Building waterproof supervision acceptance system and method
By constructing a permeability network model and a material family knowledge graph, and simulating dynamic loads, the shortcomings of dynamic environmental assessment in waterproofing acceptance are addressed, enabling efficient leakage risk location and maintenance decisions.
Patent Information
- Application Number
- CN202511908189.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-17
- Publication Date
- 2026-02-06
AI Technical Summary
Existing waterproofing acceptance techniques cannot fully assess the actual performance of waterproofing materials in complex dynamic environments, and reliance on static testing methods leads to decision-making biases.
By constructing a permeation network model, combining microchannel impedance tomography and material family knowledge graphs, a composite test parameter set is generated to simulate dynamic loads. By combining spatiotemporal correlation algorithms, high-risk leakage areas are located, and maintenance decision-making schemes are generated.
It enables dynamic performance evaluation of waterproof materials in complex environments, overcomes the decision bias of traditional methods, and provides full-chain performance prediction and disaster resistance analysis.
Smart Images
Figure CN121476017A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of building technology, and more specifically, to a building waterproofing supervision and acceptance system and method. Background Technology
[0002] Waterproofing is a crucial aspect of ensuring the structural safety and functionality of buildings. Its core function is to prevent water seepage into the building's interior, protect the structure from water erosion, extend the building's lifespan, and provide a dry and comfortable indoor environment. The quality of waterproofing directly affects the building's durability, safety, and the living or usage experience.
[0003] After construction is completed, waterproofing acceptance is a crucial step in ensuring the waterproofing project is up to standard. The main functions of acceptance include: ensuring building quality: confirming that the waterproofing construction meets design requirements and preventing structural problems caused by waterproofing issues, such as concrete cracking and steel reinforcement corrosion. Reducing later maintenance costs: timely detection and repair of waterproofing defects reduces subsequent repair costs and construction delays caused by leaks. Extending building lifespan: ensuring the integrity and durability of the waterproofing layer, extending the overall lifespan of the building. Protecting the indoor environment: preventing water leakage from damaging interior decorations, furniture, electrical equipment, etc., ensuring the safety of users' property and quality of life.
[0004] However, existing waterproofing acceptance technologies still have the following shortcomings: Limitations of testing methods: Traditional acceptance methods rely heavily on static testing (such as water pressure testing, surface observation, etc.), which cannot comprehensively evaluate the actual performance of waterproofing materials under complex dynamic environments (such as rainstorm impact, temperature cycling, structural deformation, etc.).
[0005] There is currently no effective solution to the above problems. Summary of the Invention
[0006] This application provides a building waterproofing supervision and acceptance system and method to solve the above-mentioned technical problems.
[0007] This application provides a building waterproofing supervision and acceptance system, including: The permeation network model construction module is used to scan waterproof material samples using a microchannel impedance tomography strategy, extract pore connectivity paths and defect distribution characteristics, and construct a three-dimensional permeation network model that includes pore size, connectivity paths, and defect distribution. The test parameter set generation module is used to map the water pressure and crack propagation laws of concrete structures in historical engineering projects to the current material's permeability network feature space based on the material family knowledge graph, and generate a composite test parameter set that includes rainstorm impact waveforms, temperature cycle gradients and structural deformation rates. The dynamic load application module is used to synchronously apply dynamic loads within the climate simulation test chamber according to the composite test parameter set. The acceptance module is used to locate high-risk leakage areas based on test results using a spatiotemporal correlation algorithm. It then combines a material durability database with regional extreme climate data to generate a decision-making plan that includes maintenance priorities, carbon footprint assessment, and disaster resilience prediction.
[0008] This application provides a method for supervising and accepting building waterproofing, including: By using a microchannel impedance tomography strategy, waterproof material samples were scanned to extract pore connectivity paths and defect distribution characteristics, and a three-dimensional permeation network model including pore size, connectivity paths, and defect distribution was constructed. Based on the material family knowledge graph, the water pressure and crack propagation laws of concrete structures in historical engineering projects are mapped to the current material permeation network feature space, generating a composite test parameter set that includes rainstorm impact waveforms, temperature cycling gradients, and structural deformation rates. Based on the composite test parameter set, dynamic loads are simultaneously applied inside the climate simulation test chamber; Based on the test results, high-risk leakage areas were located using a spatiotemporal correlation algorithm. Combined with a material durability database and regional extreme climate data, a decision-making scheme was generated that included maintenance priorities, carbon footprint assessment, and disaster resilience prediction.
[0009] Furthermore, after constructing the three-dimensional permeation network model, a defect enhancement step is also included: A fractal generative adversarial network is used to synthesize hidden defects. The generator input includes: crack fractal growth paths generated based on the Monte Carlo method; a library of pore connectivity patterns from historical leakage cases; and an interface debonding deformation field constrained by material viscoelastic parameters. Defect samples are reconstructed using a variational autoencoder, and Darcy's law permeability constraint is applied to the latent space. A 3D graph attention network is used to generate a penetration vulnerability heatmap, marking the 3D influence range of high-risk defect clusters.
[0010] Furthermore, the implementation of the Darcy's law permeability constraint includes: A pore network flow model is established, where pores are abstracted as nodes and throats are connected as weighted edges; The equivalent permeability resistance coefficient of each node is calculated using a graph convolutional network. The variance of the permeation resistance of the reconstructed sample is constrained to not exceed the preset proportion of the original sample, and the diameter of the largest connected pore cluster is less than the critical leakage threshold.
[0011] Furthermore, the construction of the material family knowledge graph includes: Define multidimensional node attributes: Material gene node: Fourier descriptor of chemical composition, extrusion molding process parameter curve; Failure characteristic node: Acoustic emission energy release rate spectrum, hot spot diffusion path topology; Environmental load node: regional 50-year rainstorm intensity and duration distribution characteristics; Establish heterogeneous graph relationships: process defect edge: quantify the nonlinear correlation between extrusion temperature fluctuation and pore connectivity; load failure edge: encode the mapping weights between water pressure pulse spectrum characteristics and crack propagation rate; When introducing new materials, their UV curing curves are aligned with the thermoforming parameter space of existing materials through a capsule network.
[0012] Furthermore, the parameter space alignment method includes: A dual-flow feature extraction network was constructed to process the rheological curves of photocurable materials and the melt index spectra of polymer materials. The Wasserstein distance between characteristic distributions was calculated using optimal transport theory. When the distance exceeds the threshold, a cross-modal attention mechanism is triggered to generate a virtual transition material sample set.
[0013] Furthermore, the generation of the rainstorm impact waveform includes: A physical constraint generative adversarial network is constructed: the generator's hidden layer embeds the modified Navier-Stokes equation residuals to simulate the turbulent energy dissipation process; the discriminator introduces dynamic contact angle measurements to filter out abnormal waveforms that do not conform to the material's wetting characteristics. A pulse sequence is generated by a time-temperature controller: the main frequency is set to 1.3 to 1.7 times the material's natural frequency to excite the resonance effect; alternating stress is superimposed on the hot spot area identified by the infrared thermogram, and the amplitude is positively correlated with the local temperature rise rate.
[0014] Furthermore, the following is executed when applying a temperature cyclic gradient: An exponential correlation model between acoustic emission energy and crack propagation rate was established; When the real-time crack rate exceeds the predicted value by 15%, an adaptive temperature control strategy is activated: a low-temperature constraint field of -5℃ to 10℃ is applied within a 0.5mm radius of the crack tip; the heating power in the surrounding 2mm area is increased to 80 to 250W / cm² according to a hyperbolic function. After each cycle, residual strain is measured using digital image correlation to update temperature control parameters and verify crack suppression effect.
[0015] Furthermore, when locating high-risk areas for leakage: The acoustic emission time spectrum was converted into a Mel-scale spectrogram to extract abrupt changes in energy release rate; the isotherm curvature anomaly region in the infrared thermogram was analyzed; and strain gradient distribution data from a distributed fiber optic sensor were fused. Spatial correlation weights are calculated using a 3D convolutional attention network. When the acoustic emission energy is greater than 5 kJ / m³ and accompanied by annular isotherm distribution, it is determined to be a penetrating leakage channel.
[0016] Furthermore, when generating decision-making options, the following are included: Constructing a four-dimensional assessment space: Dimension 1: Construction cost; Dimension 2: Expected lifespan extension rate; Dimension 3: Carbon emissions throughout the entire life cycle; Dimension 4: Probability of failure due to extreme weather. A constrained non-dominated sorting genetic algorithm is used to generate the Pareto solution set; The optimal solution is recommended based on the improved TOPSIS method, and the weight matrix dynamically reflects the owner's environmental preference index.
[0017] Based on the embodiments provided in this application, a three-dimensional permeable network model and a material family knowledge graph are constructed to achieve multi-field coupled dynamic loading of water pressure, temperature, and mechanical stress, overcoming the shortcomings of traditional static detection in simulating complex actual working conditions. By combining spatiotemporal correlation algorithms with regional extreme climate data, maintenance plans including carbon footprint assessment are generated, solving the decision-making bias problem caused by the reliance on experience in traditional methods. By integrating a material durability database, full-chain performance prediction from material production to service maintenance is achieved, filling the gap in existing technologies regarding the lack of long-term disaster resistance analysis. Attached Figure Description
[0018] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a structural diagram of an optional building waterproofing supervision and acceptance system according to an embodiment of this application. Figure 2 This is a flowchart of an optional building waterproofing supervision and acceptance method according to an embodiment of this application; Figure 3 This is a flowchart of an optional building waterproofing supervision and acceptance method according to an embodiment of this application.
[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0021] Optionally, such as Figure 1 As shown, this application provides a building waterproofing supervision and acceptance system, including: The permeation network model construction module 101 is used to scan waterproof material samples using a microchannel impedance tomography strategy, extract pore connectivity paths and defect distribution characteristics, and construct a three-dimensional permeation network model that includes pore size, connectivity paths and defect distribution. The test parameter set generation module 102 is used to map the water pressure and crack propagation laws of concrete structures in historical engineering projects to the current material's permeation network feature space based on the material family knowledge graph, and generate a composite test parameter set including rainstorm impact waveform, temperature cycle gradient and structural deformation rate. The dynamic load application module 103 is used to synchronously apply dynamic loads in the climate simulation test chamber according to the composite test parameter set. The acceptance module 104 is used to locate high-risk leakage areas based on test results using a spatiotemporal correlation algorithm, and generate a decision plan that includes maintenance priorities, carbon footprint assessment, and disaster resistance prediction by combining the material durability database and regional extreme climate data.
[0022] Optionally, such as Figure 2 As shown, this application provides a method for supervising and accepting building waterproofing, including: S201 uses a microchannel impedance tomography strategy to scan waterproof material samples, extract pore connectivity paths and defect distribution characteristics, and construct a three-dimensional permeation network model that includes pore size, connectivity paths and defect distribution. S202, based on the material family knowledge graph, maps the water pressure and crack propagation laws of concrete structures in historical engineering projects to the current material's permeation network feature space, generating a composite test parameter set that includes rainstorm impact waveforms, temperature cycling gradients, and structural deformation rates; S203, based on the composite test parameter set, synchronously applies dynamic loads in the climate simulation test chamber; S204. Based on the test results, a spatiotemporal correlation algorithm is used to locate high-risk leakage areas. Combined with the material durability database and regional extreme climate data, a decision-making scheme is generated that includes maintenance priorities, carbon footprint assessment, and disaster resilience prediction.
[0023] Based on the embodiments provided in this application, a three-dimensional permeable network model and a material family knowledge graph are constructed to achieve multi-field coupled dynamic loading of water pressure, temperature, and mechanical stress, overcoming the shortcomings of traditional static detection in simulating complex actual working conditions. By combining spatiotemporal correlation algorithms with regional extreme climate data, maintenance plans including carbon footprint assessment are generated, solving the decision-making bias problem caused by the reliance on experience in traditional methods. By integrating a material durability database, full-chain performance prediction from material production to service maintenance is achieved, filling the gap in existing technologies regarding the lack of long-term disaster resistance analysis.
[0024] Furthermore, after constructing the three-dimensional penetration network model, a defect enhancement step is also included: A fractal generative adversarial network is used to synthesize hidden defects. The generator input includes: crack fractal growth paths generated based on the Monte Carlo method; a library of pore connectivity patterns from historical leakage cases; and an interface debonding deformation field constrained by material viscoelastic parameters. Defect samples are reconstructed using a variational autoencoder, and Darcy's law permeability constraint is applied to the latent space. A 3D graph attention network is used to generate a penetration vulnerability heatmap, marking the 3D influence range of high-risk defect clusters.
[0025] In this embodiment, the constraint equation for defect fractal growth is:
[0026] in, The fractal dimension of the crack represents the geometric complexity of the defect morphology. The specific value is calculated based on the fractal growth path of the crack. The equivalent permeability resistance coefficient of the pores is calculated using Darcy's law; The viscoelastic coupling coefficient (value 1.5-3.0) is determined by dynamic thermomechanical analysis (DMA) experiments and reflects the hysteresis effect of interfacial debonding. It is the viscoelastic stress tensor, reflecting the debonding deformation at the material interface; This is the viscoelastic stress field gradient, used to describe the change in stress distribution within a material; Represents a time variable; The defect generation rate factor (value 0.2-0.8) is optimized using gradient descent based on material type. Higher values are used for brittle materials (such as ceramic tiles), and lower values are used for flexible roll materials. This is the throat connectivity matrix, used to describe the flow characteristics of the pore network; is the norm of the throat connectivity matrix, used to describe the flow characteristics of pore networks; The critical leakage threshold is set according to the material's impermeability level; This is a time decay factor that controls the stability of the defect generation process. Its specific value is set based on experimental data or experience. For example, Take 0.1; This represents an exponential function.
[0027] Through fractal dimension With permeation resistance The dynamic constraints accurately reproduce the topological features of the real leakage path in the virtual defect generation, solving the problem of large deviation between the morphology of traditional artificial defects and natural defects, and improving the physical rationality of hidden defect detection.
[0028] Based on the embodiments provided in this application, physical rationality is maintained during the virtual defect synthesis process by using fractal generative adversarial networks and Darcy's law constraints, which significantly improves the sensitivity of identifying micron-level hidden defects.
[0029] Furthermore, the implementation of Darcy's law penetration constraint includes: A pore network flow model is established, where pores are abstracted as nodes and throats are connected as weighted edges; The equivalent permeability resistance coefficient of each node is calculated using a graph convolutional network. The variance of the permeation resistance of the reconstructed sample is constrained to not exceed a preset proportion of the original sample, and the diameter of the largest connected pore cluster is less than the critical leakage threshold. The preset proportion can be 12%, 20%, etc.
[0030] Furthermore, the construction of the materials family knowledge graph includes: Define multidimensional node attributes: Material gene node: Fourier descriptor of chemical composition, extrusion molding process parameter curve; Failure characteristic node: Acoustic emission energy release rate spectrum, hot spot diffusion path topology; Environmental load node: regional 50-year rainstorm intensity and duration distribution characteristics; Establish heterogeneous graph relationships: process defect edge: quantify the nonlinear correlation between extrusion temperature fluctuation and pore connectivity; load failure edge: encode the mapping weights between water pressure pulse spectrum characteristics and crack propagation rate; When introducing new materials, their UV curing curves are aligned with the thermoforming parameter space of existing materials through a capsule network.
[0031] Furthermore, the parameter space alignment method includes: A dual-flow feature extraction network was constructed to process the rheological curves of photocurable materials and the melt index spectra of polymer materials. The Wasserstein distance between characteristic distributions was calculated using optimal transport theory. When the distance exceeds the threshold, a cross-modal attention mechanism is triggered to generate a virtual transition material sample set.
[0032] Furthermore, the generation of the rainstorm impact waveform includes: A physical constraint generative adversarial network is constructed: the generator's hidden layer embeds the modified Navier-Stokes equation residuals to simulate the turbulent energy dissipation process; the discriminator introduces dynamic contact angle measurements to filter out abnormal waveforms that do not conform to the material's wetting characteristics. A pulse sequence is generated by a time-temperature controller: the main frequency is set to 1.3 to 1.7 times the material's natural frequency to excite the resonance effect; alternating stress is superimposed on the hot spot area identified by the infrared thermogram, and the amplitude is positively correlated with the local temperature rise rate.
[0033] In this embodiment of the application, the turbulence waveform generation criterion includes:
[0034] in, For turbulent kinetic energy tensor, it describes the transient energy distribution of rainstorm impact; The dynamic contact angle reflects the wettability of the material surface; The velocity field vector is solved using the Navier-Stokes equations. The energy cascade process controlling waveform generation is the turbulent dissipation rate. The contact angle is used as a reference, and is calibrated based on the hydrophilicity or hydrophobicity of the material. The contact angle tolerance threshold is used to filter out abnormal waveforms. The integration region (the volume of the specimen is taken as the value) ), representing the three-dimensional spatial range of the waterproof material sample; The dynamic viscosity coefficient (value 1.0 × 10⁻⁶) -3 Pa·s), dynamically adjusted according to the experimental water temperature to simulate the characteristics of actual rainwater; The contact angle adjustment factor (value 0.5-2.0) is calibrated through material wetting experiments to control the filtering intensity of the discriminator for abnormal waveforms. The Laplace term of the velocity field is discretized using the finite volume method in numerical solutions.
[0035] It should be noted that the integrand is ;in, This is a viscous dissipation term, characterizing the loss of turbulent energy due to fluid shear stress (unit: Pa / m). This is the inertial term, describing the convective transfer of turbulent kinetic energy (unit: N / m). Turbulent dissipation rate (unit: m / s) is used to convert molecular dynamics energy into a dimensionless ratio, ensuring the consistency of physical dimensions in the integral term.
[0036] After integrating the volume, It represents the average turbulence intensity per unit volume (unit: Pa), directly reflecting the equivalent dynamic pressure distribution on the material surface caused by rainstorm impact.
[0037] By coupling the Navier-Stokes equations with contact angle constraints The generated rainstorm impact waveform conforms to the laws of fluid mechanics and is adapted to the wetting characteristics of the material surface, solving the problem that traditional hydrostatic tests cannot simulate the transient impact of real rainstorms. Specifically, the spatial distribution of turbulent energy in the three-dimensional structure of the material is quantified by volume integration, overcoming the deficiency of traditional point pressure sensors in reflecting the overall load characteristics. Normalization processing makes the test results of specimens of different sizes comparable. For example, for roofing rolls (2mm thick), Ω is taken as the volume of the thin layer within 0.5mm below the surface; for waterproof coatings (5mm thick), the entire thickness is integrated.
[0038] The integral results and the material permeability resistance are solved simultaneously, which can be directly used to optimize the rainstorm waveform parameters, ensuring the dynamic similarity between the experimental load and the actual wind and rain load.
[0039] Based on the embodiments provided in this application, a waveform generation technique based on physical constraints is used to embed material wetting characteristic criteria in turbulence simulation to ensure that the impact waveform conforms to the transient characteristics of actual rainstorms.
[0040] Furthermore, such as Figure 3 As shown, the following is executed when applying a temperature cyclic gradient: S301, establish an exponential correlation model between acoustic emission energy and crack propagation rate; S302, when the real-time crack rate exceeds the predicted value by 15%, an adaptive temperature control strategy is activated: a low-temperature constraint field of -5℃ to 10℃ is applied within a 0.5mm radius of the crack tip; the heating power in the surrounding 2mm area is increased to 80 to 250W / cm² according to a hyperbolic function. S303: After each cycle, residual strain is measured based on digital image correlation method, temperature control parameters are updated, and crack suppression effect is verified.
[0041] In this embodiment of the application, the crack propagation-temperature control coupling model is as follows:
[0042] in, Acoustic emission energy density, characterizing the instantaneous energy release during crack propagation; The real-time crack propagation rate is calculated from DIC measurement data; To constrain the temperature field locally, the low-temperature region suppresses the activity of crack tips; This represents the radial coordinate from the crack tip; The characteristic length of the heat-affected zone is used to control the distribution of heating power. The thermal activation energy of the material reflects its temperature-sensitive properties. The crack propagation sensitivity coefficient (value ranges from 0.05 to 0.15) is calculated based on the fracture toughness of the material, with a lower value for high-toughness materials. The activation energy adjustment constant (valued at 8.314 J / mol·K) is dynamically corrected in conjunction with the material's thermal expansion coefficient. The origin of the crack tip coordinates is located in real time using digital image correlation technology, with an accuracy of ±0.1mm; It is a hyperbolic secant square function that describes the spatial distribution of heating power and suppresses excessive diffusion in the heat-affected zone.
[0043] pass The function controls the local heating power distribution, forming a gradient temperature field at the crack tip, actively suppressing asymmetric crack propagation, and solving the problem that traditional uniform temperature control cannot accurately intervene in damage evolution.
[0044] Based on the embodiments provided in this application, a local low-temperature confinement field is formed at the crack tip through acoustic emission energy feedback and hyperbolic function temperature control strategy, thereby actively suppressing the propagation of asymmetric cracks.
[0045] Furthermore, when locating high-risk areas for leakage: The acoustic emission time spectrum was converted into a Mel-scale spectrogram to extract abrupt changes in energy release rate; the isotherm curvature anomaly region in the infrared thermogram was analyzed; and strain gradient distribution data from a distributed fiber optic sensor were fused. Spatial correlation weights are calculated using a 3D convolutional attention network. When the acoustic emission energy is greater than 5 kJ / m³ and accompanied by annular isotherm distribution, it is determined to be a penetrating leakage channel.
[0046] In this embodiment of the application, the multimodal leakage location criterion is:
[0047] in, The Mel-scale spectrogram matrix encodes the time-frequency features of acoustic emissions. The strain gradient tensor is acquired by a distributed fiber optic sensor. Use a 3D convolution kernel to extract spatial correlation features; This represents the isotherm curvature eigenvector of the infrared thermogram; The characteristic mapping of the strain gradient distribution; Attention weights reflect the priority of multimodal information fusion. The kernel index (values from 1 to N) represents feature extractors at different scales; N=8 is set according to the sensor array density. The attention weights (ranging from 0 to 1) are optimized through training with experimental data, and high weights are assigned to the region of abrupt changes in acoustic emission. The kernel is a 3D convolution (3×3×3), and prior knowledge of material defects (such as crack orientation preference) is injected during initialization; dynamic weights are used to... By integrating multi-source data from sound, heat, and force, sub-millimeter-level spatial positioning of leakage paths can be achieved, solving the problem that traditional single-modal detection is easily affected by environmental noise.
[0048] Based on the embodiments provided in this application, a three-dimensional convolutional attention network is used to associate acoustic, thermal, and mechanical multi-source information to achieve sub-millimeter-level spatial positioning accuracy for penetrating leakage channels.
[0049] Furthermore, when generating decision-making options, the following are included: Constructing a four-dimensional assessment space: Dimension 1: Construction cost; Dimension 2: Expected lifespan extension rate; Dimension 3: Carbon emissions throughout the entire life cycle; Dimension 4: Probability of failure due to extreme weather. A constrained non-dominated sorting genetic algorithm is used to generate the Pareto solution set; The optimal solution is recommended based on the improved TOPSIS method, and the weight matrix dynamically reflects the owner's environmental preference index.
[0050] In this embodiment, dimension one: construction cost includes material loss rate and labor efficiency coefficient; dimension two: expected life extension rate includes calculations based on accelerated aging tests and the Weibull model; dimension three: life cycle carbon emissions cover production energy consumption and maintenance and transportation carbon footprint; dimension four: extreme weather failure probability integrates regional rainstorm recurrence interval and material fatigue damage accumulation model. Based on the embodiments provided in this application, a four-dimensional evaluation space and a non-dominated sorting algorithm are used to quantify and balance cost, lifespan, environmental protection and disaster resistance indicators in the maintenance plan, thereby meeting the requirements of green building.
[0051] It should be noted that the embodiments implemented on the construction waterproofing supervision and acceptance system side in this application can be referenced with the embodiments implemented on the construction waterproofing supervision and acceptance method side, and will not be described in detail here.
[0052] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A building waterproofing supervision and acceptance system, characterized in that, include: The permeation network model construction module is used to scan waterproof material samples using a microchannel impedance tomography strategy, extract pore connectivity paths and defect distribution characteristics, and construct a three-dimensional permeation network model that includes pore size, connectivity paths, and defect distribution. The test parameter set generation module is used to map the water pressure and crack propagation laws of concrete structures in historical engineering projects to the current material's permeability network feature space based on the material family knowledge graph, and generate a composite test parameter set that includes rainstorm impact waveforms, temperature cycle gradients and structural deformation rates. The dynamic load application module is used to synchronously apply dynamic loads within the climate simulation test chamber according to the composite test parameter set. The acceptance module is used to locate high-risk leakage areas based on test results using a spatiotemporal correlation algorithm. It then combines a material durability database with regional extreme climate data to generate a decision-making plan that includes maintenance priorities, carbon footprint assessment, and disaster resilience prediction.
2. A method for supervising and accepting building waterproofing, characterized in that, include: By using a microchannel impedance tomography strategy, waterproof material samples were scanned to extract pore connectivity paths and defect distribution characteristics, and a three-dimensional permeation network model including pore size, connectivity paths, and defect distribution was constructed. Based on the material family knowledge graph, the water pressure and crack propagation laws of concrete structures in historical engineering projects are mapped to the current material permeation network feature space, generating a composite test parameter set that includes rainstorm impact waveforms, temperature cycling gradients, and structural deformation rates. Based on the composite test parameter set, dynamic loads are simultaneously applied inside the climate simulation test chamber; Based on the test results, high-risk leakage areas were located using a spatiotemporal correlation algorithm. Combined with a material durability database and regional extreme climate data, a decision-making scheme was generated that included maintenance priorities, carbon footprint assessment, and disaster resilience prediction.
3. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, Following the construction of the three-dimensional permeation network model, a defect enhancement step is also included: A fractal generative adversarial network is used to synthesize hidden defects. The generator input includes: crack fractal growth paths generated based on the Monte Carlo method; a library of pore connectivity patterns from historical leakage cases; and an interface debonding deformation field constrained by material viscoelastic parameters. Defect samples are reconstructed using a variational autoencoder, and Darcy's law permeability constraint is applied to the latent space. A 3D graph attention network is used to generate a penetration vulnerability heatmap, marking the 3D influence range of high-risk defect clusters.
4. The building waterproofing supervision and acceptance method according to claim 3, characterized in that, The implementation of the Darcy's law permeability constraint includes: A pore network flow model is established, where pores are abstracted as nodes and throats are connected as weighted edges; The equivalent permeability resistance coefficient of each node is calculated using a graph convolutional network. The variance of the permeation resistance of the reconstructed sample is constrained to not exceed the preset proportion of the original sample, and the diameter of the largest connected pore cluster is less than the critical leakage threshold.
5. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, The construction of the material family knowledge graph includes: Define multidimensional node attributes: Material gene node: Fourier descriptor of chemical composition, extrusion molding process parameter curve; Failure characteristic node: Acoustic emission energy release rate spectrum, hot spot diffusion path topology; Environmental load node: regional 50-year rainstorm intensity and duration distribution characteristics; Establish heterogeneous graph relationships: process defect edge: quantify the nonlinear correlation between extrusion temperature fluctuation and pore connectivity; load failure edge: encode the mapping weights between water pressure pulse spectrum characteristics and crack propagation rate; When introducing new materials, their UV curing curves are aligned with the thermoforming parameter space of existing materials through a capsule network.
6. The building waterproofing supervision and acceptance method according to claim 5, characterized in that, The parameter space alignment method includes: A dual-flow feature extraction network was constructed to process the rheological curves of photocurable materials and the melt index spectra of polymer materials. The Wasserstein distance between characteristic distributions was calculated using optimal transport theory. When the distance exceeds the threshold, a cross-modal attention mechanism is triggered to generate a virtual transition material sample set.
7. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, The generation of the rainstorm impact waveform includes: A physical constraint generative adversarial network is constructed: the generator's hidden layer embeds the modified Navier-Stokes equation residuals to simulate the turbulent energy dissipation process; the discriminator introduces dynamic contact angle measurements to filter out abnormal waveforms that do not conform to the material's wetting characteristics. A pulse sequence is generated by a time-temperature controller: the main frequency is set to 1.3 to 1.7 times the material's natural frequency to excite the resonance effect; alternating stress is superimposed on the hot spot area identified by the infrared thermogram, and the amplitude is positively correlated with the local temperature rise rate.
8. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, Execute when applying a temperature cyclic gradient: An exponential correlation model between acoustic emission energy and crack propagation rate was established; When the real-time crack rate exceeds the predicted value by 15%, an adaptive temperature control strategy is activated: a low-temperature constraint field of -5℃ to 10℃ is applied within a 0.5mm radius of the crack tip; the heating power in the surrounding 2mm area is increased to 80 to 250W / cm² according to a hyperbolic function. After each cycle, residual strain is measured using digital image correlation to update temperature control parameters and verify crack suppression effect.
9. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, When locating high-risk areas for leakage: The acoustic emission time spectrum was converted into a Mel-scale spectrogram to extract abrupt changes in energy release rate; the isotherm curvature anomaly region in the infrared thermogram was analyzed; and strain gradient distribution data from a distributed fiber optic sensor were fused. Spatial correlation weights are calculated using a 3D convolutional attention network. When the acoustic emission energy is greater than 5 kJ / m³ and accompanied by annular isotherm distribution, it is determined to be a penetrating leakage channel.
10. The building waterproofing supervision and acceptance method according to claim 2, characterized in that, When generating decision-making options, the following are included: Constructing a four-dimensional assessment space: Dimension 1: Construction cost; Dimension 2: Expected lifespan extension rate; Dimension 3: Carbon emissions throughout the entire life cycle; Dimension 4: Probability of failure due to extreme weather. A constrained non-dominated sorting genetic algorithm is used to generate the Pareto solution set; The optimal solution is recommended based on the improved TOPSIS method, and the weight matrix dynamically reflects the owner's environmental protection preference index.