A method and system for detecting the strength of a colored stone metal tile

Abstract

The present application relates to the technical field of building material detection, in particular to a color stone metal tile strength detection method and system, comprising: obtaining surface reflection light distribution data and internal structure stress distribution data of the sample to be detected, analyzing surface coating uniformity and micro-morphology characteristics to generate surface quality evaluation data set; fusing surface data and internal stress data, constructing a comprehensive stress-strain analysis model, and calculating the theoretical deformation parameters and the limit bearing threshold of the sample under the preset load; according to the above calculation results, accurately guiding the loading position, force value and process in the subsequent physical strength detection. The method combines the surface microstate and internal structure information to carry out predictive modeling, realizing the transformation of the detection process from unified loading, post-judgment to predictive guidance and intelligent loading.

Description

Technical Field

[0001] This invention relates to the field of building material testing technology, and in particular to a method and system for testing the strength of colored stone metal tiles. Background Technology

[0002] Currently, the strength performance evaluation of colored stone metal tiles generally adopts direct physical and mechanical testing methods, such as bending or impact tests. These methods measure the final breaking strength or deformation through standardized loading procedures, which are destructive end-stage verifications. They cannot perform rapid, non-destructive strength screening and performance prediction on a large number of samples during production or on-site inspection, resulting in low testing efficiency and high costs. The core limitation of existing technologies lies in focusing only on macroscopic mechanical response results, lacking means to quantitatively analyze and correlate the intrinsic quality defects of the samples before applying load.

[0003] At the level of specific testing techniques, the evaluation of surface coatings largely relies on average thickness measurement or visual inspection, making it difficult to quantitatively characterize the microscopic uniformity of coating distribution and surface morphology. Furthermore, the measurement of internal residual stress is typically conducted independently. Existing technologies treat surface condition analysis and internal structure analysis as isolated processes, failing to establish a quantitative correlation model between these two aspects and the overall mechanical properties. This disconnect prevents the prediction of mechanical behavior based on the unique defect characteristics of individual samples, forcing subsequent physical strength testing to employ uniform, fixed loading strategies, lacking specificity and predictability.

[0004] A method is needed that can integrate surface microstructure and internal structural stress information in advance and build a predictive model based on this. This method should calculate the theoretical mechanical parameters of individual samples before implementing destructive physical testing, thereby guiding the generation of adaptive, non-standardized loading schemes and transforming the testing process from passive verification to proactive predictive guidance. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and to propose a method and system for testing the strength of colored stone metal tiles.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for testing the strength of colored stone metal tiles, comprising: Acquire surface reflected light distribution data and internal structural stress distribution data of the colored stone metal tile sample to be tested; Based on the surface reflected light distribution data, the surface coating uniformity and micromorphological characteristics of the tested colored stone metal tile sample are analyzed to generate a surface quality assessment dataset. By integrating the surface quality assessment dataset and the internal structural stress distribution data, a comprehensive stress-strain analysis model is constructed for the colored stone metal tile sample to be tested. Based on the comprehensive stress-strain analysis model, the theoretical deformation parameters and ultimate bearing capacity threshold of the tested colored stone metal tile sample under the preset load mode are calculated. The theoretical deformation parameters and the ultimate bearing threshold are used as standard judgment criteria to guide the loading strategy in the subsequent physical strength testing process of the colored stone metal tile sample to be tested.

[0007] As a further aspect of the present invention, the acquisition of surface reflected light distribution data and internal structural stress distribution data of the colored stone metal tile sample to be tested specifically includes: The surface of the colored stone metal tile sample under test is illuminated by a multi-band controllable light source, and the reflected light signal is received by a high-resolution spectral imaging sensor. The surface reflected light intensity and spatial distribution information under different bands are collected to form the surface reflected light distribution data. A laser speckle field is projected onto the sample of colored stone metal tile to be tested using a digital speckle interferometry device, and the initial speckle pattern of the sample of colored stone metal tile to be tested under no-load conditions is recorded. A known small preload is applied to the sample of colored stone metal tile to be tested, and the current speckle pattern of the sample of colored stone metal tile under the preload state is recorded again; The current speckle pattern and the initial speckle pattern are subjected to digital correlation calculation to solve the microscopic displacement field of the sample surface caused by the preload. Based on the constitutive relationship of material mechanics, the internal structural stress distribution data of the shallow surface layer of the sample is deduced from the microscopic displacement field.

[0008] As a further aspect of the present invention, the step of analyzing the surface coating uniformity and microstructure characteristics of the tested colored stone metal tile sample based on the surface reflected light distribution data to generate a surface quality assessment dataset specifically includes: From the surface reflected light distribution data, extract the reflectance spectrum curves of different spatial locations in multiple bands; For each of the aforementioned reflectance spectrum curves, characteristic peaks are identified and peak areas are calculated. The peak area ratio of each characteristic peak is matched with the spectral feature library of the standard colored stone coating to obtain the coating composition consistency index at each location point. The spatial variation coefficient of the coating composition consistency index at all locations is calculated, and the spatial variation coefficient is used to quantitatively characterize the surface coating uniformity. Simultaneously, spatial frequency analysis is performed on the surface reflected light distribution data to separate the low-frequency component characterizing macroscopic smoothness and the high-frequency component characterizing microscopic roughness. Calculate the energy intensity and distribution entropy value of the high-frequency component, and use the energy intensity and distribution entropy value together as a quantitative parameter to describe the surface micromorphological characteristics; The spatial variation coefficient of the surface coating uniformity, as well as the energy intensity and distribution entropy value of the surface micromorphology features, are integrated into structured data to generate the surface quality assessment dataset.

[0009] As a further aspect of the present invention, the method of integrating the surface quality assessment dataset and the internal structural stress distribution data to construct a comprehensive stress-strain analysis model for the tested colored stone metal tile sample is as follows: A parametric finite element mesh model describing the laminated structure of colored stone metal tiles is established. The parametric finite element mesh model includes a metal substrate layer, an adhesive layer, and a colored stone coating. The spatial variation coefficient of surface coating uniformity in the surface quality assessment dataset is mapped to the spatial distribution function of the material properties of the colored stone coating element in the parameterized finite element mesh model. The energy intensity and distribution entropy value of the surface micromorphology features are converted into the boundary geometric roughness parameters of the colored stone coating surface in the parameterized finite element mesh model. The internal structural stress distribution data is used as the initial stress field and imported into the corresponding position of the parameterized finite element mesh model. Based on the imported material property spatial distribution function, the boundary geometric roughness parameters, and the initial stress field, the calculation engine of the parameterized finite element mesh model is run to construct the comprehensive stress-strain analysis model that can reflect the material non-uniformity and initial stress state.

[0010] As a further aspect of the present invention, the theoretical deformation parameters and ultimate bearing capacity threshold of the tested colored stone metal tile sample under a preset load mode are calculated based on the comprehensive stress-strain analysis model, specifically as follows: Define one or more of the preset load modes, including uniformly distributed surface load, concentrated line load, or dynamic wind pressure load; Each of the preset load modes is applied to the comprehensive stress-strain analysis model as a boundary condition. By performing nonlinear static solution or dynamic time history analysis using the comprehensive stress-strain analysis model, the stress field evolution cloud map, strain field evolution cloud map, and overall deformation curve of the tested colored stone metal tile sample during the entire load application process are calculated. Feature points are extracted from the overall deformation curve, including the proportional limit point, yield point and maximum load point. The load value corresponding to the proportional limit point is recorded as the initial failure threshold, and the load value corresponding to the maximum load point is recorded as the final failure threshold. The theoretical deformation parameters include the overall deflection value, the maximum strain value and its location predicted by the comprehensive stress-strain analysis model under the specified load level; The ultimate load-bearing threshold is a load range, with its lower limit being the initial failure threshold and its upper limit being the final failure threshold.

[0011] As a further aspect of the present invention, the theoretical deformation parameters and the ultimate bearing capacity threshold are used as standard judgment criteria to guide the loading strategy in the subsequent physical strength testing process of the colored stone metal tile sample to be tested, specifically as follows: Based on the ultimate bearing threshold range, the load spectrum of the physical strength testing machine is set, wherein the peak load of the load spectrum is between the initial failure threshold and the final failure threshold, and includes multiple increasing load steps; Based on the predicted overall deflection value and its changing trend in the theoretical deformation parameters, a displacement monitoring and early warning line for the physical strength testing machine during the loading process is set, and the displacement monitoring and early warning line is slightly lower than the predicted overall deflection value under the corresponding load. During the physical strength testing process, the physical strength testing machine is controlled to apply loads according to the load spectrum; The actual deformation of the colored stone metal tile sample under test is monitored in real time. When the actual deformation approaches or reaches the displacement monitoring warning line under the corresponding load step, the loading rate slowing down or load holding command is triggered. The load is continuously applied until the sample is damaged or a preset termination condition is reached, and the actual damage load and damage mode are recorded.

[0012] As a further aspect of the present invention, after the physical strength detection process, a step of feedback correction of the comprehensive stress-strain analysis model based on the detection results is further included: Acquire the actual load-displacement curve recorded during the physical strength test, as well as image information when the sample is damaged; Extract the actual proportional limit load, maximum load, and corresponding displacement value from the actual load-displacement full-process curve; The actual proportional limit load and maximum load are compared with the initial failure threshold and final failure threshold predicted by the comprehensive stress-strain analysis model, and the prediction error is calculated. Based on the prediction error, the values ​​of key material parameters in the comprehensive stress-strain analysis model are adjusted in reverse. The key material parameters include the equivalent elastic modulus of the colored stone coating and the shear strength of the adhesive layer. The integrated stress-strain analysis model is updated using the adjusted key material parameters, and the updated model is validated using another set of independent validation sample data to complete the feedback correction closed loop of the model.

[0013] As a further aspect of the present invention, before illuminating the surface of the colored stone metal tile sample to be tested with a multi-band controllable light source, the method further includes a pre-processing and positioning step for the colored stone metal tile sample to be tested: The sample of the colored stone metal tile to be tested is placed in a testing station with a constant temperature and humidity environment and left to stand for a preset time to achieve thermal and humidity equilibrium with the environment. The edge contour of the colored stone metal tile sample to be tested is identified and the preset positioning mark points are matched using a visual positioning system. Based on the identified edge contours and the positioning markers, the spatial pose of the colored stone metal tile sample to be tested relative to the multi-band controllable light source and the digital speckle interferometry measurement device is calculated. Adjust the illumination angle of the multi-band controllable light source and the projection angle of the digital speckle interferometry measurement device to ensure that the light source and the measurement beam are perpendicularly incident on the center of the test area of ​​the colored stone metal tile sample.

[0014] As a further aspect of the present invention, when defining one or more of the preset load modes, the step of equivalent simplification based on the load spectrum of the actual service environment is included: Long-term load monitoring data of colored stone metal tiles under typical building roofing environments were collected. The long-term load monitoring data included wind pressure time history data, snow load data, and temperature difference change data. Extreme value statistical analysis was performed on the wind pressure time history data to extract representative and enveloping wind pressure distribution patterns and peak values; The snow load data is converted into a uniformly distributed surface load according to the roof slope. The difference in thermal expansion coefficients between the metal substrate and the colored stone coating caused by the temperature difference change data was analyzed, and the resulting equivalent temperature stress load was calculated. The wind pressure distribution pattern, the converted uniformly distributed snow load, and the equivalent temperature stress load are combined to form the load combination condition of the preset load pattern.

[0015] As a further aspect of the present invention, the present invention also includes a strength testing system for colored stone metal tiles, the system including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of the strength testing method for colored stone metal tiles as described above.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: By acquiring and analyzing surface reflected light distribution data, the microscopic uniformity and morphological characteristics of the coating are transformed into a quantifiable evaluation dataset. This technique can non-contactly and with high sensitivity capture coating thickness variations, uneven particle distribution, and microscopic defect textures that are difficult to detect with the naked eye. These surface features directly affect stress concentration and crack initiation. Introducing them as key parameters into the strength evaluation system allows for the early identification of potential weak areas due to poor surface conditions before any physical load is applied, thus extending the capabilities from macroscopic mechanical testing to early warning of microscopic conditions.

[0017] By fusing surface quality assessment datasets with internal structural stress distribution data, a comprehensive stress-strain analysis model was constructed. This model integrates multi-source information from the surface and substrate, enabling a more realistic simulation of the overall mechanical response of the tile under load. The theoretical deformation parameters and ultimate bearing capacity thresholds calculated based on this model provide individualized prediction benchmarks for subsequent physical strength testing. This allows for dynamic adjustment of the loading location, force magnitude, loading rate, or termination conditions during actual testing based on the unique model prediction results for each sample. This transforms physical testing from a fixed-procedure verification process into a targeted and precise verification process driven by digital model predictions, enhancing the intelligence and specificity of the testing. Attached Figure Description

[0018] Figure 1 This is a flowchart of the strength testing method for colored stone metal tiles according to the present invention; Figure 2 A flowchart for acquiring surface reflected light and internal stress data; Figure 3 Flowchart generated for surface quality assessment dataset; Figure 4 A statistical analysis chart of wind pressure data in the strength testing project of colored stone metal tiles; Figure 5 Loading strategy and monitoring and early warning diagram for physical strength testing of colored stone metal tiles. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] See Figure 1 The method involves acquiring surface reflected light distribution data and internal structural stress distribution data of the colored stone metal tile sample to be tested; based on the surface reflected light distribution data, analyzing the surface coating uniformity and microstructure characteristics of the sample to generate a surface quality assessment dataset; fusing the surface quality assessment dataset and internal structural stress distribution data to construct a comprehensive stress-strain analysis model for the sample; calculating the theoretical deformation parameters and ultimate bearing threshold of the sample under a preset load mode based on the comprehensive stress-strain analysis model; and using the theoretical deformation parameters and ultimate bearing threshold as standard judgment criteria to guide the loading strategy in the subsequent physical strength testing process of the sample.

[0022] In one embodiment of the present invention, see [reference] Figure 2The sample of the colored stone metal tile to be tested is placed in a testing station with a constant temperature and humidity environment and left to stand for a preset time to achieve thermal and humidity equilibrium with the environment. The edge contour and preset positioning marks of the sample are identified using a visual positioning system. Based on the identified edge contour and positioning marks, the spatial pose of the sample relative to the multi-band controllable light source and the digital speckle interferometry device is calculated. The illumination angle of the multi-band controllable light source and the projection angle of the digital speckle interferometry device are adjusted to ensure that the light source and the measurement beam are perpendicularly incident on the center of the test area of ​​the sample. The surface of the sample is illuminated by the multi-band controllable light source, and the reflected light signal is received by a high-resolution spectral imaging sensor. The surface reflected light intensity and spatial distribution information at different wavelengths are collected to form surface reflected light distribution data. A laser speckle field is projected onto the sample using the digital speckle interferometry device, and the initial speckle pattern of the sample under no-load conditions is recorded. A known small preload is applied to the sample of colored stone metal tile to be tested, and the current speckle pattern of the sample under the preload is recorded again. The current speckle pattern is compared with the initial speckle pattern by digital correlation to calculate the microscopic displacement field on the sample surface caused by the preload. Based on the constitutive relationship of material mechanics, the internal structural stress distribution data of the shallow surface layer of the sample is deduced from the microscopic displacement field.

[0023] In practice, the test samples undergo standardized preprocessing and precise spatial positioning, based on which surface reflected light distribution data and internal structural stress distribution data are collected. The colored stone metal tile samples to be tested are placed in a constant temperature and humidity testing station with a temperature controlled at 23 degrees Celsius and a relative humidity maintained at 50%, and left to stand for 120 minutes to achieve thermal and humidity equilibrium between the sample material and the testing environment. After the visual positioning system is activated, its high-resolution industrial camera captures a complete top-down image of the sample. The image processing algorithm automatically identifies the four straight edge contours of the sample and the center pixel coordinates of the four circular positioning markers pre-printed on the sample surface. In some embodiments, based on the identified edge contours and the pixel coordinates of the positioning markers, combined with the pre-calibrated camera internal parameters and the world coordinate system of the testing station, the three-dimensional spatial pose of the colored stone metal tile sample under test in the world coordinate system is calculated, including the normal vector of the sample plane and its translation and rotation relationships relative to the multi-band controllable light source output port and the laser emitter of the digital speckle interferometry device. The robotic arm supporting the multi-band controllable light source and the gimbal of the digital speckle interferometry measurement device are automatically adjusted according to the received spatial pose data. The illumination axis of the multi-band controllable light source and the laser projection axis of the digital speckle interferometry measurement device are both adjusted to be parallel to the calculated normal vector of the sample plane, ensuring that the beam is incident perpendicularly and that the center of the light spot coincides with the center of the sample area to be measured.

[0024] In practice, after adjustments are made, a multi-band controllable light source sequentially emits narrow-band light with wavelengths of 450 nm, 550 nm, and 650 nm to illuminate the sample surface according to a preset program. A high-resolution spectral imaging sensor simultaneously acquires the reflected light signal under each band, records the light intensity value of each pixel, and finally generates a three-dimensional data matrix containing spatial coordinates and band information. This matrix represents the surface reflected light distribution data. A digital speckle interferometry device projects a laser speckle field with a wavelength of 632.8 nm onto the same test area. A high-precision CCD camera records the speckle interference pattern on the surface of the colored stone metal tile sample under test, which is not subjected to any external force. This pattern is stored as the initial speckle pattern.

[0025] Optionally, a uniformly distributed air pressure of 0.5 kPa is applied as a small preload to the colored stone metal tile sample under test using a precision pneumatic loading device. After the load stabilizes, the high-precision CCD camera of the digital speckle interferometry device re-captures and records the current speckle pattern. The current speckle pattern and the initial speckle pattern are imported into digital image correlation processing software. The software performs sub-pixel precision digital correlation calculations on the two images to calculate the full-field two-dimensional microscopic displacement vector generated on the sample surface due to the applied 0.5 kPa preload, with a displacement resolution of 0.1 pixels. Based on the constitutive relationship of linear elasticity and the plane stress assumption, combined with the known Poisson's ratio of the colored stone metal tile material, the internal structural stress distribution data of the shallow surface layer of the sample is deduced from the calculated microscopic displacement field through spatial differentiation operations. This data includes the normal stress and shear stress components at each point.

[0026] It is understandable that the aforementioned preprocessing and positioning operations ensure that all optical measurements are performed under uniform and stable reference conditions. The introduction of preload enables digital speckle interferometry to sensitively capture the material's response at extremely low stress levels, thereby retrieving the initial internal structural stress state. In practical implementation, the acquisition of surface reflected light distribution data depends on the spectral purity of the multi-band controllable light source and the sensitivity of the high-resolution spectral imaging sensor. The data quality can be evaluated by calculating the signal-to-noise ratio (SNR) of the image. The SNR is calculated as follows: in: Indicates the signal-to-noise ratio. Indicates the average power of the signal. This represents the average power of the noise. Only when the signal-to-noise ratio is greater than a preset threshold is the collected surface reflection light distribution data considered valid and allowed to proceed to the next analysis step.

[0027] In one embodiment of the present invention, see [reference] Figure 3From the surface reflected light distribution data, reflectance spectral curves at different spatial locations in multiple wavelength bands are extracted. For each reflectance spectral curve, characteristic peaks are identified and peak areas are calculated. The peak area ratio of each characteristic peak is matched with the spectral feature library of standard colored stone coatings to obtain the coating composition consistency index at each location. The spatial variation coefficient of the coating composition consistency index at all locations is calculated; this spatial variation coefficient is used to quantitatively characterize the surface coating uniformity. Simultaneously, spatial frequency analysis is performed on the surface reflected light distribution data to separate the low-frequency component characterizing macroscopic smoothness and the high-frequency component characterizing microscopic roughness. The energy intensity and distribution entropy value of the high-frequency component are calculated and used together as quantitative parameters describing the surface micromorphological characteristics. The spatial variation coefficient of surface coating uniformity, along with the energy intensity and distribution entropy values ​​of surface micromorphological characteristics, are integrated into structured data to generate a surface quality assessment dataset.

[0028] In practice, the collected surface reflected light distribution data is deeply analyzed to generate a surface quality assessment dataset. This data is a three-dimensional matrix containing spatial coordinates and multi-band light intensity information. Reflectance spectral curves for different spatial locations in multiple bands are extracted from the surface reflected light distribution data. This is achieved by traversing the light intensity sequence of each pixel in the data matrix along the wavelength dimension. Each location corresponds to a discrete reflection spectral curve composed of light intensity values ​​at 450 nm, 550 nm, and 650 nm bands. For each reflection spectral curve, characteristic peaks are identified and peak areas are calculated. The characteristic peak identification algorithm uses a sliding window and local maximum detection to determine the positions of prominent peaks in the spectral curve. Peak area calculation employs a numerical integration method to integrate the band range covered by each identified characteristic peak.

[0029] In some embodiments, the peak area ratios of each characteristic peak are matched with a spectral feature library of standard colored stone coatings. This library pre-stores the range of characteristic peak area ratios for qualified colored stone coating samples in the same wavelength band. The matching process calculates the Euclidean distance between the peak area ratios of the current curve and the reference values ​​in the feature library. After normalization, the distance values ​​are converted into a coating composition consistency index ranging from zero to one. The closer the coating composition consistency index is to one, the more consistent the coating composition is with the standard. The spatial variation coefficient of the coating composition consistency index at all locations is calculated. This spatial variation coefficient is used to quantify the surface coating uniformity. The calculation of the spatial variation coefficient involves first obtaining the arithmetic mean and standard deviation of the coating composition consistency index at all locations, and then dividing the standard deviation by the arithmetic mean.

[0030] Spatial frequency analysis was performed on the surface reflected light distribution data. This analysis employed a two-dimensional Fourier transform to convert the data from the spatial domain to the frequency domain, obtaining the corresponding two-dimensional spectrum. Low-frequency components characterizing macroscopic smoothness and high-frequency components characterizing microscopic roughness were separated. This separation was achieved by designing a two-dimensional digital filter. The cutoff frequency of this filter was chosen based on prior knowledge of the surface characteristics of the colored stone metal tile. The remaining components after filtering out those below the cutoff frequency in the spectrum were identified as high-frequency components. The energy intensity and distribution entropy of the high-frequency components were calculated. The energy intensity was obtained by summing the squares of all amplitudes in the high-frequency component spectrum, while the distribution entropy was calculated by analyzing the probability distribution characteristics of the high-frequency components in the spatial frequency domain.

[0031] The calculation of the distribution entropy requires first normalizing the amplitude of the high-frequency component spectrum to a probability distribution, as expressed by the following formula: in: This represents the distribution entropy value of the high-frequency components. This represents the total number of discrete frequency units in the high-frequency spectrum. Indicates the first The normalized amplitude probability of each frequency unit is obtained by dividing the amplitude of that unit by the sum of the amplitudes of all units. Energy intensity and distribution entropy are used together as quantitative parameters to describe the surface microstructure characteristics. The spatial variation coefficient of surface coating uniformity, along with the energy intensity and distribution entropy values ​​of surface microstructure characteristics, are integrated into structured data to generate a surface quality assessment dataset. This structured data is stored in JSON format or a database table, containing fields to record the spatial variation coefficient, energy intensity, and distribution entropy values.

[0032] In some embodiments, the characteristic peak identification step can integrate an automatic baseline correction function to eliminate the influence of spectral background noise. Automatic baseline correction is achieved by fitting the lower envelope of the spectral curve and subtracting this envelope from the original curve. Optionally, the cutoff frequency in the spatial frequency analysis can be dynamically adjusted according to the surface texture period of the specific colored stone metal tile product. The adjustment is based on the typical texture spacing obtained by pre-microscopic measurement of the standard sample surface. Optionally, in addition to using Euclidean distance matching, the coating composition consistency index can also be calculated using the correlation coefficient method, obtaining the index value by calculating the Pearson correlation coefficient between the current spectral curve and the feature library reference curve. It can be understood that the generation of the surface quality assessment dataset is a fully automated data processing flow executed by dedicated analysis software. After reading the surface reflected light distribution data file, the software sequentially calls the spectral analysis module and the spatial frequency analysis module, ultimately outputting a structured surface quality assessment dataset file for subsequent steps.

[0033] In one embodiment of the present invention, a parametric finite element mesh model describing the laminated structure of colored stone metal tiles is established. The parametric finite element mesh model includes a metal substrate layer, an adhesive layer, and a colored stone coating. The spatial variation coefficient of surface coating uniformity in the surface quality assessment dataset is mapped to the spatial distribution function of material properties of the colored stone coating elements in the parametric finite element mesh model. The energy intensity and distribution entropy values ​​of the surface micromorphology features are transformed into boundary geometric roughness parameters of the colored stone coating surface in the parametric finite element mesh model. The internal structural stress distribution data is used as the initial stress field and imported into the corresponding position of the parametric finite element mesh model. Based on the imported spatial distribution function of material properties, boundary geometric roughness parameters, and initial stress field, the calculation engine of the parametric finite element mesh model is run to construct a comprehensive stress-strain analysis model that reflects material non-uniformity and initial stress state.

[0034] In practical implementation, a comprehensive stress-strain analysis model is constructed by integrating multi-source data. The construction of this model begins with establishing a parametric finite element mesh model describing the laminated structure of the colored stone metal tile. Using the preprocessor module of commercial finite element software, the parametric finite element mesh model is established. This model clearly defines three layers: a metal substrate layer, an adhesive layer, and a colored stone coating. These layers are connected via shared nodes. Shell or solid elements suitable for laminate analysis are selected, and the model's geometric dimensions strictly correspond to the actual dimensions of the colored stone metal tile sample under test. In some embodiments, the spatial variation coefficient of the surface coating uniformity in the surface quality assessment dataset is mapped to a spatial distribution function of the material properties of the colored stone coating elements in the parametric finite element mesh model. Specifically, the mapping process involves reading the spatial variation coefficient values ​​recorded in the surface quality assessment dataset. These values ​​are correlated with the spatial fluctuation of the colored stone coating's elastic modulus. A transformation function converts the spatial variation coefficient into an adjustment factor for the local elastic modulus of each colored stone coating element. The transformation function is defined as follows: in: Indicates the position coordinates of the colored stone coating. The local equivalent elastic modulus at that point This is the reference value for the nominal elastic modulus of the colored stone coating. It is a material sensitivity coefficient. It is the spatial variation coefficient of surface coating uniformity in the surface quality assessment dataset. It is a location-dependent random field function used to introduce a random distribution within the model that conforms to the statistical characteristics of spatial coefficient of variation. The energy intensity and distribution entropy values ​​of the surface microstructure features are transformed into boundary geometric roughness parameters of the colored stone coating surface in the parameterized finite element mesh model. The transformation operation maps the energy intensity values ​​to the amplitude parameters of the surface roughness and the distribution entropy values ​​to the spatial distribution complexity parameters of the surface roughness. These two parameters are used together to define small geometric perturbations on the outer surface of the colored stone coating in the finite element model to simulate the actual microstructure.

[0035] In practice, the internal structural stress distribution data is used as the initial stress field and imported into the corresponding location of the parametric finite element mesh model. The internal structural stress distribution data is a data file containing the coordinates of each measurement point and its corresponding stress components. Using the initial condition definition function of the finite element software, an interpolation algorithm is employed to distribute the discrete measurement point stress data to each element integration point of the parametric finite element mesh model, thus establishing the initial stress state before the model solution begins. Based on the imported material property spatial distribution function, boundary geometric roughness parameters, and initial stress field, the computational engine of the parametric finite element mesh model is run to construct a comprehensive stress-strain analysis model that reflects both material inhomogeneity and the initial stress state.

[0036] It is understandable that the application of the material property spatial distribution function means that the colored stone coating is no longer a homogeneous material in the model; its elastic properties vary spatially according to the spatial variation coefficient of the surface coating homogeneity. The introduction of the boundary geometric roughness parameter is equivalent to applying microscale geometric irregularities with specific statistical characteristics to the surface of the colored stone coating. The introduction of the initial stress field ensures that the starting point of the model analysis includes residual stresses introduced by previous measurement or manufacturing processes. The construction of the integrated stress-strain analysis model is a data-driven process that unifies the surface quality assessment dataset from optical measurements with the internal structural stress distribution data from mechanical measurements into the same numerical analysis framework.

[0037] In some embodiments, a random field function is used to describe the spatial distribution function of material properties. The generation can be based on the Karun-Louis expansion method to ensure that the generated spatial distribution pattern has specified correlation length and variance characteristics, which match the actual spatial distribution characteristics of the surface coating uniformity. Optionally, the boundary geometric roughness parameters can be applied by defining small random displacements of surface nodes in the finite element software through user subroutines. The amplitude distribution of the displacements follows a probability density function determined by both energy intensity and distribution entropy. It can be understood that the successful operation of the comprehensive stress-strain analysis model depends on the accuracy and consistency of all input data. The operation process is executed by the finite element solver, completing steps such as model assembly, boundary condition application, and nonlinear equation solving, ultimately generating a complete numerical model that can be used to predict the stress, strain, and deformation response of colored stone metal tile samples under different external loads.

[0038] In one embodiment of the present invention, one or more preset load modes are defined, including uniformly distributed surface load, concentrated line load, or dynamic wind pressure load. Long-term load monitoring data of the colored stone metal tile under typical building roof conditions are collected. This long-term load monitoring data includes wind pressure time history data, snow load data, and temperature difference variation data. Extreme value statistical analysis is performed on the wind pressure time history data to extract representative and enveloping wind pressure distribution patterns and peak values. The snow load data is converted into a uniformly distributed surface load according to the roof slope. The difference in thermal expansion coefficients between the metal substrate and the colored stone coating caused by temperature difference variation data is analyzed to calculate the resulting equivalent temperature stress load. The wind pressure distribution pattern, the converted snow load uniformly distributed surface load, and the equivalent temperature stress load are combined to form a load combination condition of the preset load modes. Each preset load mode is applied to the comprehensive stress-strain analysis model as boundary conditions. Nonlinear static solutions or dynamic time history analyses are performed using the comprehensive stress-strain analysis model to calculate the stress field evolution cloud map, strain field evolution cloud map, and overall deformation curve of the tested colored stone metal tile sample throughout the entire load application process. Feature points are extracted from the overall deformation curve, including the proportional limit point, yield point, and maximum load point. The load value corresponding to the proportional limit point is recorded as the initial failure threshold, and the load value corresponding to the maximum load point is recorded as the final failure threshold. The theoretical deformation parameters include the overall deflection value, maximum strain value, and their location predicted by the comprehensive stress-strain analysis model at a specified load level. The ultimate bearing capacity threshold is a load range, with its lower limit being the initial failure threshold and its upper limit being the final failure threshold.

[0039] In the specific implementation, long-term load monitoring data of colored stone metal tiles under typical building roof environments were collected. This data came from one year of continuous records from pressure sensors, snow thickness monitors, and temperature sensors installed at specific locations on similar building roofs. Specifically, the long-term load monitoring data included wind pressure time-history data collected at a frequency of ten points per second, daily recorded snow load data, and hourly recorded temperature difference data. Extreme value statistical analysis was performed on the wind pressure time-history data. A generalized extreme value distribution model was used to fit the wind pressure peak value, extracting the wind pressure peak value corresponding to a 50-year return period and the typical spatial distribution pattern of wind pressure under that peak value. The snow load data was converted into a uniformly distributed surface load component perpendicular to the tile plane based on a 30-degree roof slope using a cosine relationship, resulting in the converted uniformly distributed snow load. The difference in thermal expansion coefficients between the metal substrate and the colored stone coating caused by temperature variation data was analyzed. The thermal expansion coefficient of the metal substrate is 23.2 x 10 to the power of -6 per degree Celsius, while that of the colored stone coating is 8.5 x 10 to the power of -6 per degree Celsius. The resulting equivalent temperature stress load was calculated.

[0040] In practical implementation, the wind pressure distribution pattern, the converted uniformly distributed snow load, and the equivalent temperature stress load are combined to form a load combination case with a preset load pattern. The load combination case follows the basic combination principles in the building structure load code. A typical example of a load combination case is shown in Table 1. One or more preset load patterns are defined. The preset load patterns include the combination case of uniformly distributed surface load and dynamic wind pressure load in Table 1, as well as the concentrated line load pattern, which simulates the local load generated by roof installation clamps or foot traffic. Each preset load pattern is applied to the comprehensive stress-strain analysis model as boundary conditions. For the uniformly distributed surface load, it is applied as pressure to the surface elements of the colored stone coating; for the dynamic wind pressure load, it is applied as a surface pressure function that varies with time; for the concentrated line load, it is applied as a line load to a specific set of nodes at the edge of the model, as shown in Table 1.

[0041] Table 1: Typical Load Combination Conditions Nonlinear statics or dynamic time history analysis was performed using a comprehensive stress-strain analysis model. For combined load conditions A and B, a nonlinear statics solver was used; for combined load condition C, a dynamic time history analyzer was used. This resulted in the calculation of stress field evolution contour maps, strain field evolution contour maps, and the overall deformation curve of the tested colored stone metal tile sample throughout the entire load application process. Feature points were extracted from the overall deformation curve, including the proportional limit point, yield point, and maximum load point. The proportional limit point was determined by observing the endpoint of the initial straight line segment of the load-displacement curve. The yield point was determined using the 0.2% residual strain method. The maximum load point was the peak point of the curve. The load value corresponding to the proportional limit point was recorded as the initial failure threshold, and the load value corresponding to the maximum load point was recorded as the final failure threshold.

[0042] In some embodiments, the calculation of equivalent temperature stress load needs to consider the effect of temperature gradient, and its equivalent stress calculation formula is expressed as: in: Represents equivalent temperature stress. This represents the difference in the coefficients of thermal expansion between the metal substrate and the colored stone coating. This represents the equivalent elastic modulus of the colored stone coating. This represents the absolute value of the maximum positive or negative temperature difference derived from long-term load monitoring data. Theoretical deformation parameters include the overall deflection value, maximum strain value, and their location predicted by the integrated stress-strain analysis model under a specified load level. The specified load level can be 50% or 80% of the lower limit of the ultimate bearing capacity threshold. The ultimate bearing capacity threshold is a load range, with its lower limit being the initial failure threshold and its upper limit being the final failure threshold. For example, if the initial failure threshold is 2.5 kPa and the final failure threshold is 3.8 kPa, then the ultimate bearing capacity threshold range is from 2.5 kPa to 3.8 kPa.

[0043] It is understandable that the definition of the preset load mode relies heavily on load monitoring and statistical analysis of the actual service environment. The construction of load combination conditions makes the simulation conditions of the comprehensive stress-strain analysis model closer to engineering reality. Solving through the comprehensive stress-strain analysis model can obtain full-field stress-strain information and the complete nonlinear deformation process that cannot be directly obtained through simple experiments. In some embodiments, the application function of dynamic wind pressure load can be simulated and generated by the harmonic superposition method to match the characteristics of the target wind pressure spectrum. Optionally, for large-span colored stone metal tiles, the concentrated line load mode can be set at key locations such as the mid-span and quarter-span. It is understandable that extracting feature points from the overall deformation curve is a key step in determining the ultimate bearing capacity threshold. The accuracy of the extraction process depends on the convergence accuracy of the model solution and the robustness of the curve post-processing algorithm. Theoretical deformation parameters and ultimate bearing capacity thresholds provide quantitative prediction targets and loading control benchmarks for subsequent physical strength testing.

[0044] See Figure 4 This is a statistical analysis chart of wind pressure data from a strength testing project for colored stone metal tiles, used to support subsequent load model construction and strength simulation analysis. The blue dashed line marks the "50-year return period wind pressure peak (4.5 kPa)," a key reference value for extreme weather protection in building structure design. The peak value of the common wind pressure distribution is approximately 2 kPa, representing the typical wind pressure level under daily weather conditions. The hurricane wind pressure time history curve fluctuates between 2 kPa and 8 kPa, reflecting the drastic dynamic changes in wind pressure under strong winds. This chart is the core basis for constructing the "preset load model," clarifying the wind pressure characteristics under different scenarios through statistical analysis. The distribution of the 50-year return period extreme wind pressure and the common wind pressure is used to set the static load conditions; the hurricane time history curve is used as the simulation input for dynamic wind pressure loads. These data will be combined with snow load and equivalent temperature stress to form load conditions that closely match the actual service environment, providing accurate boundary conditions for subsequent stress-strain analysis.

[0045] In one embodiment of the present invention, a load spectrum for the physical strength testing machine is set according to the ultimate bearing threshold range. The peak load of the load spectrum is between the initial failure threshold and the final failure threshold, and includes multiple increasing load steps. Based on the predicted overall deflection value and its changing trend in the theoretical deformation parameters, a displacement monitoring warning line for the physical strength testing machine during the loading process is set. The displacement monitoring warning line is slightly lower than the predicted overall deflection value under the corresponding load. During the physical strength testing process, the physical strength testing machine is controlled to apply load according to the load spectrum. The actual deformation of the tested colored stone metal tile sample is monitored in real time. When the actual deformation approaches or reaches the displacement monitoring warning line under the corresponding load step, a loading rate reduction or load holding command is triggered. Loading continues until the sample fails or reaches a preset termination condition, and the actual failure load and failure mode are recorded. The actual load-displacement curve recorded during the physical strength testing process, as well as the image information at the time of sample failure, are obtained. From the actual load-displacement curve, the actual proportional limit load, maximum load, and corresponding displacement value are extracted. The actual proportional limit load and maximum load are compared with the initial failure threshold and final failure threshold predicted by the comprehensive stress-strain analysis model, and the prediction error is calculated. Based on the prediction error, the values ​​of key material parameters in the integrated stress-strain analysis model are adjusted in reverse. These key material parameters include the equivalent elastic modulus of the colored stone coating and the shear strength of the adhesive layer. The integrated stress-strain analysis model is then updated using the adjusted key material parameters, and the updated model is validated using another set of independent validation sample data to complete the feedback correction closed loop of the model.

[0046] In practical implementation, the load spectrum of the physical strength testing machine is set according to the ultimate bearing capacity threshold range. The lower limit of the ultimate bearing capacity threshold range is the initial failure threshold of 2.5 kPa, and the upper limit is the final failure threshold of 3.8 kPa. The peak load of the load spectrum is set to 3.2 kPa, which is between the initial failure threshold and the final failure threshold. The load spectrum contains five increasing load steps, with each load step incrementing by 0.64 kPa. The displacement monitoring and warning line of the physical strength testing machine during loading is set based on the predicted overall deflection value and its changing trend from the theoretical deformation parameters. The displacement monitoring and warning line is set slightly lower than the overall deflection value predicted by the comprehensive stress-strain analysis model under the corresponding load. For example, under the first load step of 1.28 kPa, the model predicts a deflection of 2.1 mm, so the displacement monitoring and warning line for this level is set to 1.9 mm.

[0047] During the physical strength testing process, the physical strength testing machine applies loads according to the load spectrum. The loads are applied uniformly to the surface of the colored stone metal tile sample under test in the form of air pressure or mechanical loading. The actual deformation of the sample is monitored in real time, and the actual deformation is measured by a high-precision laser displacement sensor installed on the back of the sample. When the actual deformation approaches or reaches the displacement monitoring warning line of the corresponding load step, the control system of the physical strength testing machine triggers a loading rate reduction or load holding command. The loading rate is reduced from 0.05 kPa to 0.02 kPa, or the current load value is maintained for 60 seconds to observe the deformation trend. Loading continues until the sample fails or reaches the preset termination condition. The actual failure load and failure mode are recorded. The failure mode is recorded by a high-speed camera to identify specific forms such as coating peeling, substrate buckling, or adhesive layer failure.

[0048] The actual load-displacement curve during the physical strength testing process, along with image information at sample failure, was acquired. The actual load-displacement curve was recorded by the testing machine's data acquisition system at a frequency of 100 points per second. The actual proportional limit load, maximum load, and corresponding displacement values ​​were extracted from the actual load-displacement curve. The proportional limit load was determined by finding the end point of the initial straight line segment of the curve, with a value of 2.7 kPa, and the maximum load was 3.9 kPa. The actual proportional limit load and maximum load were compared with the initial failure threshold and final failure threshold predicted by the comprehensive stress-strain analysis model to calculate the prediction error. The formula for calculating the prediction error is expressed as follows: in: This represents the percentage of prediction error for a specific load index. This represents the predicted value from the comprehensive stress-strain analysis model. This represents the actual value extracted from the full-process curve of actual load displacement. The values ​​of key material parameters in the comprehensive stress-strain analysis model are adjusted in reverse based on the prediction error. Key material parameters include the equivalent elastic modulus of the colored stone coating and the shear strength of the adhesive layer. For example, if the prediction error of the final failure threshold is 2.6%, the input value of the equivalent elastic modulus of the colored stone coating is proportionally reduced.

[0049] In some embodiments, the displacement monitoring warning line can be set as a fixed proportion based on the predicted deflection, for example, set to 90% of the overall deflection value predicted by the model under the corresponding load. Optionally, the number and increment of the load spectrum steps can be dynamically designed according to the width of the ultimate bearing threshold range to ensure sufficient data acquisition points within the critical load range. It can be understood that the rate slowing or holding command during the loading process is an active protection mechanism designed to prevent the sample from unexpectedly failing before reaching the target peak due to prediction deviation. The integrated stress-strain analysis model is updated using the adjusted key material parameters, and the updated model is validated using another set of independent validation sample data to complete the feedback correction closed loop of the model. The independent validation sample data comes from the same batch of colored stone metal tile samples that did not participate in the aforementioned modeling and testing process. The implementation of the feedback correction closed loop enables the integrated stress-strain analysis model to self-calibrate through actual physical test data, thereby improving its accuracy in predicting the mechanical behavior of the same batch or similar products.

[0050] In some embodiments, the image information of the failure mode can be quantitatively analyzed using digital image correlation techniques to more accurately locate the failure initiation point and compare it with stress concentration areas in the model. Optionally, the prediction error can be calculated separately for the initial failure threshold and the final failure threshold, and the difference between the two errors determines which type of key material parameter should be adjusted first. For example, a large error in the initial failure threshold may indicate that coating or adhesive layer parameters need adjustment, while a large error in the final failure threshold may indicate that substrate material parameters need adjustment. It is understood that the model feedback correction is not a one-time step, but rather an iterative process. As more sample detection data accumulates, the material parameter library of the comprehensive stress-strain analysis model will be continuously optimized, and the predictive ability will gradually improve.

[0051] See Figure 5This is a loading strategy and monitoring and early warning diagram for the physical strength testing of colored stone metal tiles. It intuitively illustrates the stepped loading process, safety thresholds, and deformation early warning logic. This diagram serves as the loading control benchmark for the strength testing of colored stone metal tiles. By presetting stepped loads and corresponding displacement early warning lines, it can be implemented in physical experiments. When the load reaches step 4 (initial failure threshold), the system triggers an early warning and slows down the loading rate. When the load approaches step 5 (final failure threshold), the load is maintained and deformation is closely monitored until the sample is damaged or the experiment is terminated. Through clearly defined initial and final failure thresholds, it provides clear safety boundaries for physical strength testing, avoiding equipment damage or sudden sample failure caused by blind loading.

[0052] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for testing the strength of colored stone metal tiles, characterized in that, include: Acquire surface reflected light distribution data and internal structural stress distribution data of the colored stone metal tile sample to be tested; Based on the surface reflected light distribution data, the surface coating uniformity and micromorphological characteristics of the tested colored stone metal tile sample are analyzed to generate a surface quality assessment dataset. By integrating the surface quality assessment dataset and the internal structural stress distribution data, a comprehensive stress-strain analysis model is constructed for the colored stone metal tile sample to be tested. Based on the comprehensive stress-strain analysis model, the theoretical deformation parameters and ultimate bearing capacity threshold of the tested colored stone metal tile sample under the preset load mode are calculated. The theoretical deformation parameters and the ultimate bearing threshold are used as standard judgment criteria to guide the loading strategy in the subsequent physical strength testing process of the colored stone metal tile sample to be tested.

2. The method for testing the strength of colored stone metal tiles according to claim 1, characterized in that, The acquisition of surface reflected light distribution data and internal structural stress distribution data of the colored stone metal tile sample to be tested specifically involves: The surface of the colored stone metal tile sample under test is illuminated by a multi-band controllable light source, and the reflected light signal is received by a high-resolution spectral imaging sensor. The surface reflected light intensity and spatial distribution information under different bands are collected to form the surface reflected light distribution data. A laser speckle field is projected onto the sample of colored stone metal tile to be tested using a digital speckle interferometry device, and the initial speckle pattern of the sample of colored stone metal tile to be tested under no-load conditions is recorded. A known small preload is applied to the sample of colored stone metal tile to be tested, and the current speckle pattern of the sample of colored stone metal tile under the preload state is recorded again; The current speckle pattern and the initial speckle pattern are subjected to digital correlation calculation to solve the microscopic displacement field of the sample surface caused by the preload. Based on the constitutive relationship of material mechanics, the internal structural stress distribution data of the shallow surface layer of the sample is deduced from the microscopic displacement field.

3. The method for testing the strength of colored stone metal tiles according to claim 2, characterized in that, The process involves analyzing the surface coating uniformity and microstructure characteristics of the tested colored stone metal tile sample based on the surface reflected light distribution data, and generating a surface quality assessment dataset. Specifically: From the surface reflected light distribution data, extract the reflectance spectrum curves of different spatial locations in multiple bands; For each of the aforementioned reflectance spectrum curves, characteristic peaks are identified and peak areas are calculated. The peak area ratio of each characteristic peak is matched with the spectral feature library of the standard colored stone coating to obtain the coating composition consistency index at each location point. The spatial variation coefficient of the coating composition consistency index at all locations is calculated, and the spatial variation coefficient is used to quantitatively characterize the surface coating uniformity. Simultaneously, spatial frequency analysis is performed on the surface reflected light distribution data to separate the low-frequency component characterizing macroscopic smoothness and the high-frequency component characterizing microscopic roughness. Calculate the energy intensity and distribution entropy value of the high-frequency component, and use the energy intensity and distribution entropy value together as a quantitative parameter to describe the surface micromorphological characteristics; The spatial variation coefficient of the surface coating uniformity, as well as the energy intensity and distribution entropy value of the surface micromorphology features, are integrated into structured data to generate the surface quality assessment dataset.

4. The method for testing the strength of colored stone metal tiles according to claim 3, characterized in that, The comprehensive stress-strain analysis model for the tested colored stone metal tile sample is constructed by integrating the surface quality assessment dataset and the internal structural stress distribution data, specifically as follows: A parametric finite element mesh model describing the laminated structure of colored stone metal tiles is established. The parametric finite element mesh model includes a metal substrate layer, an adhesive layer, and a colored stone coating. The spatial variation coefficient of surface coating uniformity in the surface quality assessment dataset is mapped to the spatial distribution function of the material properties of the colored stone coating element in the parameterized finite element mesh model. The energy intensity and distribution entropy value of the surface micromorphology features are converted into the boundary geometric roughness parameters of the colored stone coating surface in the parameterized finite element mesh model. The internal structural stress distribution data is used as the initial stress field and imported into the corresponding position of the parameterized finite element mesh model. Based on the imported material property spatial distribution function, the boundary geometric roughness parameters, and the initial stress field, the calculation engine of the parameterized finite element mesh model is run to construct the comprehensive stress-strain analysis model that can reflect the material non-uniformity and initial stress state.

5. The method for testing the strength of colored stone metal tiles according to claim 4, characterized in that, Based on the comprehensive stress-strain analysis model, the theoretical deformation parameters and ultimate bearing capacity threshold of the tested colored stone metal tile sample under a preset load mode are calculated, specifically as follows: Define one or more of the preset load modes, including uniformly distributed surface load, concentrated line load, or dynamic wind pressure load; Each of the preset load modes is applied to the comprehensive stress-strain analysis model as a boundary condition. By performing nonlinear static solution or dynamic time history analysis using the comprehensive stress-strain analysis model, the stress field evolution cloud map, strain field evolution cloud map, and overall deformation curve of the tested colored stone metal tile sample during the entire load application process are calculated. Feature points are extracted from the overall deformation curve, including the proportional limit point, yield point and maximum load point. The load value corresponding to the proportional limit point is recorded as the initial failure threshold, and the load value corresponding to the maximum load point is recorded as the final failure threshold. The theoretical deformation parameters include the overall deflection value, the maximum strain value and its location predicted by the comprehensive stress-strain analysis model under the specified load level; The ultimate load-bearing threshold is a load range, with its lower limit being the initial failure threshold and its upper limit being the final failure threshold.

6. The method for testing the strength of colored stone metal tiles according to claim 5, characterized in that, The method of using the theoretical deformation parameters and the ultimate bearing threshold as standard judgment criteria to guide the loading strategy in the subsequent physical strength testing process of the colored stone metal tile sample to be tested is as follows: Based on the ultimate bearing threshold range, the load spectrum of the physical strength testing machine is set, wherein the peak load of the load spectrum is between the initial failure threshold and the final failure threshold, and includes multiple increasing load steps. Based on the predicted overall deflection value and its changing trend in the theoretical deformation parameters, a displacement monitoring and early warning line for the physical strength testing machine during the loading process is set, and the displacement monitoring and early warning line is slightly lower than the predicted overall deflection value under the corresponding load. During the physical strength testing process, the physical strength testing machine is controlled to apply loads according to the load spectrum; The actual deformation of the colored stone metal tile sample under test is monitored in real time. When the actual deformation approaches or reaches the displacement monitoring warning line under the corresponding load step, a loading rate reduction or load holding command is triggered. The load is continuously applied until the sample is damaged or a preset termination condition is reached, and the actual damage load and damage mode are recorded.

7. The method for testing the strength of colored stone metal tiles according to claim 6, characterized in that, Following the physical strength testing process, the method further includes a step of feedback correction of the comprehensive stress-strain analysis model based on the test results: Acquire the actual load-displacement curve recorded during the physical strength test, as well as image information when the sample is damaged; Extract the actual proportional limit load, maximum load, and corresponding displacement value from the actual load-displacement full-process curve; The actual proportional limit load and maximum load are compared with the initial failure threshold and final failure threshold predicted by the comprehensive stress-strain analysis model, and the prediction error is calculated. Based on the prediction error, the values ​​of key material parameters in the comprehensive stress-strain analysis model are adjusted in reverse. The key material parameters include the equivalent elastic modulus of the colored stone coating and the shear strength of the adhesive layer. The integrated stress-strain analysis model is updated using the adjusted key material parameters, and the updated model is validated using another set of independent validation sample data to complete the feedback correction closed loop of the model.

8. The method for testing the strength of colored stone metal tiles according to claim 2, characterized in that, Before illuminating the surface of the colored stone metal tile sample under test with a multi-band controllable light source, the method further includes a pre-processing and positioning step for the colored stone metal tile sample under test: The sample of the colored stone metal tile to be tested is placed in a testing station with a constant temperature and humidity environment and left to stand for a preset time to achieve thermal and humidity equilibrium with the environment. The edge contour of the colored stone metal tile sample to be tested is identified and the preset positioning mark points are matched using a visual positioning system. Based on the identified edge contours and the positioning markers, the spatial pose of the colored stone metal tile sample to be tested relative to the multi-band controllable light source and the digital speckle interferometry measurement device is calculated. Adjust the illumination angle of the multi-band controllable light source and the projection angle of the digital speckle interferometry measurement device to ensure that the light source and the measurement beam are perpendicularly incident on the center of the test area of ​​the colored stone metal tile sample.

9. The method for testing the strength of colored stone metal tiles according to claim 5, characterized in that, When defining one or more of the preset load modes, the step includes equivalent simplification based on the load spectrum of the actual service environment: Long-term load monitoring data of colored stone metal tiles under typical building roofing environments were collected. The long-term load monitoring data included wind pressure time history data, snow load data, and temperature difference change data. Extreme value statistical analysis was performed on the wind pressure time history data to extract representative and enveloping wind pressure distribution patterns and peak values; The snow load data is converted into a uniformly distributed surface load according to the roof slope. The difference in thermal expansion coefficients between the metal substrate and the colored stone coating caused by the temperature difference change data is analyzed, and the resulting equivalent temperature stress load is calculated. The wind pressure distribution pattern, the converted uniformly distributed snow load, and the equivalent temperature stress load are combined to form the load combination condition of the preset load pattern.

10. A strength testing system for colored stone metal tiles, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the strength testing method for colored stone metal tiles as described in any one of claims 1 to 9.

Images