structural deterioration rate is ; wherein, is the structural health index at the initial time, is the health index at the current time, is the interval time; The material aging coefficient is ; wherein, is the measured strength of the concrete, is the design strength of the concrete.

4. The building structure quality monitoring based digital twin autonomous optimization system of claim 1, wherein, The steps of the dynamic iterative updating include: An effectiveness coefficient E is calculated and its effect is evaluated, specifically: ； When E≥0.2: determined as primary; When 0.1≤E<0.2: determined as secondary; When E<0.1: determined as tertiary; When the building structure is changed, the geometric, physical and behavioral parameters of the digital twin are automatically adjusted to maintain synchronization with the physical building.

5. The building structure quality monitoring based digital twin autonomous optimization system of claim 1, wherein, The digital twin specifically integrates the geometric twin, the physical twin and the behavioral twin to construct the digital twin. wherein the synchronization mapping accuracy of the digital twin Once a day: ； wherein M is the total number of parameters participating in the check, is a twin parameter, is a body parameter.

Images